Palliative home care for cancer patients in Denmark ...

FACULTY OF HEALTH SCIENCES AARHUS UNIVERSITY

Palliative home care for cancer patients in Denmark - with a particular focus on the primary care sector, GPs and community nurses PhD Thesis

Mette Asbjørn Neergaard

Faculty of Health Sciences Aarhus University 2009

Palliative home care for cancer patients in Denmark - with a particular focus on the primary care sector, GPs and community nurses PhD thesis

Mette Asbjørn Neergaard

In memory of my late husband Poul Raben Jensen (1958 – 1992)

Research Unit and Department of General Practice Faculty of Health Sciences Aarhus University

PhD thesis Palliative home care for cancer patients in Denmark - with a particular focus on the primary care sector, GPs and community nurses Mette Asbjørn Neergaard, MD, GP

This thesis has been accepted for PhD defence by the Faculty of Health Sciences, Aarhus University and th was defended on 29 May 2009.

Supervisors: Professor, Dr.Med.Sci., Frede Olesen, Research Director of The Research Unit for General Practice, University of Aarhus Professor, PhD, Jens Søndergaard, The Research Unit for General Practice, University of Southern Denmark PhD, Consultant, Anders Bonde Jensen, Department of Oncology, Aarhus University Hospital

Opponents: Associate professor, PhD, Dr.Med.Sci., Mogens Grønvold, Health Services Research Unit, Department of Public Health, University of Copenhagen Professor, Dr.Med.Sci., Marianne Ewertz Kvistgaard, Department of Oncology, Odense University Hospital Professor, PhD, Bo Christensen, Department of General Practice, School of Public Health, University of Aarhus (chairman)

Print: Fællestrykkeriet for Sundhedsvidenskab, Aarhus University

ISBN: 978-87-90004-29-3

Address: The Research Unit for General Practice University of Aarhus Bartholins Allé 2 DK-8000 Aarhus C Email: [email protected] www.alm.au.dk

Preface

PREFACE

I

Preface

Outline of this thesis: This thesis is based on the project: “Palliative home care for cancer patients in Denmark - with a particular focus on the present and future role of general practice” – a mixed method study conducted in the period 2005 -2008 in the former Aarhus County, Denmark. The project emanated from the Research Unit and Department for General Practice, Aarhus University. The definitions of issues concerning this thesis are introduced in Chapter 1. A general introduction to the subject is given in Chapter 2 and the aims of the projects are outlined in Chapter 3. Chapter 4 offers a description of material and methods.

The five papers are presented in Chapters 5 to 9.

In Chapters 10 and 11 a more general discussion of the methods and results are given and Chapters 12 to 14 gather the conclusions pertaining to the aims of the project and raise perspectives and offer suggestions for future research.

Finally, Chapters 15 to 17 give English and Danish summaries and a full reference list of the thesis. The Appendices include interview guides, cover letters and questionnaires.

The PhD study was funded by the Aarhus County Research Fund for the Clinical Development and Research in General Practice and across the Primary and Secondary Health Care Sectors (4-01-304), the Danish National Research Foundation for Primary Care (585-457808) and The Multipractice Study Committee (585-04/2072).

II

Preface

This PhD thesis is based on the following five papers:

1. Neergaard MA, Olesen F, Jensen AB, Sondergaard J. Palliative care for cancer patients in a primary health care setting: Bereaved relatives' experience, a qualitative group interview study. BMC Palliat Care 2008;7:1

2. Neergaard MA, Olesen F, Jensen AB, Sondergaard J. Shared care in basic level palliative home care – Organisational and interpersonal challenges. Submitted to Family Practice

3. Neergaard MA, Vedsted P, Olesen F, Sokolowski I, Jensen AB, Sondergaard J Associations between home death and general practitioner involvement in palliative care. Submitted to British Journal of General Practice

4. Neergaard MA, Vedsted P, Olesen F, Sokolowski I, Jensen AB, Sondergaard J Associations between successful palliative cancer pathways and general practitioner involvement. Submitted to Scandinavian Journal of Primary Health Care

5. Neergaard MA, Vedsted P, Olesen F, Sokolowski I, Jensen AB, Sondergaard J Associations between successful palliative cancer pathways and community nurse involvement. Submitted to BMC Palliative Care

III

Preface

Motivation My interest in palliation began when my then beloved husband died of cancer in 1992. Working for the general practitioners Carl Erik Mabeck and Henrik Læssøe, Højbjerg in 1998, I was entrusted with the task of providing palliative care, for which I am very grateful to patients, relatives and my colleagues. Afterwards, I was lucky to work at the Department of Oncology at Aarhus University Hospital where I soon discovered that my main interest lay with terminally ill patients and their families. Thus, in the final phase of my studies to become a general practitioner in 2003, I contacted my colleague, Marianne Bjerager, who at the time was a PhD-student at the Research Unit for General Practice in Aarhus, and she incited me to contact Frede Olesen, professor, GP and director of the Research Unit for General Practice in Aarhus. Fortunately, Frede supported my idea of a research project on palliative and primary care and I started at the Research Unit in August 2004. Acknowledgements Throughout my employment at The Research Unit for General Practice in Aarhus, I have had a feeling of being “at home”. Frede Olesen‟s trust in me and my abilities is priceless and I am deeply grateful to have had the opportunity to work with him and under his “wings”. Likewise, my two other supervisors Jens Søndergaard and Anders Bonde Jensen have provided immeasurable supports throughout the whole period. Furthermore, I would like to thank Peter Vedsted for his help with the quantitative articles, and his never failing sense of perspective in every research issue. My special, deep-felt thanks go to the participating bereaved relatives, community nurses, hospital consultants and GPs, who all played an irreplaceable role in the study. I also owe thanks to the Palliative Care team and community nurses in Aarhus for locating potential relatives to participate in the interviews. I also wish to thank the administrative and technical staff of the Research Unit, Birthe Brauneiser, Eva Højmark, Hanne Beyer, Lars Olesen and Eva Munkholm for their patience and involvement. Furthermore, I wish to thank Eva Højmark, Morten Pilegaard, Ellen Asbjørn Jensen and Kim Neergaard for revising my manuscripts, Lone Brøcker, Elsebeth S. Christensen and Pia Høj Christiansen for transcribing interviews and Josefine, Natasja and Maja for packing, scanning and verifying the questionnaires. I am very grateful to my “office-mates”, Ineta Sokolowski, Kaj S. Christensen and Marianne Rosendal for providing a pleasant office environment with fruitful discussions. Particular thanks are extended to Ineta for her friendship and help with statistics. I also owe special thanks to Rikke Pilegaard Hansen and Marianne Bjerager who were a great help, when I first started at the Research Unit. Since Trine Brogaard and Mai-Britt Guldin started at the Research unit, I have not been alone with my passion for palliative care and I thank them for being soul mates on this and a lot of other issues. I wish to thank Rikke Sand Andersen for rewarding discussions about qualitative research methods and my very good and “old” friend at 1st floor, Bodil Hammer Beck for always being there for me. There is no doubt that my colleagues in building 1260 are the best one can imagine and they all contributed to make my PhD period a wonderful journey. The greatest thank of all goes to my family: My parents, my brothers, their families and my familyin-law have always been a tremendous support. With all my heart I thank my girls, “The Fantastic Four”, Vilma, Karla, Josefine and Aja, my sons-in-law Emil and Lau and my two wonderful grandchildren Sille and Otto for bringing much happiness, laughter and a whole lot of meaning into my life. Lastly my ultimate deep-felt appreciation goes to Kim Neergaard, my wonderful husband, for love and care and his never failing trust in my abilities. Mette Asbjørn Neergaard, Aarhus 2009

IV

Preface

Abbreviations and acronyms CI CN FGI FO GP HDR IQI JS MAN PR RCD SD SWOT OR QD WHO

Confidence interval Community nurse Focus group interview Frede Olesen General practitioner The county hospital discharge registry Inter quartile interval Jens Søndergaard Mette Asbjørn Neergaard Prevalence ratio The Danish Register of Causes of Death Standard deviation Strengths, weaknesses, opportunities and threats Odds ratio Qualitative description The World Health Organisation

V

“How people die remains in the memories of those who live on” Dame Cicely Saunders

Contents

Contents 1. 1.1. 1.2. 1.3. 1.4. 1.5. 1.6. 1.7. 1.8.

Definitions.............................................................................................................. 5 Palliative care Basic and specialist palliative care level The palliative period Patients in need of palliative care Families in palliative care Palliative home care Shared care References of Chapter 1

2. 2.1. 2.2. 2.3.

General introduction ............................................................................................ 11 General introduction Introduction at a glance References of Chapter 2

3. 3.1. 3.1.1. 3.1.2. 3.2. 3.2.1. 3.2.2. 3.2.3. 3.2.4. 3.2.5.

Aims ....................................................................................................................... 19 The overall aim of the thesis Research question A Research question B The aims of the five studies Aim 1 Aim 2 Aim 3 Aim 4 Aim 5

4. 4.1. 4.2. 4.2.1. 4.2.1.1. 4.2.1.2. 4.2.1.3. 4.2.2. 4.2.2.1. 4.2.2.2. 4.2.2.3. 4.2.2.4. 4.2.2.5. 4.2.2.6. 4.3. 4.4. 4.5.

Materials and methods.......................................................................................... 23 Setting Methods Focus group interviews Identifying informants The interviews Analysis of the qualitative studies Quantitative surveys The questionnaires Pilot testing Study population and sampling Data collection Data entry Analysis of the quantitative studies Economic incentives Approvals References of Chapter 4

1

Contents

5. Paper 1.................................................................................................................... 35 Palliative care for cancer patients in a primary health care setting - Bereaved relatives' experience. A qualitative group interview study 5.1. Abstract 5.2. Background 5.3. Methods 5.4. Results 5.5. Discussion 5.6. Conclusion 5.7. Tables 5.8. References of Chapter 5 6. Paper 2.................................................................................................................... 53 Shared care in basic level palliative home care - organizational and interpersonal challenges 6.1 Abstract 6.2. Introduction 6.3. Material and method 6.4. Results 6.5. Discussion 6.6. Conclusion 6.7. Tables 6.8. References of Chapter 6 7. Paper 3................................................................................................................... 71 Associations between home death and general practitioner involvement in palliative care 7.1. Abstract 7.2. Introduction 7.3. Materials and methods 7.4. Results 7.5. Discussion 7.6. Conclusion 7.7. Tables 7.8. References of Chapter 7 8. Paper 4.................................................................................................................... 85 Associations between successful palliative cancer pathways and general practitioner involvement 8.1. Abstract 8.2. Introduction 8.3. Materials and methods 8.4. Results 8.5. Discussion 8.6. Conclusion 8.7. Tables 8.8. References of Chapter 8

2

Contents

9. Paper 5.................................................................................................................... 101 Associations between successful palliative cancer pathways and community nurse involvement 9.1. Abstract 9.2. Introduction 9.3. Method 9.4. Results 9.5. Discussion 9.6. Conclusion 9.7. Tables 9.8. References of Chapter 5 10. Discussion of methods......................................................................................... 119 10.1. Mixed methods design 10.1.1. Choice of informants / participants 10.2. Qualitative study 10.2.1. The focus group interviews 10.2.2. Selection of informants 10.2.3. Analysis 10.2.4. Validity 10.2.4.1. Credibility 10.2.4.2. Dependability and conformability 10.2.4.3. Transferability 10.3. Quantitative surveys 10.3.1. Study design 10.3.2. Questionnaires 10.3.3. Variables in the models 10.3.4. Response rates 10.3.5. Internal validity 10.3.5.1. Data from registers 10.3.5.2. Sampling errors 10.3.5.3. Selection bias because of non-responding GPs 10.3.5.4. Selection bias because of non-responding CNs 10.3.5.5. Selection bias because of non-responding relatives 10.3.5.6. Selection bias because of relatives excluded by GPs or CNs 10.3.5.7. Social desirability bias (Information bias) 10.3.5.8. Recall bias (Information bias) 10.3.5.9. Confounding 10.3.6. Generalisability 10.4. Ethical considerations 10.5. References of Chapter 10 11. 11.1. 11.2. 11.3. 11.3.1. 11.3.2. 11.4.

General discussion of results.............................................................................. 145 Quality of care at the end of life Relatives‟ versus patients‟ evaluation Discussion of main results from Papers 1 - 5 Research question A Research question B References of Chapter 11

3

Contents

12. 12.1. 12.2. 12.3. 12.4. 12.5. 12.6. 12.7.

Conclusion Thesis................................................................................................. 157 Conclusion in relation to the overall aim of the thesis Conclusion in relation to aim 1 Conclusion in relation to aim 2 Conclusion in relation to aim 3 Conclusion in relation to aim 4 Conclusion in relation to aim 5 Conclusion of thesis

13.

Perspectives and implication for health care .................................................... 161

14.

Future research .................................................................................................... 163

15.

English summery.................................................................................................. 167

16.

Danish summery.................................................................................................... 171

17.

References listed alphabetically by author......................................................... 175

Appendices .............................................................................................................................. 187 A: B: C: D: E: F: G: H: I: J: K: L: M: N: O:

Interview guide, final, relatives (in Danish) Interview guide, final, GPs (in Danish) Interview guide, final, mixed professionals (in Danish) Interview guide, final, Community nurses (in Danish) Topic guide themes in interviews with professionals Scales and questionnaires developed for relatives in palliative care, incl. reference list Themes in questionnaires Variables with data retried from official registers (Papers 3 – 5) Variables from questionnaires with coherent questions (Papers 3 – 5) Cover letter, GPs (in Danish) Questionnaire, GP (in Danish) Cover letter, Community nurses (in Danish) Questionnaire, Community nurses (in Danish) Cover letter, Relatives (in Danish) Questionnaire, Relatives (in Danish)

4

Chapter 1. Definitions

CHAPTER 1 DEFINITIONS

5


1.1. Palliative Care The World Health Organisation (WHO) defines palliative care as follows: “Palliative care improves the quality of life of patients and families who face life-threatening illness, by providing pain and symptom relief, spiritual and psychosocial support from diagnosis to the end of life and bereavement. Palliative Care: provides relief from pain and other distressing symptoms; affirms life and regards dying as a normal process; intends neither to hasten or postpone death; integrates the psychological and spiritual aspects of patient care; offers a support system to help patients live as actively as possible until death; offers a support system to help the family cope during the patient‟s illness and in their own bereavement; uses a team approach to address the needs of patients and their families, including bereavement counselling, if indicated; will enhance quality of life, and may also positively influence the pathway of illness; is applicable early in the pathway of illness, in conjunction with other therapies that are intended to prolong life, e.g. chemotherapy or radiation therapy, and includes investigations needed to better understand and manage distressing clinical complications“ (1)

1.2. Basic and specialist palliative care level Palliative care is ideally divided into a basic level and a specialized level. The Danish Health Authorities have defined the basic palliative care level as follows: “The palliative effort at basic level concerns the effort at general hospital wards and in the patients‟ homes. The effort at the hospitals is performed by the hospital staff and in the patients‟ homes by the general practitioner (GP) and home care services”. The specialist palliative care level is defined as follows: “The palliative effort at specialist level concerns the patient who has a complex symptomatology and who demands a specialised and / or a inter-professional effort, including support for psychosocial and existential problems ... The tasks of the palliative specialist care teams are, in collaboration with the GP and the community nurse, to improve the patient‟s ability to stay at home during the palliative period, to die at home and to avoid unnecessary admittances to hospitals” (2).

6


1.3. The palliative period The palliative period is very difficult to define in terms of a precise time frame. It covers the time period from the decision is made to discontinue all curatively intended treatment until death. The time span varies from diagnosis to diagnosis, from stage to stage of the disease at time of diagnosis and from individual to individual. Defining the palliative period is further complicated by the fact that many groups of cancer patients receive life-prolonging treatments (e.g. chemotherapy and hormonal therapy), just like palliative treatment, to a far greater extent now than previously.

1.4. Patients in need of palliative care According to the WHO, all patients with a life-limiting disease are potential recipients of palliative care (1). The number of patients in Denmark who receive palliative care is not known, but the majority of patients in need of palliative care are cancer patients, and 15 626 patients died from cancer in Denmark in 2006 (3). The majority of these patients probably needed palliative care (4). Terminally ill non-cancer patients, e.g. patients with terminal heart failure, terminal chronic obstructive lung disease, terminal neurological diseases and HIV, are just as important in palliative care. However, so far these patients account for only a small fraction of patients in palliative care in Denmark. 1.5. Families in palliative care As it appears from the WHO definition of palliative care, the families are in focus as well as the patients (1). Relatives seem to cope better with bereavement after a loved one's death if sufficient support is provided during the palliative stage of disease at home (5-7). The close relative is in a dual position being both an important participant in palliative home care and a person who is suffering from the coming loss of a loved relative. Relatives want to be involved in the care, especially if sufficient support is provided by the professionals (8). In this thesis, the relative is referred to as the closest relative, meaning the family member or friend who is closest to the patient in the palliative phase. 1.6. Palliative home care Palliative home care includes all palliative care services performed in the patients‟ homes. For patients living in a nursing home, this is seen as their homes. In this thesis, the palliative care effort performed at nursing homes is therefore included in “palliative home care”.

7


1.7. Shared care Since patients in palliative home care are often both in contact with GPs, CNs, hospital consultants, and sometimes also with a palliative specialist team, the cooperation between these professionals is important. Shared care may be an approach to organise this collaboration:

Pritchard and Hughes have defined shared care as follows: “... the responsibility for the health care of the patient is shared between individuals or teams who are part of separate organizations or where substantial organizational boundaries exist” (9)

Hickman has a more narrow definition: “.... the joint participation of general practitioners and hospital consultants in the planned delivery of care for patients with a chronic condition, informed by an enhanced information exchange over and above routine discharge and referral letters” (10)

8


1.8. References of Chapter 1 (1) World Health Organization. Pain relief and palliative care. National cancer control programmes, policies and managerial guidelines. 2nd ed. ed. Geneva: 2002. p. 86-91. (2) The National Board of Health. National Cancer Plan II • Denmark National Board of Health recommendations for improving cancer healthcare services. Copenhagen: The National Board of Health; 2005. (3) The Danish National Board of Health. The Danish Register of Causes of Death [In Danish]. 2008. (4) Lidstone V, Butters E, Seed PT, Sinnott C, Beynon T, Richards M. Symptoms and concerns amongst cancer outpatients: identifying the need for specialist palliative care. Palliat Med 2003 Oct;17(7):588-95. (5) Grande GE, Farquhar MC, Barclay SI, Todd CJ. Valued aspects of primary palliative care: content analysis of bereaved carers' descriptions. Br J Gen Pract 2004 Oct;54(507):772-8. (6) Jansma FF, Schure LM, de Jong BM. Support requirements for caregivers of patients with palliative cancer. Patient Educ Couns 2005 Aug;58(2):182-6. (7) Milberg A, Strang P. What to do when 'there is nothing more to do'? A study within a salutogenic framework of family members' experience of palliative home care staff. Psychooncology 2007 Aug;16(8):741-51. (8) Weibull A, Olesen F, Neergaard MA. Caregivers' active role in palliative home care - to encourage or to dissuade? A qualitative descriptive study. BMC Palliat Care 2008 Sep 16;7(1):15. (9) Pritchard P, Hughes J. Shared Care The Future Imperative? 1.ed. ed. London: Royal Society of Medicine Press; 1995. (10) Hickman M, Drummond N, Grimshaw J. A taxonomy of shared care for chronic disease. J Public Health Med 1994;16(4):447-54.

9

10

Chapter 2. General introduction

CHAPTER 2 GENERAL INTRODUCTION

11


2.1. General introduction Approximately 55 000 persons die each year in Denmark (1) and between one third and one fourth of them die from cancer. Hence, 15 626 persons died from cancer in Denmark in 2006.

International studies from countries that Denmark is normally compared with, show that most patients and their relatives wish terminal care and death to take place at home (2-6), but no data yet exists detailing where Danish cancer patients wish to die.

The place of death is reported in the Danish Register of Causes of Death (RCD) and Table 2.1 shows the place of death for Danish cancer patients in the years 2000–2005. It is seen that these years have seen little change in the distribution of places of death (1). Approximately half of all cancer patients die at hospitals, about one fourth die at nursing homes and one fourth die at home. Table. 2.1. Percentage distribution s of cancer deaths divided by place of death in 2000–2005 (1) Place of death (%)

2000

2001

2002

2003

2004

2005

Hospital 57.3 55.4 * * * 55.1 Nursing home and hospice** 13.8 14.2 * * * 17.9 Home 25.0 25.7 * * * 25.8 Other place 3.9 4.7 * * * 0.8 Not registered 0.0 0.0 * * * 0.5 In total 100.0 100.0 * * * 100.1 * These numbers are not yet available from The Danish national Board of Health who administers the Danish Register of Causes of Death ** In 1995-2005 deaths at hospices were registered as deaths at nursing homes.

Hence, there seems to be a mismatch between cancer patients‟ wish for and their actual place of death.

One has to keep in mind that for a huge part of those cancer patients who died at nursing homes, they also died “at home”, since they lived at a nursing home even before the palliative period. However, some of the nursing home deaths are deaths of patients who were admitted to the nursing home in the palliative period and represent a kind of “institutional death”. We have, however, no possibility to separate between these two types of “nursing home deaths” in RCD, but in both groups the doctor involved is the patient‟s own GP. Furthermore, the data does not allow us to separate between the place of care and place of death in the palliative period which is also an important information (7).

12


Since palliative care is often needed during the period just before death of a cancer patient (8), the services provided in this period is crucial in achieving home death, and maybe more importantly, in achieving a good palliative pathway. In this thesis, we define the palliative period as the last period of the patients‟ life where all curative and life prolonging treatments are stopped and the care and treatment are merely palliative.

In caring for dying cancer patients, one has to remember that, on the one hand, they form a homogenous group because they share a course of disease that progresses through phases of diagnosis, curative treatment, realisation of incurability and a palliative pathway; yet on the other hand, they are also heterogeneous in terms of symptoms, socioeconomic relations and health professional involvement. Both patient and relatives therefore need individual support during the palliative pathway. Since the palliative effort comprises both pain symptom relief, spiritual and psychosocial issues (9), the effort should be team-based and interdisciplinary (10;11).

Dying patients and their relatives request symptom control, access to appropriate care, availability of professionals, time to listen and continuity in their contact with the health care system (12-16). The wish for continuity may be particularly strong in palliative care where patients are in contact with an array of health providers during a period (17), where they are in dire need of peace and safety (14;16). Both relational and informational continuity in palliative pathways seem important to terminally ill patients (16).

Hence, it can be seen from above that high quality of care in the palliative pathway demands that the professionals involved exercise flexibility, tailor the care to the needs of the individual patients and relatives, and commit themselves to cooperation in a team, provide symptom control, availability and continuity.

The palliative care effort in Denmark is officially organised at a basic and a specialist palliative care level (see Chapter 1 for definitions) (11). Some patients require specialist palliative care because they suffer from severe symptoms or have serious problems of a physical, psychological, emotional or spiritual nature. However, most patients are taken well care of at the basic palliative care level, where GPs, CNs and home care services provide primary professional palliative care services in the patients‟ homes. In Denmark there is no central registration of palliative pathways and nobody knows how the task distribution between the two levels of palliative care actually is. Furthermore, there is no official guideline on how to organise the palliative care effort nationally like in other surrounding countries, e.g The United Kingdom (18). In Denmark only recommendations

13


have been given (11). The present organisation of palliative care in Denmark has therefore evolved locally and it essentially rests not on evidence but on especially interested professionals‟ views and traditions, political trends and goodwill in the different areas of the country.

As mentioned above, the primary health care team, e.g. the GPs and CNs, is often involved in the palliative pathway when the terminally ill patient stays at home. The GPs and CNs are frontline workers in the patients‟ homes - sometimes in cooperation with a palliative specialist team. Many GPs consider palliative care to be typical GP work and one of the more rewarding parts of the job (19-25). The traditions and values of general medicine corroborate the intentions of palliative care (9;26), e.g. in seeking a continuing relationship between patient, family and professionals, the continuity of care (27) and the ability to perform care in co-operation with other health professionals (28). The Danish authorities recommend that the GP has a central, coordinating role (29).

It has been postulated that the low percentage distribution of home deaths may be rooted in poor delivery of primary healthcare (4). Thus, earlier studies identified considerable dissatisfaction with symptom control in a primary care setting (5;30-32); however, satisfaction with the GP was still rated high (5) and relatives valued the involvement of GPs and CNs (33). Furthermore, research has shown that involvement of GPs or CNs is positively associated with home death (34-39). All in all, it suggests that good palliative care amounts to more than simply a certain degree of symptom control.

However, GPs and CNs do not necessarily act like a multidisciplinary team in palliative care. GPs and CNs have been found to appreciate each others‟ efforts in palliative care (40), but barriers to co-operation have also been found: e.g. CNs criticised GPs for involving them too late and some were critical of GPs‟ knowledge of pain treatment (40;41). Likewise, GPs regarded CNs so overstretched that they tended to defer involving them and felt that too many different CNs were involved in each case (40).

GPs and CNs have also pointed out that co-operation with the secondary health care system did not always work and that shortfalls included e.g. inadequate information handover (40;42), lack of shared decision-making (43) and lack of available specialists in palliative care (21;44). It seems that barriers to successful co-operation or shared care in palliative care exist both within the primary healthcare sector and across sectoral borders.

14


To sum up, the present organisation of palliative care does not support home death to the extent requested by patients and relatives even if specialist palliative care team have sprung up in many major cities in Denmark over the past decade. The problem may lie in poor delivery of primary healthcare, but may also be due to the working conditions of the primary healthcare professionals, which do not adequately support co-operation with the secondary health care system.

Hence, knowledge of how the palliative home care ideally is best organised is inadequate and more insight into the quality of the palliative effort in completed palliative pathways in the primary care sector is required. The patients and relatives are recipients of palliative care and their experiences, wishes and suggestions for improvement are pivotal to efforts at improving the entire palliative effort in the primary care health sector. Furthermore, the views of the professionals involved are important to determine facilitators and barriers to achieving a good palliative pathway and home death. To conclude, knowledge is needed of patients‟, bereaved relatives‟, GPs‟, CNs‟ and other health care professionals‟ experiences, wishes and suggestions on how to improve the entire palliative effort in the primary care health sector. Particularly important is to obtain knowledge about the primary health care sector‟s effort in respect of palliative care and how the primary and the secondary health care sectors cooperate in palliative care. 2.2. Introduction at a glance •

About one fourth of all cancer deaths in Denmark occur in the patients‟ homes and added to this is the deaths that occur in nursery homes often cared for by the GPs

•

Most patients and their relatives wish terminal care and death to take place at home

•

GPs and CNs are frontline workers in the patients‟ homes in the palliative pathway

•

Palliative care demands flexibility, cooperation, symptom control, availability and continuity from the professionals involved

•

The present organisation of palliative care does not succeed in achieving homedeath of patients to the extent it is requested by patients and relatives

•

There is a need for a deeper insight into experiences, barriers and facilitators of good palliative home care, especially in relation to primary health care

15


2.3. References of Chapter 2 (1) The Danish National Board of Health. The Danish Register of Causes of Death [In Danish]. 2008. (2) Thomas C, Morris SM, Clark D. Place of death: preferences among cancer patients and their carers. Soc Sci Med 2004 Jun;58(12):2431-44. (3) Thomas C. The place of death of cancer patients: can qualitative data add to known factors? Soc Sci Med 2005 Jun;60(11):2597-607. (4) Grande GE, Farquhar MC, Barclay SI, Todd CJ. Valued aspects of primary palliative care: content analysis of bereaved carers' descriptions. Br J Gen Pract 2004 Oct;54(507):772-8. (5) Hanratty B. Palliative care provided by GPs: the carer's viewpoint. Br J Gen Pract 2000 Aug;50(457):653-4. (6) Brazil K, Howell D, Bedard M, Krueger P, Heidebrecht C. Preferences for place of care and place of death among informal caregivers of the terminally ill. Palliat Med 2005 Sep;19(6):492-9. (7) Agar M, Currow DC, Shelby-James TM, Plummer J, Sanderson C, Abernethy AP. Preference for place of care and place of death in palliative care: are these different questions? Palliat Med 2008 Oct;22(7):787-95. (8) Franks PJ, Salisbury C, Bosanquet N, Wilkinson EK, Kite S, Naysmith A, et al. The level of need for palliative care: a systematic review of the literature. Palliat Med 2000 Mar;14(2):93-104. (9) World Health Organization. Pain relief and palliative care. National cancer control programmes, policies and managerial guidelines. 2nd ed. ed. Geneva: 2002. p. 86-91. (10) Palliativ behandling i almen praksis og lokalområdet. Klinisk vejledning (in press). 1 ed. København: DSAM; 2004. (11) The National Board of Health. National Cancer Plan II • Denmark National Board of Health recommendations for improving cancer healthcare services. Copenhagen: The National Board of Health; 2005. (12) Baker R. Development of a questionnaire to assess patients' satisfaction with consultations in general practice. Br J Gen Pract 1990;40(341):487-90. (13) Lecouturier J, Jacoby A, Bradshaw C, Lovel T, Eccles M. Lay carers' satisfaction with community palliative care: results of a postal survey. South Tyneside MAAG Palliative Care Study Group. Palliat Med 1999 Jul;13(4):275-83. (14) Norman A, Sisler J, Hack T, Harlos M. Family physicians and cancer care. Palliative care patients' perspectives. Can Fam Physician 2001 Oct;47:2009-12,;%2015-6.:2009-12. (15) Guthrie B, Wyke S. Personal continuity and access in UK general practice: a qualitative study of general practitioners' and patients' perceptions of when and how they matter. BMC Fam Pract 2006;7:11.

16


(16) Michiels E, Deschepper R, Van Der Kelen G, Bernheim JL, Mortier F, Vander SR, et al. The role of general practitioners in continuity of care at the end of life: a qualitative study of terminally ill patients and their next of kin. Palliat Med 2007 Jul;21(5):409-15. (17) Borgsteede SD, Deliens L, van der Wal G, Francke AL, Stalman WA, van Eijk JT. Interdisciplinary cooperation of GPs in palliative care at home: a nationwide survey in The Netherlands. Scand J Prim Health Care 2007 Dec;25(4):226-31. (18) Department of Health U. End of Life Care Strategy. Promoting high quality care for all adults at the end of life. 2008. (19) Field D. Special not different: general practitioners' accounts of their care of dying people. Soc Sci Med 1998 May;46(9):1111-20. (20) Abom BM, Obling NJ, Rasmussen H, Kragstrup J. [Unplanned emergency admission of dying patients. Causes elucidated by focus group interviews with general practitioners]. Ugeskr Laeger 2000 Oct 23;162(43):5768-71. (21) Burge F, McIntyre P, Twohig P, Cummings I, Kaufman D, Frager G, et al. Palliative care by family physicians in the 1990s. Resilience amid reform. Can Fam Physician 2001 Oct;47:1989-95. (22) Mitchell GK. How well do general practitioners deliver palliative care? A systematic review. Palliat Med 2002 Nov;16(6):457-64. (23) Groot MM, Vernooij-Dassen MJ, Crul BJ, Grol RP. General practitioners (GPs) and palliative care: perceived tasks and barriers in daily practice. Palliat Med 2005 Mar;19(2):111-8. (24) Burt J, Shipman C, White P, Addington-Hall J. Roles, service knowledge and priorities in the provision of palliative care: a postal survey of London GPs. Palliat Med 2006 Jul;20(5):487-92. (25) O'Connor M, Lee-Steere R. General practitioners' attitudes to palliative care: a Western Australian rural perspective. J Palliat Med 2006 Dec;9(6):1271-81. (26) Allen J, Gay B, Crebolder H, Heyrman J, Svab I, Ram P. The European Definitions of the Key Features of the Discipline of General Practice: the role of the GP and core competencies. Brit J Gen Pract 2002;52(479):526-7. (27) von B, I, Eliasson G, Sarvimaki A, Mattsson B, Hjortdahl P. Patients' views on interpersonal continuity in primary care: a sense of security based on four core foundations. Fam Pract 2006 Apr;23(2):210-9. (28) Nielsen JD, Palshof T, Mainz J, Jensen AB, Olesen F. Randomised controlled trial of a shared care programme for newly referred cancer patients: bridging the gap between general practice and hospital. Qual Saf Health Care 2003 Aug;12(4):263-72. (29) Hjælp til at leve til man dør. Rapport fra arbejdsgruppe om palliativ indsats i amter og kommuner. Amtsrådsforeningen, Sundhedsministeriet, Kommunernes Landsforening; 2001.

17


(30) Addington-Hall JM, MacDonald LD, Anderson HR, Freeling P. Dying from cancer: the views of bereaved family and friends about the experiences of terminally ill patients. Palliat Med 1991;5:207-14. (31) Jones RV, Hansford J, Fiske J. Death from cancer at home: the carers' perspective. BMJ 1993;306:249-51. (32) Sykes NP, Pearson SE, Chell S. Quality of care of the terminally ill: the carer's perspective. Palliative Medicine 1992;6:227-36. (33) Brazil K, Bedard M, Krueger P, Abernathy T, Lohfeld L, Willison K. Service preferences among family caregivers of the terminally ill. J Palliat Med 2005 Feb;8(1):69-78. (34) Cantwell P, Turco S, Brenneis C, Hanson J, Neumann CM, Bruera E. Predictors of home death in palliative care cancer patients. J Palliat Care 2000;16(1):23-8. (35) Brazil K, Bedard M, Willison K. Factors associated with home death for individuals who receive home support services: a retrospective cohort study. BMC Palliat Care 2002 Mar 25;1(1):2-7. (36) Burge F, Lawson B, Johnston G, Cummings I. Primary care continuity and location of death for those with cancer. J Palliat Med 2003 Dec;6(6):911-8. (37) Aabom B, Kragstrup J, Vondeling H, Bakketeig LS, Stovring H. Population-based study of place of death of patients with cancer: implications for GPs. Br J Gen Pract 2005 Sep;55(518):684-9. (38) Fukui S, Fukui N, Kawagoe H. Predictors of place of death for Japanese patients with advanced-stage malignant disease in home care settings: a nationwide survey. Cancer 2004 Jul 15;101(2):421-9. (39) Howat A, Veitch C, Cairns W. A retrospective review of place of death of palliative care patients in regional north Queensland. Palliat Med 2007 Jan;21(1):41-7. (40) Cartwright A. The relationship between general practitioners, hospital consultants and community nurses when caring for people in the last year of their lives. Fam Pract 1991 Dec;8(4):350-5. (41) Cartwright A. Balance of care for the dying between hospitals and the community: perceptions of general practitioners, hospital consultants, community nurses and relatives. Br J Gen Pract 1991 Aug;41(348):271-4. (42) Shipman C, Addington-Hall J, Barclay S, Briggs J, Cox I, Daniels L, et al. Providing palliative care in primary care: how satisfied are GPs and district nurses with current out-ofhours arrangements? Br J Gen Pract 2000 Jul;50(455):477-8. (43) Wakefield MA, Beilby J, Ashby MA. General practitioners and palliative care. Palliat Med 1993;7(2):117-26. (44) Barclay S, Todd C, Grande G, Lipscombe J. Controlling cancer pain in primary care: the prescribing habits and knowledge base of general practitioners. J Pain Symptom Manage 2002 May;23(5):383-92.

18

Chapter 3. Aims

CHAPTER 3 AIMS

19

Chapter 3. Aims

3.1. The overall aim of the thesis The overall aim of the study was to evaluate the basic palliative care level in the primary health care sector in general with a particular focus on the role of the GP and CN and their cooperation with the secondary health care sector and their relation to patient and relatives.

The following research questions were addressed:

3.1.1.

Research question A

Which suggestions for improvement and which wishes to the function and organisation of palliative care at home do bereaved relatives, GPs, CNs and consultants at relevant hospital wards have, and which factors are related to experiences of successful and unsuccessful palliative pathways?

3.1.2.

Research question B

How is a series of newly finished palliative pathways at home evaluated, and which factors are associated with experiences of successful and unsuccessful palliative pathways?

3.2.

The aims of the five studies

3.2.1.

Aim 1

The aim was to analyse experiences and preferences of bereaved relatives to terminally ill cancer patients in a primary care setting to explore barriers and facilitators for delivery of good palliative home care (Paper 1).

3.2.2. The

Aim 2 aim

was

to

illuminate

involved

health-professionals‟

experiences

and

views

on

interprofessional co-operation in basic level palliative home-care (Paper 2).

20

Chapter 3. Aims

3.2.3.

Aim 3

The aim was to describe cancer patients in palliative home care in relation to demographic characteristics, disease pathway, degree of GP involvement, and to examine the association between home death and GP involvement (Paper 3).

3.2.4.

Aim 4

The aim was to examine associations between the bereaved relatives‟ evaluation of palliative pathways, place of death and GPs‟ involvement (Paper 4).

3.2.5.

Aim 5

The aim was to examine the association between the bereaved relatives‟ evaluation of palliative pathways and the place of death and the CNs‟ involvement (Paper 5).

21

22

Chapter 4. Materials and methods

CHAPTER 4 MATERIALS AND METHODS

23


We conducted a mixed method study comprising three focus group interviews with bereaved relatives, seven focus group interviews with GPs, CNs and hospital consultants and a combined questionnaire and register survey with bereaved relatives, GPs and CNs in relation to 599 consecutive cases of cancer patients who died from cancer. 4.1. Setting We conducted the study in 2005-2008 in Aarhus County (at that time approximately 640,000 inhabitants, 12% of the Danish population, 26 municipalities and 1,680 deaths from cancer (2005)) (1).

The Danish health care system is tax-financed and more than 98% of Danes are registered with GPs and receive free medical care (2). Danish GPs act as gatekeepers for access to specialist treatment and are responsible for frontline care 24 hours a day. Large GP cooperations provide out-of-hours care (4 pm to 8 am and Sat/Sun), staffed and run by ordinary GPs, covering an entire county. Denmark has no formal national agreement on task distribution in palliative care, but the Aarhus County provided a special service fee for GPs‟ palliative care involvement.

CNs employed by the municipalities are often involved in palliation and visit patients on a 24-hour basis.

Palliative specialist outgoing teams based at the major hospitals are available during daytime hours. Specialist advice from the palliative specialist team can also be obtained by GPs or community nurses by telephone. 4.2.

Methods

In the PhD study we chose a sequential, mixed method approach (qualitative  quantitative) (Figure 10.1., Chapter 10), firstly conducting focus group interviews and subsequently doing a combined register and questionnaire survey, since our research aims invited both qualitative and quantitative research approaches.

4.2.1.

Focus group interviews

We chose focus group interviews as the method to answer research question A (See Chapter 3), using a qualitative description analysis approach (See later). In our study, focus group interviews seemed as the right method to use since we wanted to obtain broad insight into the subject and since the shared feeling among the informants of a difficult

24


experience can take the focus away from the individual experience and create a common understanding and approach to an emotionally difficult topic.

4.2.1.1. Identifying informants In Paper 1 (Chapter 5) informants were close relatives of recently deceased cancer patients because we wanted to evaluate completed palliative pathways (see also 10.1.1). The inclusion criteria can be seen in Table 5.2., Chapter 5. The home care services in the 26 municipalities of Aarhus County and The Specialist Palliative Team at Aarhus University Hospital were asked to identify possible informants meeting the inclusion criteria. The specialist palliative team identified two possible informants who were both invited and accepted to participate. The other 16 possible informants were identified by the home care services. In all, we invited 18 relatives of whom 14 agreed to be interviewed (78 %) and the three interview groups comprised three, five and six informants, respectively. We ensured a wide range of demographic characteristics and in time since the death of the patient (3). Informant demographics can be seen in Tables 5.3 and 5.4, Chapter 5. The four relatives, who did not participate, were not able to participate at the given dates.

The informants in Paper 2 (Chapter 6), GPs, hospital consultant physicians and CNs were all invited by mail to participate in the interviews. Home-care services in the 26 municipalities of Aarhus County were asked to identify CNs as possible informants. The GPs and the consultant physicians were found in the database of Danish Medical Association (3). In all three groups we chose the informants purposefully, trying to ensure a wide range of demographic characteristics (3). Demographics of the informants can be seen in Table 6.1, Chapter 6. We invited 74 GPs, 17 consultant physicians and 21 home-care nurses, and 23 (31 %), five (29 %) and 15 (71 %) agreed to be interviewed, respectively. The main reason given for not participating was lack of time.

4.2.1.2. The interviews All group interviews were conducted at The Research Unit for General Practice, University of Aarhus. The group interviews were conducted by MAN, who was supervised by the co-authors, JS or FO. The group interviews, which were tape-recorded with the informants‟ consent, were guided by a semi-structured topic guide based on clinical experiences and literature studies (Appendix A - E). Overall themes in interviews with bereaved relatives were „The health professionals‟ management „ (i.e. „The role and management of the primary health care professionals‟, „Discharge from hospital care to home care‟, „Interpersonal relation‟ and „Organisation of palliative care‟) and „The role of the

25


relatives‟ (Table 5.5, Chapter 5). Overall themes in interviews with professionals were „The health professionals‟ interpersonal relations „, „Experiences of professionals‟ barriers‟, „The effort of health professionals‟ and „Organisation of palliative care‟ (Appendix E). Open-ended questions were used and we took our starting point in the informants‟ own experiences, asking what was good and what was bad according to a given issue, and how it could be improved. In this way we revealed barriers and facilitators of the issues. This way of asking questions was inspired by the SWOT analysis (Acronym for: Strengths, weaknesses, opportunities and threats) (4;5). The informants were then encouraged to speak freely and to raise issues of importance to them. A summary was given at the end of each group interview to obtain an immediate validation of the themes identified by the researcher. Before the bereaved relatives left the room of the interview, we incited them to contact us if any issues or emotional reactions appeared after the interview. No informants contacted us subsequently. The interview guide was further developed according to the themes emerging during the analyses conducted after each group interview.

4.2.1.3. Analysis of the qualitative studies The group-interviews were transcribed verbatim, the first interview by one of the authors (MAN), and the rest by trained secretaries, and all transcripts were read repeatedly validating the transcription and getting an overall impression before initial coding (6).

We used qualitative description (QD) to analyse the interviews, since our study was a mixed method study, since research question A (Chapter 3) was to describe the experiences, suggestions for improvement and wishes to the function and organisation of the involved relatives and professionals, and since we wanted to use the results to qualify our quantitative survey and QD is particular useful for all of these purposes (7).

All meaningful text units were identified and coded. Examining information in the text units, they were gathered in groups of main codes and labelled with a name. Again the information in each main code was examined and they were grouped into categories, and subsequently into main categories. In this way we allowed the main categories to evolve from the data instead of imposing a framework a priori (8). No new categories emerged from the analysis of the last group interviews. Agreement on the analysis was reached among the authors after thorough discussion. The software package NVivo (9) was used to assist in the coding, sorting, and retrieval of data.

26


4.2.2.

Quantitative surveys

We chose a combined register and questionnaire study as the method to answer research question B (see Chapter 3).

4.2.2.1. The questionnaires Three questionnaires were developed: One for GPs (Appendix K), one for CNs (Appendix M) and one for bereaved relatives (Appendix O). Our preceding interviews, clinical experience and literature studies were used to identify the themes of the questionnaires (10). At the time of conducting our survey, no validated questionnaire directly applicable to our study was found in the literature search (See also 10.3.5., Chapter 10 and appendix F). However, former questionnaires designed to evaluate patients‟ satisfaction with family physicians services were used as inspiration (11-13). The themes in the final questionnaires can be seen in Appendix G.

4.2.2.2. Pilot testing When agreement of the questions in the questionnaires was achieved in the research group, the questionnaires were reviewed by the scientific staff and lay persons at the Department and Research Unit for General Practice, Aarhus University. The questionnaires were revised before a regular pilot test was performed. The questionnaires for relatives were pilot-tested among the 14 participants who had participated in the interview study. The GP questionnaire was pilot-tested among 30 GPs in the County of Northern Jutland and the 15 CNs who had participated in the interview study were asked to complete and comment on the nurses‟ questionnaire. Only minor changes to the phrasings of the questions were made after the pilot test and no questions or themes were dropped or added.

4.2.2.3. Study population and sampling We aimed to include adults in Aarhus County who died from cancer from 1 st March to 30th November 2006 and who had received some palliative home care either from GPs, community nurses or a palliative specialist team. Since no database on palliative patients was available and since the Danish Register of Causes of Death was still not updated on deaths in 2006, we sampled the patients by combining several official registers. In Denmark, all citizens are registered with unique civil registration numbers (14) and by means of these numbers questionnaire data were linked to health register information.

27


From the county hospital discharge registry we identified 29,043 individuals above 18 years who had been registered with at least one cancer diagnosis (ICD-10) (excluding non-melanoma skin cancers) during the period November 2006 and ten years back (Figure 4.1). In December 2006, using the Civil Registration System database, we identified 813 patients among the 29,043 who had died from 1st March to 30th November 2006 (nine months). From the regional health authority‟s register we identified their GPs. Eight (1.0%) patients were not registered with a GP and 18 (2.2%) had moved out of the county after having been diagnosed, leaving 787 deceased cancer patients. The GPs, CNs and relatives of these patients were included in the study at the time.

In late 2008 data from the Danish Register of Causes of Death for 2006 was available. Merging the information of cancer deaths with our study-database, we excluded 188 patients who were not registered with cancer as the reason of death, which reduced our study population to 599 deceased cancer patients (Figure 4.1).

Figure 4.1. Sampling of cases included in the study using health registers.

Sampling

29043 Patients with a cancer diagnose in Aarhus County within 10 years before17th November 2006

28230 Alive or dead before 1st March 2006 or < 18 years old

813 Cancer patients in Aarhus County died between 1 st March and 30th November 2006 8 has no GP registered 18 moved from Aarhus County since cancer diagnosis

787 Cases of deceased cancer patients in Aarhus County registered with a GP

188 patients not registered with cancer as cause of death

599 Cases of patients died from cancer in Aarhus County registered with a GP

28


4.2.2.4. Data collection A GP questionnaire was sent to the general practice of each patient included in the study-database (Appendix K). The cover letter (Appendix J) was addressed to the GP in the general practice who was most familiar with the patient. The GP was asked first about cause of death and if palliative care had been provided in the patient‟s home, secondly the GP was asked to state the closest relative and to decide whether the relative would be able to fill out a questionnaire, and if not to state the reason for this decision. Furthermore, the GP was asked to state a name on the CNs involved.

A questionnaire was also sent to the CN, or if the GP did not state the name of the nurse, it was sent to the home care centre in the area of the patient‟s address (Appendix M). If the GP did not answer or did not want to participate, a CN questionnaire was also sent to the home care centre in the area of the patient‟s address. The cover letter (Appendix L) was addressed to the CN who was most familiar with the patient. The nurse was asked to state the closest relative and to decide whether the relative would be able to fill out a questionnaire, and if not to state the reason for this decision.

After receiving the GP and CN questionnaires, a questionnaire was sent to the relative that the GP or CN had named (Appendix O). The cover letter to the bereaved relative (Appendix N) was formulated in a way that if the person identified was not a relative to the deceased, he or she could contact the Research Unit directly. However, nobody contacted us with this information. If one of the GPs and CNs advised against sending the relative a questionnaire, it was not sent. The closest relative was defined by the professionals, but if they stated that we could contact the relative but without providing name or address, we sent the questionnaire to the relative in the following order: Spouse, child above 18 years, oldest sibling and parent (data from the Civil Registration System database). The flowchart of the total quantitative study is seen in Figure 4.2.

4.2.2.5. Data entry The questionnaires were designed and processed in the computer programme Teleform Enterprise Version 8.0 (Cardiff software Inc., San Marcos, CA, USA) for data capture by optical scanning. Handwritten data were coded by MAN. An assistant scanned all returned questionnaires, and whenever the programme was in doubt of an answer, the questionnaire was examined thoroughly by the assistant and MAN. A previous study has documented the accuracy of this processing (15). Data were transferred to the statistical software program STATA 10 (16) and checked for errors. If

29


errors were encountered, the original questionnaire was inspected and the study-database entry corrected. Figure 4.2. Flowchart of questionnaire study

General practitioners

Community nurses

Bereaved relatives

(GPs)

(CNs)

(BRs)

599 GP questionnaires sent out

194 CN questionnaires sent out

56 cases

194 GP non-responders 55: GP not wanting participation 105: GP did not answer 9: GP died, were sick or retired 25: GP stated that patient did not die from cancer

72 Excluded by GP 72: No palliative home care

95 CN non-responders 74: CN did not fill in questionnaire 2: CN no time to participate 1: CN stated that patient did not die from cancer 4: CN left the nursing centre 13: No CNs knew the patient 1: Head of nursing centre not wanting nurse to participate 43 Excluded 30: No palliative home care 13: Nursing home resident

56 CN questionnaires filled in

333 GP questionnaires filled in 106 cases excluded by GP: 24: Relative mentally disabled 2: Relative linguistically disabled 3: BR had alcohol problem 2: Patient was divorced 1: No relatives 1: Relative living abroad 1: Relative did not participate 1: Patient not registered with GP 71: No reason stated

15 cases excluded by CN 2: BR mentally disabled 1: BR had alcohol problem 1: Patient was divorced 11: No reason stated

41 BR questionnaires sent out 10 BR non-responders 10: BR did not respond

31 BR questionnaires filled in

333 CN questionnaires sent out 125 CN non-responders 1: CN stated that patient did not die from cancer 13: CN left the nursing centre 7: No CNs knew the patient 10: Head of nursing centre not wanting nurse to participate 5: CN no time to participate 89: CN did not respond 63 Excluded 63: No palliative home care

145 CN questionnaires filled in 227 BR questionnaires sent out 70 Non-responders 8: Returned unanswered 2: BR too sick 55: BR did not respond 5: Postal services did not find BR on the given address 4 Excluded 4: BR: Not relevant course

153 BR questionnaires filled in

30


4.2.2.6. Analysis of the quantitative studies In the descriptive data, a p-value < 0.05 level was chosen to decide statistically significant differences between distributions within variables in different groups (e.g. responders versus nonresponders). In Paper 3 „Home death‟ was defined as the outcome measure and associations with patient, GP and palliative pathway variables were calculated. Place of death in 2006 is defined in The Danish Register of Causes of Death as „home‟, „nursing home‟, „hospital or hospice‟ and „death in other places‟ (Se also 10.3.5.1) Since the patient‟s own GP provides care for the patient at home as well as in nursing homes, we defined „home death‟ as death either at home or at a nursing home. GPs are only to a limited extend involved in palliative pathways at hospices. In Papers 4 and 5, ‟A successful palliative pathway‟ was defined as the outcome measure and associations with patient, relative, palliative pathway and GP and CN variables, respectively were calculated. As for the evaluation of the palliative pathway, the relatives were asked: „How, in your own words, was the entire period at home during which the deceased was dying compared with how you feel it should have been?‟ with the options: „Very well‟, „Well‟, „Fairly well‟, „Bad‟, „Very bad‟. In the analysis answers were dichotomized into „not successful‟ („Fairly well‟, „Bad‟, „Very bad‟) and „successful‟ („Very well‟, „Well‟)).

The variables used in the analysis can be seen in Appendices H and I. The reasons for choosing the specific variables in the models are discussed in Chapter 10.

Unadjusted and adjusted associations were calculated. When patients are treated by the same health care person, certain conditions are the same among this group of patients, and when analysing on the total data one has to adjust for the possibility of patient clustering (18). Unfortunately, in our study we could not adjust for clustering of patients in relation to each GP, since most of the GPs did not state their name on the questionnaire. Instead we adjusted for clustering at practice level in the models in Papers 3 and 4. In Paper 5 we adjusted for clustering of patients of the same CN. Prevalence ratios (PRs) with 95% confidence intervals (95% CI) were used as a measure of association as the prevalence of the outcomes in our data was very high (19). Usually, when outcomes are binary, logistic regression is used for the data analysis and odds ratios are frequently reported in situations where incidences or prevalence ratios are estimable. The use of odds ratios is absolutely correct; however, when working with frequent outcomes, the odds ratios can strongly overestimate the association (20;21). PRs were calculated with

31


generalised linear models (GLM) with log link and the Bernoulli family, and when the model could not converge, we used the Poisson regression model (20;22). The model fit for multivariate models was assessed using the Hosmer-Lemeshow test for goodness of fit (23). The variables were assessed for collinearity (Pearson's correlation coefficient > 0.4) and multicollinearity (Variance inflation factor < 10) (24;25). Data were analyzed using STATA 10 (16). 4.3. Economic incentives No payment was given for participation in the focus groups, but transport expenses were covered for all participants. Remuneration was made possible (360,- DKR, approximately 48 Euro per completed GP questionnaire) through funds provided by The Multipractice Study Committee (585-04/2072) and The Aarhus County Research Fund for the Clinical Development and Research in General Practice and across the Primary and Secondary Health Care Sectors (4-01-3-04) Nurses and relatives received no payment for filling in questionnaires. 4.4. Approvals According to the Scientific Committee for the county of Aarhus, the Biomedical Research Ethics Committee System Act does not apply to this project as no active intervention was involved in the study (j.no. 2005-2.0/14). The study was recommended by the Multi-Practice Committee of the Danish Society of General Practitioners and the Organisation of General Practitioners in Denmark (MPU 6-2005). The overall study was approved by the Danish Data Protection Agency (j.no. 200541-4967) and the quantitative study by the Danish National Board of Health (j.no. 7-505-291007/1).

32


4.5. References of Chapter 4 (1) The Danish National Board of Health. The Danish Register of Causes of Death [In Danish]. 2008. (2) de Fine Olivarius N, Hollnagel H, Krasnik A, Pedersen PA, Thorsen H. The Danish National Health Service Register. Dan Med Bull 1997;44(4):449-53. (3) Crabtree BF, Miller WL. Doing qualitative research. Newbury Park: Sage Publications; 1992. (4) Johnson G, Scholes K. Exploring Corporate Strategy. 4th edition ed. Prentice Hall; 1997. (5) Saaty RW. The analytic hierarchy process and SWOT analysis-what it is and how it is used. Math Model 1987;9:161-78. (6) Kvale S. Interviews: An Introduction to Qualitative Research Interviewing. 1 ed. Thousand Oaks: SAGE Publications; 1996. (7) Sandelowski M. Whatever happened to qualitative description? Res Nurs Health 2000 Aug;23(4):334-40. (8) Pope C, Ziebland S, Mays N. Qualitative research in health care. Analysing qualitative data. BMJ320(7227):114-6. (9) Guizzo BS, Krziminski CO, de Oliveira DL. [The software QSR Nvivo 2.0 in qualitative data analysis: a tool for health and human sciences researches]. Rev Gaucha Enferm 2003 Apr;24(1):53-60. (10) Olesen F, Mainz J. Krav til spørgeskemaer II. Hvorledes udarbejdes et spørgeskema? Månedsskr Prakt Lægegern 1994 Nov;72(Nov.):1461-8. (11) Nielsen JD, Palshof T, Olesen F. [Cross-sectorial cooperation regarding cancer patients in a recently started care program. Ideas and themes based on focus group interviews with general practitioners and oncologists]. Ugeskr Laeger 1999 Apr 5;161(14):2074-8. (12) Heje HN. Patient evaluation in general practice. Methodological aspects, influence of patient and GP characteristics and the GPs' experiences with the evaluations [thesis]. 1 ed. Aarhus: Research Unit and Department of General Practice, Faculty of Health Sciences, University of Aarhus; 2006. (13) Mikkelsen TH, Søndergaard J, Sokolowski I, Jensen AB, Olesen F. Cancer survivors' rehabilitation needs in a primary healthcare context. submitted 2008. (14) Pedersen CB, Gotzsche H, Moller JO, Mortensen PB. The Danish Civil Registration System. A cohort of eight million persons. Dan Med Bull 2006 Nov;53(4):441-9. (15) Jørgensen CK, Karlsmose B. Validation of automated forms processing. A comparison of Teleform™ with manual data entry. Comput Biol Med 1998 Nov;28(6):659-67. (16) Stata Statistical Software: Release 9.0. College Station, TX: StataCorp LP; 2005.

33


(17) Centeno C, Clark D, Lynch T, Racafort J, Praill D, De Lima L, et al. Facts and indicators on palliative care development in 52 countries of the WHO European region: results of an EAPC Task Force. Palliat Med 2007 Sep;21(6):463-71. (18) Donner A, Klar N. Design and Analysis of Cluster Randomisation Trials in Health Research. 1 ed. London: Hodder Arnold; 2000. (19) Clayton D, Hills M. Statistical Models in Epidemiology. 1 ed. Oxford: Oxford University Press; 1993. (20) Barros AJ, Hirakata VN. Alternatives for logistic regression in cross-sectional studies: an empirical comparison of models that directly estimate the prevalence ratio. BMC Med Res Methodol 2003 Oct 20;3(1):21. (21) Thompson ML, Myers JE, Kriebel D. Prevalence odds ratio or prevalence ratio in the analysis of cross sectional data: what is to be done? Occup Environ Med 1998;55(4):272-7. (22) Zou G. A modified poisson regression approach to prospective studies with binary data. Am J Epidemiol 2004 Apr 1;159(7):702-6. (23) Hosmer DW, Lemeshow S. Applied logistic regression, second edition. New York: John Wiley & Sons; 2000. (24) Armitage P, Berry G, Matthews JNS. Statistical Methods in Medical Research. Fourth edition. 4 ed. Oxford: Blackwell Science; 2005. (25) O'Brien RM. A Caution Regarding Rules of Thumb for Variance Inflation Factors. Quality and Quantity 2007;41(5):673-90.

34

Chapter 5. Paper 1

CHAPTER 5 PAPER 1 Palliative care for cancer patients in a primary health care setting Bereaved relatives' experience. A qualitative group interview study Neergaard MA, Olesen F, Jensen AB, Sondergaard J BMC Palliat Care 2008;7:1

35

Chapter 5. Paper 1

5.1. Abstract Background: Knowledge about the quality and organisation of care to terminally ill cancer patients with a relatives' view in a primary health care setting is limited. The aim of the study is to analyse experiences and preferences of bereaved relatives to terminally ill cancer patients in a primary care setting to explore barriers and facilitators for delivery of good palliative home care. Methods: Three focus group interviews with fourteen bereaved relatives in Aarhus County, Denmark. Results: Three main categories of experience were identified: 1) The health professionals' management, where a need to optimize was found. 2) Shared care, which was lacking. 3) The relatives' role, which needs an extra focus. Conclusion: Relatives experience insufficient palliative care mainly due to organizational and cultural problems among professionals. Palliative care in primary care in general needs improvement and attention should be drawn to the "professionalization" of the relatives and the need to strike a balance between their needs, wishes and resources in end-of-life care and bereavement.

5.2. Background A palliative course of disease often involves several different parts of the healthcare system, and the collaboration or shared care therefore becomes an important issue (Table 1)[1]. One group of health care professionals often involved is the primary care professionals, e.g. the general practitioners (GPs). GPs are expected by their patients, policy makers and health organisations to be involved in palliative home care [2,3]. Many GPs consider palliative care to be typical GP work and one of the best parts of the job [4]. The traditions and values of general medicine corroborate the intentions of palliative care [5,6], e.g. in seeking a continuing relationship between patient, family and professionals, the continuity of care [7] and the ability to perform care in co-operation with other health professionals [8]. GPs' management of palliative care, especially symptom control, has often been a subject of debate. It has been suggested that poor delivery of care in the primary sector may be a major reason why the majority of cancer patients die in hospitals despite their and their relatives' preference for a home-death [9-11]. Hence, there is a profound need for a deeper insight into barriers and facilitators for delivery of good palliative home care. Evaluation of palliative care has often focused on clinical measures, e.g. the professionals' ability to provide symptom control, at the expense of the cancer patient's and relatives' perception of the

36

Chapter 5. Paper 1

total care delivery in the terminal phase. However, the bereaved relatives have invaluable experience of primary health care delivery from the terminal diagnosis until bereavement. We therefore aimed to analyse experiences and preferences of bereaved relatives to terminally ill cancer patients in a primary care setting to explore barriers and facilitators for delivery of good palliative home care. 5.3. Methods We conducted three qualitative, semi-structured group interviews with fourteen relatives of recently deceased cancer patients in Aarhus County (640 000 inhabitants), Denmark (5.5 million inhabitants) from June to October 2005. Setting The Danish Health care system is financed through taxes and provides free and equal access to health care services. More than 98% of the Danes are registered with a GP and receive free medical care [12]. Danish GPs provide most of the health care themselves and act as gatekeepers for access to specialist treatment.

Specialist palliative teams with consultants, nurses and other healthcare providers are attached to the three major hospitals in the County of Aarhus, and the GPs can refer patients to these teams or ask the specialists for advice. These teams are only available during normal daytime hours.

Palliative home care is divided into a basic and a specialist level [13]. The basic level professionals are the primary care sector, e.g. the GPs, the home care nurses and home care services. The professionals involved in specialist level home care are the primary care sector plus a palliative specialist team, working either as consultants or as active professionals in the patient's home.

At the moment there is no formalised education in palliative care of significance at the pre-graduate level of doctors in Denmark. At the level of specialising toward General Medicine there is a oneday course of theoretical aspects of palliative care. The Danish GPs can participate in noncompulsory courses in palliative care, but these courses are in competition with courses in all other aspects concerning general medicine.

37

Chapter 5. Paper 1

Sample The informants were close relatives of recently deceased cancer patients, and the inclusion criteria can be seen in Table 2. The criteria that the patient died less than one year before the inclusion of the relatives was used to minimise memory decay.

The home care services in the 26 municipalities of Aarhus County and the specialist palliative team at Aarhus University Hospital were asked to identify possible informants meeting the inclusion criteria. The specialist palliative team identified two possible informants who were both invited and accepted to participate. The other 16 possible informants were identified by the home care services. In all, we invited 18 relatives of whom 14 agreed to be interviewed (78%). Four relatives were not able to participate on the proposed dates, which gave us the problem that only three informants participated in the first interview. However this did not influence on the interaction in the groups that were just as spirited in all three groups. Hence, the three interview groups comprised three, five and six informants, respectively (Table 3). We ensured a wide range of demographic characteristics and in time since death of the patient (Table 4) [14]. One participant (nr. 13) were only one month past bereavement. This relative had specifically stated to the home care service that he wanted to help others in the same situation if possible.

Group interviews All group interviews were conducted at The Research Unit for General Practice, The University of Aarhus. The group interviews were conducted by MAN, who was supervised by the co-authors, JS or FO. All three are specialists in general practice and JS and FO are GPs. MAN has completed courses in interview technique and analysis, and JS and FO are experienced in qualitative research methods.

The group interviews, which were tape-recorded with the informants' consent, were guided by a topic guide based on clinical experiences and literature studies (Table 5). The topic guide covered the following main areas: The health professionals' management, including the role and management of the GP, the organisation, interpersonal relation and co-operation in palliative care [15]. The guide was further developed according to the themes emerging during the analyses conducted after each group interview, e.g. after the first group interview the role of the relatives was added to the topic guide. No new themes were added after the two following interviews. Openended questions were used. The informants were encouraged to speak freely and to raise issues of importance to them. The group interviews lasted from 111 to 129 minutes. A summary was

38

Chapter 5. Paper 1

given at the end of each group interview to obtain an immediate validation of the themes identified by the researcher. Analysis The group interviews were transcribed verbatim by a trained professional secretary. All transcripts were read simultaneously with the sound of the tapes by MAN to ensure correctness of the transcription and were read repeatedly by the authors to get an overall impression of the material before the initial coding [16]. A qualitative description approach was used for the analysis[17]. Following this analytic approach we enhanced rigour by focusing on the following strategies: 1) authenticity, the attention to the voices of participants and the ability to remain true to the phenomena under study, 2) credibility, a reflexion of how believable results are, 3) criticality, the critical appraisal of every decision made throughout the research process and 4) integrity, demonstrated by ongoing reflection and self-criticality of the researcher [18].

All meaningful text units were identified and coded. The codes were subsequently grouped into relevant categories by MAN and at the end seven categories were identified, which again were grouped into three main categories. In this way we allowed the main categories to evolve from the data instead of imposing a framework a priori [19]. No new categories emerged from the analysis of the third group interview.

Agreement was reached in the group of authors after thorough discussion of the initial coding, the categories and the main categories. We used the software package NVivo, ed. 6 (QRS international, Melbourne, Australia) to assist in the coding, sorting and retrieval of data. 5.4. Results Seven categories were identified and grouped into three main categories: the health professionals' management, shared care and the relatives' role (Figure 1).

39

Chapter 5. Paper 1

Figure 5.1. From categories to main categories

Main Categories

The professionals’ management

Categories

The professionals‟ knowledge The professionals‟ behaviour and communication skills The professionals‟ contact with patients and families Responsibility

Shared care

Inter-professional culture Inter-professional communication

The relatives’ role

The relatives‟ role

The health professionals' management The analyses revealed that the relatives found three aspects of the professionals' management of palliative home-care to be important: The professionals' knowledge, their behaviour and communication skills and their contact with patient and families.

The health professionals' knowledge Some relatives experienced that their GP was proficient in handling symptom control and other aspects of the treatment of terminally ill cancer patients.

"As soon as we were there, the GP looked things up on the computer and she knew all the time what was going on... she really did." (Informant 10)

Others felt that the GPs neither possessed sufficient knowledge nor sought assistance from the palliative care specialists.

40

Chapter 5. Paper 1

"... when the GP cannot manage, he should refer to someone who can. Our GP is the kind of person who thinks he knows best." (Informant 2)

The health professionals' behaviour and communication skills Empathy and time to listen were important qualities to the relatives.

"... a good thing about my GPs is ... both the female and the male GP in my practice are very competent and have the skills necessary to talk to the patients. They have a way with people." (Informant 1)

Some of the doctors, nurses and other health-care providers were able to communicate competently with the patients and their relatives, while others did not seem to possess these skills.

"And as I told you before, our own GP actually referred to me as "the widow" already in advance. I think that was a very improper remark." (Informant 2)

"They must have a flair for breaking bad news. Some doctors think you have to know everything, whether you want it or not" (Informant 6)

The professionals' contact with patients and families The relatives requested that health professionals be more active in establishing personal contact with the family.

"... perhaps the GP could do a little outreach work and phone us asking how we were doing ... the GP did not have the slightest idea of what was going on." (Informant 9)

The relatives described how hard it was always to be the one to take initiative.

"I think it has been very hard work the way we constantly having to keep abreast." (Informant 12)

Some GPs visited the patients at home on their own initiative, whereas others refused to make home visits at all, e.g. because they lacked time. GPs' home visits were highly appreciated because of the profound feeling of security it gave the patient and relatives.

41

Chapter 5. Paper 1

"The GP spontaneously came to visit us every day until it was over. He came and asked if there was anything he could do. And it was an incredible relief not having to call him first... Then, I really felt that he gave me all the support a doctor could give."(Informant 13)

The informants found it important to know whom to contact during the out-of-hours periods, e.g. evenings, nights, weekends and national holidays. In some cases, the GP gave the family his or her private telephone number for such situations, which was highly appreciated, while others had to use the on-call GPs who were unfamiliar with the patient. The informants also appreciated whenever the health care professionals made an active contact after bereavement, because they often felt left in limbo.

Shared care Three categories emerging from the analyses addressed shared care between the different professionals involved: Professional responsibility, inter-professional culture and inter-professional communication.

Professional responsibility Often, the GPs were not even informed about the patients' discharge from hospital to palliative care at home since they received the discharge letters very late. This failure to pass on responsibility made patients and relatives feel as though they were "left in limbo".

"You see, at first, when we were told by the hospital physician that there was nothing more to do but control the pain, at that time, I felt left alone... ... We were simply left in limbo." (Informant 2)

During the palliative course of disease in the patient's home, the relatives wanted one of the health-care professionals to take on the role of team-leader, i.e. to co-ordinate care and take control of the situation, and they preferred this professional to be the GP. They appreciated the continuity the GP provided, i.e. the GP's acquaintance with the patient and the relatives both before and after the cancer diagnosis and the GP's knowledge of the specific course of disease.

"Well, I think the GP should have the opportunity to be in charge of the coordination..." (Informant 5)

Inter-professional culture

42

Chapter 5. Paper 1

In many cases the relatives experienced a lack of respect between the professionals especially between primary health care professionals and hospital professionals.

"... it was not a comforting message we got, that our GP had referred my husband to the hospital too late ... they told us that straight out. A hospital physician told us that they always get the patients too late from that municipality." (Informant 11)

The insinuation of other professionals' incompetence gave the relatives a feeling of insecurity and they did not know whom to trust.

Inter-professional communication Many relatives experienced communication problems between the hospital and the primary care sector.

"Honestly, I sometimes think that there is a wide gap between the GPs and the hospitals ... Well, I wish they had closer contact." (Informant 1)

Furthermore, the relatives often experienced that the GPs and the home care nurses did not receive the necessary information from the hospitals. The relatives did not understand why GP and hospital records were not accessible on-line to all the professionals involved.

"Our GP asked me to call him, when I got news from the hospital, because it often took too long before he got the information himself..." (Informant 6)

Poor communication between GPs and home care nurses was perceived as a major problem. In some cases the relatives experienced that the nurses were unable to reach the GP by phone, and the relatives saw the home care nurses' frustration.

"But I know that the home nurses were very frustrated because they couldn't contact our GP." (Informant 14)

The role of the relatives All relatives stated that before the patient's death both they and the patients wished that end of life care and death should take place at home. Despite that, only half died at home (Table 4).

43

Chapter 5. Paper 1

All relatives stated that they were grateful to have had the opportunity to participate in the care of their terminally ill relatives.

"They should be better at explaining people what it actually means to take your loved ones home and care for them there. How much help you can actually get ... And it is not dangerous because the professionals are there for you, all of them." (Informant 12)

However, an experience shared among the relatives was an unspoken pressure from the professionals

to

be

"semiprofessionals"

themselves

rather

than just

relatives,

i.e.

a

"professionalization" of the relatives took place.

"My husband's family noticed that when I was with my husband, I did not act as his wife, but as a professional when it came to care and when the GP visited..." (Informant 6)

"I insisted on being allowed to be just a relative. And that was difficult although I told them I had a right to that." (Informant 14)

The pressure of being a "semi-professional" was especially a problem when the relatives happened to be health care professionals themselves, when the relatives had to provide intimate care for the patient and when the patient needed 24-hour care.

"... you have to be available twenty-four hours a day. My eighty-year-old dad cried at night when it became too much for him taking care of my terminally ill mother. "I cannot do this anymore", he cried. And then I had to take over..." (Informant 14)

In one case the relative stated that the burden of the role of a "semi-professional" meant that the family gave up on the possibility of home death. This left them with a feeling of inadequacy as relatives since they could not fulfil the patient's wish for a home death. 5.5. Discussion Since the relatives felt an unspoken pressure from the professionals to be "semi-professionals" themselves and a "professionalization" of the relatives took place, the relatives asked for health professionals who had the necessary knowledge, who were readily available to the patients and their families and more active in establishing personal contact with them. They also described both organizational and cultural problems in the health care system.

44

Chapter 5. Paper 1

Focus group interviews are pertinent, as individual interview to explore the informants' perspectives and experiences concerning a specific topic. The force of group interview is that the interaction in the group can be utilised as a factor of development of views and description among the participants and can make new themes or aspects of the topic emerge [20]. In this way one gets a broad elaboration of the topic. In our study the group setting made new themes emerge but it also gave the relatives a valuable feeling of shared experience and of being able to support each other in the interview situation.

Interviews with terminally ill patients could have added important information to our study, but arranging interviews with terminally ill patients would have entailed ethical and practical difficulties disproportionate to the added value. Choosing close relatives who had participated in the palliative care as informants gave us an insight into the patients experience as well and it gave us the opportunity to explore the role of the relatives.

There was a surprising polarization in the experiences of GPs' management of palliative care. However, most experiences were negative in nature. This may be due to the fact that the relatives who agreed to participate in the group interview were the ones who needed to express their bad experiences which the authors feel that there is a tendency towards in the Danish society or of course might be due to an actual bad performance of the GPs involved. However, this can not be determined in a qualitative study, and will require a follow-up with a quantitative study.

The fact that the interviewers (MAN, JS, and FO) were all specialists in general practice could bias the group interviews and the analyses, but all authors were aware of this potential problem [14].

In earlier studies of the professionals' management of palliative care some relatives were not satisfied with symptom control in primary care, but they still rated the overall GP care as very good [10]. The overall satisfaction with the GPs was significantly associated with symptom control, time to listen and accessibility [21]. GP home-visits may increase terminally ill cancer patients' possibility of dying at home [22]. In our study, continuity of care and good knowledge of the whole course of disease were among the reasons why the GPs were valued despite their at times poor management of palliative care.

Most studies examining the relatives' role have described that the relatives become emotionally and physically exhausted [11,23]. We found that an important reason for this could be their "professionalization" and the absence of support from health care professionals, particularly after

45

Chapter 5. Paper 1

bereavement. It is important that the professionals help the relatives strike a balance between their needs, wishes and resources [23,24]. This balance can be difficult to find in end-of-life care situations, where the relatives feel obliged to fulfil the wishes of the dying relative.

The importance of communication between GPs and specialists and the request for shared care have also been found in other palliative studies [25-28]. Also in line with our findings, these studies emphasize that continuous contact with the GP can be maintained by referring the patients back to their GPs after hospitalization, by ensuring good communication between specialists and GPs and by providing a clear definition of the GP's role [26]. However, a fundamental prerequisite for successful shared care is the presence a coordinator [25]. The need for and lack of shared care is well-described, but the reasons for this lack of shared care and ways of improving it are less wellexplored. From our study it appears that one of the main barriers to optimizing shared care may be the lack of a shared care culture among health professionals and the lack of a home-care team leader in palliative care.

The issue of "professionalization" of the relatives was an important finding of our study and the extend and significance of this aspect should be further elaborated in future research. It is a challenge to all professionals involved in palliative home care to help relatives to strike a balance between the needs, wishes and resources f the relatives.

Another main challenge to the health care system is to create a shared care culture characterised by cooperation and mutual respect among professionals within the health care system. Further research in this subject is in place, but it seems that the respect and knowledge among different types of professionals are to some extend missing, and maybe shared care is missing as a focus already in the education and training of doctors, nurses and other professionals involved.

Hence, in order to optimize palliative home care, further research is required into how to develop a shared care culture, how to improve communication along the cancer journey, how to delegate professional responsibility and on relatives role. A questionnaire survey among relatives, home care nurses and GPs is currently being conducted in order to give quantitative estimates of the distributions of opinions and experiences. 5.6. Conclusion Our study indicates that relatives experience insufficient palliative care, mainly due to organizational and cultural problems among professionals. There is a lack of shared care.

46

Chapter 5. Paper 1

Palliative care in primary care in general needs improvement, both in terms of professional knowledge, attitude, availability and communication skills. Furthermore, attention should be drawn to the "professionalization" of the relatives and the need to strike a balance between their needs, wishes and resources in end-of-life care and bereavement. Competing interests The author(s) declare that they have no competing interests. Authors' contributions MAN participated in the design of the study, conducted the interviews and the analysis and drafted the manuscript. FO and JS participated in the design of the study, as supervisors in conducting the interviews and participated in discussions of categories. ABJ participated in the design of the study. All authors read and approved the final manuscript. Acknowledgements The authors would like to express profound gratitude to all informants for participating. We thank the home care nurses in Aarhus County and The Palliative Team at the Aarhus University Hospital who helped us identify potential informants. The study was funded by The County of Aarhus and The Danish National Research Foundation for Primary Care. The study has been approved by the local ethical committee.

5.7. Tables Table 5.1. Definition and implications of Shared care Shared care »Shared care is the joint participation of general practitioners and hospital consultants in the planned delivery of care for patients with a chronic condition, informed by an enhanced information exchange over and above routine discharge and referral letters« (12) Shared care demands knowledge of the abilities and the qualities of the cooperating partners, but also accept of each others‟ roles in the delivery of palliative care(1). Table 5.2. Inclusion criteria Inclusion criteria for  bereaved relatives     

The relative must be closely related to an adult deceased cancer patient The relative must have been involved in the palliative care of the deceased patient The cancer patient must have died less than one year before the inclusion The relative must be 18 years or more The relative must speak and understand Danish The relative must be mentally and physically able to participate in a focus group interview

47

Chapter 5. Paper 1

Table 5.3. The informants Nr of Informant

Group interview number

Relation to deceased

Gender

Age of relative

Age of patient

1

1

Daughter

Female

40

2 3 4 5 6 7 8

1 1 2 2 2 2 2

Spouse Sister Spouse Spouse Spouse Sister Spouse

Female Female Female Male Female Female Female

9

3

Daughter

10 11 12 13 14

3 3 3 3 3

Spouse Spouse Daughter Spouse Daughter

Table 5.4. Demographics Deceased cancer patients: (14 patients (100 %))

Place of death: (14 patients (100 %)) Time from patient‟s death to interview with relative, median (range) Relatives interviewed (14 informants (100 %))

Type of cancer

Place of death

64

Pancreas

55 61 75 60 46 40 73

60 63 76 56 56 60 83

Kidney Mamma Urinary Ovarian Pleural Brain Prostate

Female

40

72

Lung

Male Female Female Male Female

75 70 44 45 49

65 78 78 44 83

Lung Lung Lung Colon Malignant melanoma

Daughter‟s home Home Hospice Home Hospital Home Hospice Nursing home Daughter‟s home Home Hospital Hospital Home Nursing home

Sex Age at time of death, median (range) Patients in contact with a specialist palliative care team Home Hospital Hospice Short-term nursing home

Months from death to interview 5 5 2 9 6 9 15 10 6 16 3 3 1 7

Women 7 patients (50 %) Men 7 patients (50 %) 64.5 years (44; 83) 11 patients (79 %) 7 patients (50 %) 3 patients (21 %) 2 patients (14 %) 2 patients (14 %) 6 months (1; 16)

Sex Family relationship to patient

Age at interview, median (range) Place of residence

Women Men Spouse Child Sibling Urban Semiurban Rural

11 informants (79 %) 3 informants (21 %) 8 informants (57 %) 4 informants (29 %) 2 informants (14 %) 52 years (40; 75) 3 informants (21 %) 5 informants (36 %) 6 informants (43 %)

48

Chapter 5. Paper 1

Table 5.5. Topic guide themes Themes Issues The health The role and management of the GP professionals‟ management

Discharge from hospital care to home care

Interpersonal relation

Organisation of palliative care The role of the relatives

Examples of questions How where the management of the GP involved? What was good / bad? How could it be improved? Which role did you wish the GP had played? (Following aspects were elaborated about the GP: knowledge, resources, availability, continuity, barriers) How was the quality of the discharge from the hospital? How was the co-operation in this situation among the sectors? What was good / bad? How could it be improved? How was the co-operation among the GP and the home care nurses / services? How was the co-operation among the GP and the hospital / palliative specialist teams during the course of disease at home? What was good / bad? How could it be improved? How do you think palliative care should be organised? Did you get enough support from the professionals involved as a relative? What was good / bad? How could it be improved?

49

Chapter 5. Paper 1

5.8. References of Chapter 5 1. Bliss J, Cowley S, A W: Interprofessional working in palliative care in the community: a review of the literature. Journal of interprofessional care 2000, 14:281-290. 2. Department of Health: The NHS Cancer Plan. A plan for investment, A plan for reform. London, Department of Health; 2000. 3. The Danish National Board of Health, Kræftstyregruppen: [NationalDanish Cancer Action Plan] National kræftplan 2 2005. 4. Groot MM, Vernooij-Dassen MJ, Crul BJ, Grol RP: General practitioners (GPs) and palliative care: perceived tasks and barriers in daily practice. Palliat Med 2005, 19:111-118. 5. Allen J, Gay B, Crebolder H, Heyrman J, Svab I, Ram P: The European Definitions of the Key Features of the Discipline of General Practice: the role of the GP and core competencies. Brit J Gen Pract 2002, 52:526-527. 6. Cancer Pain Relief and Palliative Care. Report of a WHO Expert Committee Geneva, World Health Organization; 1990:7-67. 7. von B I, Eliasson G, Sarvimaki A, Mattsson B, Hjortdahl P: Patients' views on interpersonal continuity in primary care: a sense of security based on four core foundations. Fam Pract 2006, 23:210-219. 8. Nielsen JD, Palshof T, Mainz J, Jensen AB, Olesen F: Randomised controlled trial of a shared care programme for newly referred cancer patients: bridging the gap between general practice and hospital. Qual Saf Health Care 2003, 12:263-272. 9. Thomas C: The place of death of cancer patients: can qualitative data add to known factors? Soc Sci Med 2005, 60:2597-2607. 10. Hanratty B: Palliative care provided by GPs: the carer's viewpoint. Br J Gen Pract 2000, 50:653-654. 11. Brazil K, Howell D, Bedard M, Krueger P, Heidebrecht C: Preferences for place of care and place of death among informal caregivers of the terminally ill. Palliat Med 2005, 19:492-499. 12. de Fine Olivarius N, Hollnagel H, Krasnik A, Pedersen PA, Thorsen H: The Danish National Health Service Register. Dan Med Bull 1997, 44:449-453. 13. The National council for palliative care: Palliative care explained. 2007 [http://www.ncpc.org.uk/palliative_care.html]. 14. Doing qualitative research Edited by: Crabtree BF and Miller WL. Newbury Park, Sage Publications; 1992. 15. Donabedian A: The quality of medical care. Science 1978, 200:856-864. 16. Giorgi A: Sketch of a psychological phenomenological method. In Phenomenology and psychological research Volume 2. 1st edition. Edited by: Giorgi A. Pittsburgh, Duquesne University Press; 1985:23-85.

50

Chapter 5. Paper 1

17. Sandelowski M: Whatever happened to qualitative description? Res Nurs Health 2000, 23:334340. 18. Milne J, Oberle K: Enhancing rigor in qualitative description: a case study. J Wound Ostomy Continence Nurs 2005, 32:413-420. 19. Pope C, Ziebland S, Mays N: Qualitative research in health care. Analysing qualitative data. BMJ 320:114-116. 20. Kitzinger J: Qualitative research. Introducing focus groups. BMJ 1995, 311:299-302. 21. Lecouturier J, Jacoby A, Bradshaw C, Lovel T, Eccles M: Lay carers' satisfaction with community palliative care: results of a postal survey. South Tyneside MAAG Palliative Care Study Group. Palliat Med 1999, 13:275-283. 22. Aabom B, Kragstrup J, Vondeling H, Bakketeig LS, Stovring H: Does persistent involvement by the GP improve palliative care at home for end-stage cancer patients? Palliat Med 2006, 20:507512. 23. Grbich C, Parker D, Maddocks I: The emotions and coping strategies of caregivers of family members with a terminal cancer. J Palliat Care 2001, 17:30-36. 24. Proot IM, Abu-Saad HH, Crebolder HF, Goldsteen M, Luker KA, Widdershoven GA: Vulnerability of family caregivers in terminal palliative care at home; balancing between burden and capacity. Scand J Caring Sci 2003, 17:113-121. 25. Steinhauser KE, Christakis NA, Clipp EC, McNeilly M, McIntyre L, Tulsky JA: Factors considered important at the end of life by patients, family, physicians, and other care providers. JAMA 2000, 284:2476-2482. 26. Norman A, Sisler J, Hack T, Harlos M: Family physicians and cancer care. Palliative care patients' perspectives. Can Fam Physician 2001, 47:2009-2012. 27. Evans R, Stone D, Elwyn G: Organizing palliative care for rural populations: a systematic review of the evidence. Fam Pract 2003, 20:304-310. 28. Borgsteede SD, Graafland-Riedstra C, Deliens L, Francke AL, van Eijk JT, Willems DL: Good end-of-life care according to patients and their GPs. Br J Gen Pract 2006, 56:20-26. 29. Hickman M, Drummond N, Grimshaw J: A taxonomy of sharedcare for chronic disease. J Public Health Med 1994, 16:447-454.

51

52

Chapter 6. Paper 2

CHAPTER 6 PAPER 2 Shared care in basic level palliative home care - organizational and interpersonal challenges Neergaard MA, Olesen F, Jensen AB, Sondergaard J

Submitted to Family Practice

53

Chapter 6. Paper 2

6.1 Abstract Background: Little is known about the existing barriers to co-operation among healthprofessionals in basic level palliative care to terminally ill cancer patients in primary health-care. Objective: The aim of this study was to analyze health-professionals‟ views on interprofessional co-operation in palliative home-care for terminally ill cancer patients. Method: This study was a qualitative, descriptive study based on seven semi-structured groupinterviews conducted in Aarhus County, Denmark. 43 health-professionals (23 GPs, five hospital consultant physicians and 15 home-care nurses) working with palliative patients were interviewed. Focus of the interview was health-professionals‟ views on interprofessional co-operation in palliative home-care for terminally ill cancer patients. Results: Two main categories of problems were identified: 1) the organization of palliative homecare (need for proactive planning from the start, clear distribution of tasks, advancement of more efficient communication pathways and improved accessibility to all health-professionals) and 2) interaction between health-professionals (increased knowledge of and respect for the competencies of other health occupations and individuals). Conclusion: This study indicates that there are problems with both the organization among healthprofessionals and working culture. The main issues are distribution of tasks, information exchange, availability, respect and personal acquaintance. These issues are pivotal to improvement of delivery of palliative home-care, to training in palliative care and warrant future research. 6.2. Introduction Since palliative care is interdisciplinary (1) basic level palliative home-care often involves many different health-professionals, e.g. General Practitioners (GPs), home-care nurses and discharging hospital staff. Co-operation between these health-professionals is important (2). In a recent study we found that relatives of deceased cancer patients feel a need for strengthened interprofessional co-operation or shared care (3). Lack of co-operation caused feelings of insecurity regarding quality of care, and often forced the relatives to take on the role as coordinators. Shared care is co-operation between health-professionals: “… applies when the responsibility for the health-care of the patient is shared between individuals or teams who are parts of separate organisations, or where substantial boundaries exist” (4). It demands knowledge and accept of each others abilities, qualities and roles in the delivery of care (2). We aimed to illuminate involved health-professionals‟ experiences and views on interprofessional co-operation in basic level palliative home-care.

54

Chapter 6. Paper 2

6.3. Material and method This study was a qualitative, descriptive study based on seven semi-structured group-interviews with in all 43 health-professionals from Aarhus County (640 000 inhabitants), Denmark (5.5 million inhabitants) conducted from May to November 2005. Setting Danish Health-care system is financed through taxes with free, equal access to health-care services and more than 98 % have a registered GP (5). Danish GPs provide most of the services and act as gatekeepers to specialist treatment. Palliative specialist teams with physicians, nurses, and other health-care workers are available during daytime hours in major hospitals in Denmark, and the GPs can refer patients or ask the team for advice. Palliative home-care is divided into basic and specialist level (6). The primary care sector composes the basic level. The palliative specialist team, working either as consultants or as active professionals in the patient‟s home with or without the GPs and home-care nurses compose the specialist level.

Sample The informants were GPs, hospital consultant physicians and home-care nurses, all invited by mail. Home-care services in the 26 municipalities of Aarhus County were asked to identify homecare nurses as possible informants. The GPs and the consultant physicians were found in the database of Danish Medical Association (7). In all three groups we chose the informants purposefully trying to ensure a wide range of demographic characteristics (7) (Table 6.1.). We invited 74 GPs, 17 consultant physicians and 21 home-care nurses, 23 (31 %), five (29 %) and 15 (71 %) agreed to be interviewed, respectively. The main reason given for not participating was lack of time.

Group-interviews All interviews were conducted by MAN at The Research Unit for General Practice, University of Aarhus, and were supervised by JS or FO. All three are GPs. MAN has completed courses in qualitative methods, and JS and FO are experienced in qualitative research. The groups interviewed can be seen in Table 6.2. Interviews lasted from 108 minutes to 138 minutes and were guided by a topic guide designed on clinical experience, consensus discussions among the authors, and extensive literature studies. The topic guide was revised after each groupinterview.

55

Chapter 6. Paper 2

The interview began with interviewers‟ and participants‟ introduction, and information about anonymization and publication were given and accepted. Open-ended questions were used (8). A summary was given at the end of each group-interview to obtain an immediate validation of the issues raised. Analysis The research approach chosen was Qualitative Description since we pursued presentation of the facts from the informants‟ point of view and wanted to stay close to the data (9). The groupinterviews were transcribed verbatim by a trained secretary, and all transcripts were read repeatedly getting an overall impression before initial coding (8). All meaningful text-units were identified and coded. Examining information in text-unit they were grouped into seven groups of main codes and labelled with a name (figure 6.1). Again the information in each main code was examined and they were grouped into five categories, and subsequently two main categories. In this way we allowed the main categories to evolve from the data instead of imposing a framework a priori (10). No new categories emerged from the analysis of the last group-interview. Since the information in the group-interview with just one professional group did not differ from the other, they were all analyzed together. Agreement on the analysis was reached among the authors after thorough discussion. To enhance rigour we focused on authenticity, credibility, criticality and integrity (11) (Table 6.3.). Software package NVivo was used to assist in the coding, sorting, and retrieval of data. 6.4. Results There was overall agreement in the group-interviews on the insufficiency of cooperation among health-professionals in palliative care. Two main categories were identified: Organisation of palliative home-care and Interaction between health-professionals (Figure 6.1).

56

Chapter 6. Paper 2

Figure 6.1. From main codes to main categories.

Main codes

Categories

Main Categories

Planning Team leader

Distribution of tasks

Basic / specialist

Information exchange


Availability

Availability

Respect

Respect

Personal acquaintance

Personal acquaintance

Organisation of palliative home care

Interaction between Health-professionals

Organisation of palliative home-care Analyses revealed three important aspects of the organisation of palliative home-care: Distribution of tasks, information exchange, and availability (Figure 6.1). Specific, proposed actions from the group-interviews can be seen in Table 6.4.

Distribution of tasks The informants‟ experiences showed that palliative care needs a special focus on careful planning, the team leader role, and separation into basic and specialist levels (Fig. 1).

Planning The importance of a clear distribution of tasks from the beginning was pointed out by the informants. The agreements ought to be clear to everyone, including the patient and the relatives.

57

Chapter 6. Paper 2

”That is, that one knew who carried the baton at the moment. Right? It should be clear, crystal clear, and be stated in the record, who is responsible. We need to communicate, to determine who is responsible for the patient.” (GP) (Group-interview 2)

Especially in the discharge process, a need for a well-defined responsibility was described. Primary care informants expressed that the general hospital doctor was the health-professional who had the responsibility to promote this agreement. ”If a patient will die over the course of a few days, it would be a great help, and I sometimes experience this, if the hospital physicians would pick up the phone and say, we‟re sending her home, and she might die within a week. - I just called to let you know...” (GP) (Group-interview 3) “We do some very good discharges in close and quick contact with the home-care services, when planning the discharge… I admit that we don‟t have much contact with the GP concerning this.” (Hospital consultant physician) (Group-interview 5)

Furthermore, GPs requested that the hospital doctors informed the patient and relatives that the GP would take over after discharge. The GPs and home-care nurses experienced that whenever this information was given, the families did not feel left in limbo, and it helped the GP to reestablish the contact to the family after discharge. ”Patients would also feel better about the transfer, if they could see that now the hospital physician is handing over to the GP, who is ready to take over. Many feel insecure. They have been discharged, and then what happen? And the patient call me, maybe, before I have received the letter, and then the handing over has not been successful.” (GP) (Group-interview 7)

58

Chapter 6. Paper 2

Team leader All informants asked for a well-defined, coordinating person, a team leader. “In any case, it‟s very important that someone is aware that they are responsible for coordination. This stops everything from falling apart because everyone believes that someone else is probably taking care of things.” (GP) (Group-interview 7)

A team leader-role requires a person in control, with knowledge of the actual situation at home, the history of the patient and the disease, and an ability to facilitate the co-operation between the health-care professionals involved. Informants from all three professional groups thought that the team leader ought to be the GP. ”When we have discharged a patient, we actually expect the GP to take charge. We should approach patients and ask about their relation to their GP as he or she becomes a central figure. You need to be realistic in secondary health-care – well, we can‟t be there for the patients like the GP can.” (Hospital consultant physician) (Group-interview 3)

The GPs in the group-interviews were willing to take on the role as team leaders. ”I believe we have a much more natural and central part to play. If we have known the patient for many years, I believe it‟s absolutely natural that we assume the responsibility.”

(GP)

(Group-interview 1)

Basic or specialist level Informants described division between basic and specialist level palliative care as inappropriately blurred and characterized by arbitrariness. It was not apparent whether the discharging hospital staff considered the patient‟s symptoms and situation to be a basic or a specialist palliative care task before deciding to involve the specialist palliative team. ”…then we have simply prescribed the palliative team, exactly like we prescribe neurologists or thorax x-rays.” (Hospital consultant physician) (Group-interview 4)

59

Chapter 6. Paper 2

”And then at the hospital, and maybe frustrated that they can‟t heal the patient, they say: “We‟ve got the palliative team.” But we prefer to use the palliative team as counsellors because, if the patient has entered their sphere, the palliative team can be hard to get rid of and now who is to do what.” (GP) (Group-interview 1)

Information exchange Informants from all three professional groups described poor information exchange between professionals. GPs and home-care nurses wanted information about diagnosis, treatment, side effects, prognosis, plan, and especially information about what was said to patient and relatives, and they wanted the information latest at the day of discharge, but these demands were rarely met. ”It the GP is advised concurrently with the discharge of the patient, then that‟s a totally different situation. We sometimes see that they are advised more than a week after discharge…” (Home-care nurse) (Group-interview 6) ”I‟ll admit that hardly any discharge report states what information has been given to the patient.” (Hospital consultant physician) (Group-interview 5)

All informants welcome a webpage of shared patient-related disease information, professional contact information, and practical information about palliative care in general. ”It‟s possible to communicate with people in Australia in a matter of split seconds. But our two systems cannot see each other… Why can‟t I enter the hospital‟s system and see, what they wrote when the patient was discharged?” (GP) (Group-interview 2) ”We spend a lot of time gathering information, no matter how willing everyone is to pass on information, it takes lots of time.” (Home-care nurse) (Group-interview 6)

60

Chapter 6. Paper 2

Availability The informants pointed out the necessity of cooperating partners being available. “I expect the GPs to be available.” (Home-care nurse) (Group-interview 4) ”Yes, when a colleague from the hospital calls, I always think it‟s important. It‟s no use for us to sit on each our high horse with no will to move, because we will never have a chance to talk with each other, will we?” (GP) (Group-interview 4)

Many informants spend much time getting in touch with each other. They asked for facilitated mutual contact, direct telephone numbers and regular telephone hours. “Unfortunately, I mostly find that doctors aren‟t there and that we need to spend an hour on the phone waiting just to get through…I think the odds are against us, at times” (Home-care nurse) (Group-interview 6) ”Everyone says that we‟re hard to get a hold of. And that‟s probably right.” (GP) (Group-interview 7)

GPs often felt alone in the treatment of terminally ill patients. They called for easier access to specialist advice and supervision. ”I feel better about being in charge if I‟m confident about the situation. On the other hand, if I‟m not confident about the situation…, then I feel it‟s good that the palliative team can take over.” (GP) (Group-interview 7) ”I believe the palliative team can do things that I can‟t... I‟ve had contact with them several times concerning pain therapy. It‟s made me feel more secure.” (GP) (Group-interview 2) ”It also makes patients feel more secure, that you have contacted the specialised department. And then they know, that my recommendations and actions have been confirmed by others.” (GP) (Group-interview 3)

61

Chapter 6. Paper 2

Interaction between health-professionals Analyses revealed two important aspects on interaction between health-professionals: Respect and personal acquaintance. Specific, proposed actions from the group-interviews can be seen in Table 6.4.

Respect Everyone agreed that trust and mutual respect, both personally and professionally, were a prerequisite for successful cooperation. ”Yes, but it is all about respecting each other.” (GP) (Group-interview 3) ”I very much agree that mutual trust makes you feel secure.” (Home-care nurse) (Group-interview 3)

Informants were aware that prejudice and lack of respect was a barrier to co-operation and could create distress among patients and relatives. ”At the hospitals we try hard not to hold others responsible for what has happened. We may internally criticize each other, but we never criticize a GP when communicating with patients. If a colleague does so, we address him or her directly, because it is unacceptable.” (Hospital consultant physician) (Group-interview 5) Several GPs and home-care nurses expressed much respect for each other‟s work. “We‟ve got a bunch of brilliant home nurses. And I agree with them on most issues. I‟ve really not got anything negative to say about any of the contacts I‟ve had. Because things have actually worked out well.” (GP) (Group-interview 2)

However, others expressed disrespect against other health-professionals, especially when they met other professionals who were not as enthusiastic about palliative care.

62

Chapter 6. Paper 2

”When things have really gone wrong, it‟s probably when a home-care nurse would rather sit on her well-padded behind. And control everything from there.” (GP) (Group-interview 2) “Do GPs not have the guts? Because this gets too close for comfort? Or is it because they say: ”I‟m so busy taking care of my practice, someone else will have to take care of this. I don‟t need to get involved.” (Home-care nurse) (Group-interview 3)

Several GPs revealed disrespect against hospital doctors, especially their unwillingness to establish teamwork with GPs. ”No, that‟s a quite normal attitude for a surgeon: “Our business is surgery, the sooner the patient leaves afterwards, the better. It‟s not our problem regardless if the patient is still ill.” (GP) (Group-interview 4) “The hospital physicians believe that they have a right to turn away patients. That‟s what it‟s all about. The best way to turn away a patient…” (GP) (Group-interview 7)

Home-care nurses and hospital doctors, on the other hand, sometimes revealed disrespect against GPs, primarily regarding the GPs competences and personal merits. Hospital doctors sometimes referred directly to the specialist palliative team because of uncertainty about GPs‟ competences and interest in palliative care. “Some GPs will and some won‟t, and maybe some can‟t because they don‟t have the adequate knowledge or training, or the interest needed…So if I hand over to a practitioner who is not really interested in this area, I can be very worried about the patient… How can I know that the GP is interested in this area?” (Hospital consultant physician) (Group-interview 5) ”The worst GPs are not up to date and have the attitude of:”I know better and will do better on my own”. They are the worst.” (Home-care nurse) (Group-interview 4)

63

Chapter 6. Paper 2

Personal acquaintance Informants in all professional groups concurrently called for personal acquaintance with each other as a method to facilitate cooperation. ”Yes, personal contact, I believe, is still preferable to electronic contact. And you may get totally different information.” (Hospital consultant physician) (Group-interview 3)

They had all experienced personal acquaintance breaking down prejudices and disrespect and causing better co-operation. ”You get a positive response when calling. It is exactly the same situation as when the GPs call. This demonstrates a warm person who gets involved with the patient‟s course of disease.” (Hospital consultant physician) (Group-interview 3) ”It‟s essential that you know who you are communicating with when you need help. Because it makes it so much easier to act. You don‟t need to start from square one to get understood.” (GP) (Group-interview 3) ”Being able to put a face on the people you work with is immensely important. You get a much better understanding of who they are…” (Home-care nurse) (Group-interview 6)

GPs and home-care nurses said that personal acquaintance promotes team-spirit and increase mutual responsibility. ”How would I ever be able to do my job, if I didn‟t have a close relation to the home-care nurses? - If we didn‟t know each other really well.” (GP) (Group-interview 1) “Well, when you know the GPs of the area and they know us… It‟s a considerable advantage in many of these patient pathways.” (Home-care nurse) (Group-interview 3)

64

Chapter 6. Paper 2

6.5. Discussion We found indication of need for improved cooperation in palliative home-care through proactive planning from the start of the palliative course, clear distribution of tasks, better and easier communication pathways, and improved accessibility to all health-professionals. Maybe the most important finding was need for better working culture and willingness to be equal team-partners with mutual respect and better knowledge of each other‟s professional competence and personal merits.

Another important issue is that division between basic and specialist level of palliative care seemed characterized by arbitrariness. This could be due to the fact that palliative care and palliative education has been given a low priority in Denmark (12). Interviews are pertinent to explore the informants‟ perspectives and experiences concerning a specific topic. The force of group-interview is that the interaction in the group can facilitate new issues in the interview and thus make new themes or aspects emerge (13).

Being aware of the fact that the interviewers (MAN, JS, and FO) were all GPs, we tried to minimise the influence this fact could give to the interviews and analysis (8). We had a sparse amount of participating hospital doctors in the study, but the five doctors worked in five different fields and still had similar opinions about the important issues.

In line with findings from other studies, we found that GPs find coordination of palliative care an essential task (14;15), but they request easier access to specialist backup (16).

Also in line with other studies, we found problems with organisation of palliative care, feelings of “struggling with the system” (17) and an agreement on the importance of naming a team leader in palliative home-care (18;19).

Similarly to other studies, we found insufficient information exchange between hospitals and community-based services and inadequate availability among health-professionals (16;17).

A review concluded that professionals must acknowledge the strengths of other healthprofessionals (2) and corresponds with the importance of respect found in the present study. It is thought-provoking that even bereaved relatives see this as a barrier to successful palliative care (3). We and others found that whenever the health-professionals met each another in person, the

65

Chapter 6. Paper 2

chance of mutual respect was increased (20). In our study, health-professionals experienced that personal acquaintance among health-professionals also engenders team-spirit leading to responsibility. Our study revealed prejudices and disrespect among the health-care professionals that need to be further exposed in future research.

Improving shared care in palliative home-care, creating a system based on cooperation and mutual respect is a major challenge and requires further research. Finally, our research showed need to improve communication; indirectly showing the need for developing better IT and common webbased information. The issues about shared care in palliative care, pointed out in Table 4, are important issues in training in palliative care. A questionnaire-survey of relatives, home-care nurses, and GPs is currently being conducted. 6.6. Conclusion The study indicates problems with both organization among health-professionals and working culture. The main issues: Distribution of tasks, Information exchange, Availability, Respect and Personal acquaintance, are pivotal to improve delivery of palliative home-care, to training in palliative care and warrant future research. Funding This work was supported by the Aarhus County Research Fund for the Clinical Development and Research in General Practice and across the Primary and Secondary Health Care Sectors [4-01-3-04] and the Danish National Research Foundation for Primary Care [585-457808]. Acknowledgements Profound gratitude is extended to participating informants. Ethics According to Scientific Committee for the County of Aarhus, the Biomedical Research Ethics Committee System Act does not apply to this study. Conflicts of interest Authors have no conflicts of interest

66

Chapter 6. Paper 2

6.7. Tables Table 6.1. Demographics of participants in the interviews. General practitioners In all (n (%)) Sex (n (%)) Women Men Age (Median (IQR)) Location (n (%)) Major city Minor city Countryside Home-care nurses In all (n (%)) Sex (n (%)) Women Men Age (Median (IQR)) Location (n (%)) Major city Minor city Countryside Hospital doctors In all (n (%)) Sex (n (%)) Women Men Age (Median (IQR)) Speciality (n (%)) Oncology Internal Medicine Gastro surgery Endocrinology Gynaecology Location (n (%)) University hospital Regional hospital

23 (100 %) 12 (52 %) 11 (48 %) 53 years (47:57) 9 (39 %) 9 (39 %) 5 (22 %) 15 (100 %) 15 (100 %) 0 49 years (43; 51) 5 (33 %) 3 (20 %) 7 (47 %) 5 (100 %) 1 (20 %) 4 (80 %) 50 years (50;53) 1 (20 %) 1 (20 %) 1 (20 %) 1 (20 %) 1 (20 %) 3 (60 %) 2 (40 %)

Table 6.2. The distribution of informants in the group-interviews Group-interview number 1 2 3 4 5 6 7

Number of informants 8 4 7 6 6 7 5

In all

43

8 4 3 1 2 5

Number of homecare nurses 2 3 3 7 -

Number of Hospital physicians 2 2 1 -

23

15

5

Number of GPs

Table 6.3. Strategies to enhance rigor in Qualitative description (11) Strategies

Explanation

Authenticity

The attention to the voices of participants or the ability to remain true to the phenomena under study

Credibility

A reflexion of how believable results are

Criticality

The critical appraisal of every decision made throughout the research process

Integrity

On-going reflection and self-criticality of the researcher

67

Chapter 6. Paper 2

Table 6.4. Results. Issues and actions to better shared care for palliative patients at home Organisation Issues that need improvement Planning the palliative course of disease from the beginning

Clear distribution of tasks and team leadership


Availability

Proposed actions The hospital doctor contacts the GP when a palliative patient is diagnosed to inform and to discuss the level of palliative care needed The hospital doctor informs the patient and relatives of diagnosis and treatments and discusses the future plan and wishes. The hospital doctor and the GP make an agreement on responsibility, team leadership, and maybe contact with a palliative specialist team The hospital nurses make an agreement with the homecare nurse concerning a contact nurse for the patient The patient and relatives are informed about the plan, i.e. the agreements with the GP and the home-care nurse. A detailed discharge letter with diagnosis, prognosis, treatments, information to patient and relatives, agreements with GP and home-care nurse, and preferably names and contact information (phone number, fax number, E-mail, phone hours) to all involved professionals Possibilities for all professionals involved to gain access to relevant hospital and GP files Easier access to specialist advice for GPs Improved availability among professionals Interactions between professionals

Issues that need improvement

Proposed actions

Respect and shared care culture

Improving the knowledge of other professionals‟ competences in general Focus on and dealing with prejudices at pre- and postgraduate education

Acquaintance

Phone-calls and meetings to support written information Conferences with all professionals involved in specific cases

68

Chapter 6. Paper 2

6.8. References of Chapter 6 (1) Cancer Pain Relief and Palliative Care. Report of a WHO Expert Committee. Geneva: World Health Organization; 1990. (2) Bliss J, Cowley S, While A. Interprofessional working in palliative care in the community: a review of the literature. Journal of interprofessional care 2000;14(3):281-90. (3) Neergaard MA, Olesen F, Jensen AB, Sondergaard J. Palliative care for cancer patients in a primary health care setting: Bereaved relatives' experience, a qualitative group interview study. BMC Palliat Care 2008 Jan 15;7(1):1. (4) Pritchard P, Hughes J. Shared Care The Future Imperative? 1.ed. ed. London: Royal Society of Medicine Press; 1995. (5) de Fine Olivarius N, Hollnagel H, Krasnik A, Pedersen PA, Thorsen H. The Danish National Health Service Register. Dan Med Bull 1997;44(4):449-53. (6) The National council for palliative care, http:, www.ncpc.org.uk/palliative_care.html. Palliative care explained. 2007. (7) Crabtree BF, Miller WL. Doing qualitative research. Newbury Park: Sage Publications; 1992. (8) Kvale S. Interviews: An Introduction to Qualitative Research Interviewing. 1 ed. Thousand Oaks: SAGE Publications; 1996. (9) Sandelowski M. Whatever happened to qualitative description? Res Nurs Health 2000 Aug;23(4):334-40. (10) Pope C, Ziebland S, Mays N. Qualitative research in health care. Analysing qualitative data. BMJ320(7227):114-6. (11) Milne J, Oberle K. Enhancing rigor in qualitative description: a case study. J Wound Ostomy Continence Nurs 2005 Nov;32(6):413-20. (12) Centeno C, Clark D, Lynch T, Racafort J, Praill D, De Lima L, et al. Facts and indicators on palliative care development in 52 countries of the WHO European region: results of an EAPC Task Force. Palliat Med 2007 Sep;21(6):463-71. (13) Kitzinger J. Qualitative research. Introducing focus groups. BMJ 1995;311(7000):299-302. (14) Borgsteede SD, Graafland-Riedstra C, Deliens L, Francke AL, van Eijk JT, Willems DL. Good end-of-life care according to patients and their GPs. Br J Gen Pract 2006 Jan;56(522):20-6. (15) Groot MM, Vernooij-Dassen MJ, Crul BJ, Grol RP. General practitioners (GPs) and palliative care: perceived tasks and barriers in daily practice. Palliat Med 2005 Mar;19(2):111-8. (16) Shipman C, Addington-Hall J, Barclay S, Briggs J, Cox I, Daniels L, et al. Providing palliative care in primary care: how satisfied are GPs and district nurses with current out-ofhours arrangements? Br J Gen Pract 2000 Jul;50(455):477-8.

69

Chapter 6. Paper 2

(17) McIlfatrick S. Assessing palliative care needs: views of patients, informal carers and healthcare professionals. J Adv Nurs 2007 Jan;57(1):77-86. (18) Steinhauser KE, Christakis NA, Clipp EC, McNeilly M, McIntyre L, Tulsky JA. Factors considered important at the end of life by patients, family, physicians, and other care providers. JAMA 2000 Nov 15;284(19):2476-82. (19) Mitchell GK. How well do general practitioners deliver palliative care? A systematic review. Palliat Med 2002 Nov;16(6):457-64. (20) Mitchell G, Cherry M, Kennedy R, Weeden K, Burridge L, Clavarino A, et al. General practitioner, specialist providers case conferences in palliative care--lessons learned from 56 case conferences. Aust Fam Physician 2005 May;34(5):389-92.

70

Chapter 7. Paper 3

CHAPTER 7 PAPER 3 Associations between home-death and general practitioner involvement in palliative care Neergaard MA, Vedsted P, Olesen F, Sokolowski I, Jensen AB, Sondergaard J

Submitted to British Journal of General Practice

71

Chapter 7. Paper 3

7.1. Abstract Background: Most cancer patients die at institutions despite their wish of home-death. General practitioner (GP)-related factors may be crucial in attaining home-death. Aim: Describing cancer patients in palliative home care and examining associations between home-death and GP involvement in the palliative pathway. Design of study: Population-based, cross-sectional combined register and questionnaire study Setting: Aarhus County, Denmark. Methods: Patient-specific questionnaires were sent to GPs of 599 cancer patients who died during a 9-months period in 2006. Information on the 333 included cases comprised socio-demography and GP-related issues, e.g. knowledge of the patient, unplanned home-visits, GPs providing their private phone number and contact with relatives. Register data were collected on patients‟ age, gender, cancer diagnosis, place of death and number of GP home-visits. Associations with homedeath were analysed in a multivariate regression model with prevalence ratios (PR) as a measure of association. Results: There was a strong association between home-death and GPs making home-visits (PR: 4.4 (95% CI: 1.3;15.2) and involvement of community nurses (PR: 1.4 (95% CI: 1.0;1.9). No other GP-related variables were statistically significantly associated with home-death. Conclusion: Active involvement of GPs providing home-visits and the use of home nurses were independently associated with a higher likelihood of home-death. The primary care team may facilitate home-death, accommodating patients‟ wishes. Future research should examine the precise mechanisms of their involvement. 7.2. Introduction Most cancer patients die at institutions even if most of them wish to die at home (1-4) but we have only little knowledge of how best to facilitate the process of dying at home (5). General practitioners (GPs) and community nurses provide frontline health care in patients‟ homes and are often involved in palliative care, but there is only sparse knowledge of the actual extent and nature of their involvement in facilitating home-death.

In an interview study, bereaved relatives requested more GP involvement (6), but even when GPs were involved, the palliative pathways were often sub-optimal. Most extant research focuses on specialised institutional based palliative care, and studies on how to support and improve GPinvolvement in palliative care remain sparse.

72

Chapter 7. Paper 3

The aim of the present study was to describe cancer patients in palliative home care in relation to demographic characteristics, palliative pathway, degree of GP involvement, and to examine the association between home-death and GP involvement in the palliative pathway in an attempt to identify factors of significance for supporting patients to die at home. 7.3. Materials and methods We conducted a survey in 2007 (January – March) of 231 general practices with 599 registered deceased cancer patients in Aarhus County (at that time approximately 640 000 inhabitants, 12% of the Danish population, 43 municipalities and 1680 persons dying from cancer each year (2005)) (7). In Denmark, all citizens are registered with unique civil registration numbers (8) and by means of these numbers questionnaire data were linked to health register information. Setting Danish health care system is tax-financed and more than 98% of Danes are registered with GPs and receive free medical care (9). Danish GPs act as gatekeepers for access to specialist treatment and are responsible for frontline care 24 hours a day. Large GP cooperations provide out-of-hours care (4 pm. to 8 am and Sat/Sun), covering an entire county.

Denmark has no formal national agreement on task distribution in palliative care, but the Aarhus County provided a special fee for GPs‟ palliative care involvement. Community nurses employed by the municipalities are often involved in palliation and visit patients on a 24-hour basis. Palliative specialist outgoing teams based at the major hospitals are available during daytime hours, and specialist advice can be obtained from these teams by GPs or community nurses by telephone. Study population and sampling We included adults in Aarhus County who died from cancer from 1st March to 30th November 2006 and who had received some palliative home care either from GPs, community nurses or a palliative specialist team. Since no database on palliative patients was available and since the Danish Register of Causes of Death was still not updated on deaths in 2006, we sampled the patients by combining official register data with questionnaires information. The questionnaires were answered by the patients‟ GPs, who were asked first about cause of death and secondly if palliative care had been provided in the patient‟s home.

From the county hospital discharge register we identified 29,043 individuals above 18 years who were registered with at least one cancer diagnosis (ICD-10) (excluding non-melanoma skin

73

Chapter 7. Paper 3

cancers) during the period November 2006 and ten years back. In December 2006, using the Civil Registration System database, we identified 813 patients among the 29,042 who died from 1st March to 30th November 2006 (nine months). From the regional health authority‟s register we identified their GPs. Eight (1.0%) patients were not registered with a GP and 18 (2.2%) had moved from the county after having been diagnosed, leaving 787 deceased cancer patients. A questionnaire was sent to their GPs.

In late 2008 data from the Danish Register of Causes of Death for 2006 was available. Merging the information of cancer deaths with our database, we excluded 188 patients who were not registered with cancer as the death reason which reduced our study population to 599 deceased cancer patients (Figure 7.1). Figure 7.1. Flow-chart of sampling of study population and GP questionnaires. Responders and non-responders.

Flow-chart 29043 Patients with a cancer diagnosis in Aarhus County within 10 years before17th November 2006 28230 Alive or dead before 1st March 2006 or < 18 years old 813 Cancer patients in Aarhus County died between 1st March and 30th November 2006 8 Are not registered with GP and 18 moved out of the area 787 Cancer patients in Aarhus County died between 1 March and 30 November 2006 registered with a GP

350 GP questionnaires were filled in

437 Questionnaires were not filled in 2008: Data from The Danish Register of Causes of Death

17 Not registered as cancer deaths


599 Patients registered as cancer deaths in Aarhus County died between 1 March and 30 November 2006 registered with a GP 333 GP questionnaires were filled in

266 Questionnaires were not filled in

194 Non-responders 9 GP retired, were ill or had died 73 GP did not want to participate 87 GP did not respond 25 GP stated :“Not cancer death”

72 Cases excluded No palliative home care

Data collection

74

Chapter 7. Paper 3

GP questionnaires included themes identified through literature studies, clinical experience and group-interview studies with bereaved relatives (6) and involved professionals (GPs, community nurses and hospital consultants). Important questionnaires themes were: GPs‟ knowledge of the patient prior to the palliative period, whether GPs provided a plan on whom to contact during outof-office hours, contact to relatives and the duration of the palliative period. The 72-item questionnaire was pilot-tested among 30 GPs in another Danish county.

In partnership practices, the GPs most familiar with the patients were asked to answer the questionnaire. GPs received a small economic compensation for their efforts. Non-responders were sent reminders four and seven weeks following the first questionnaire. We retrieved register data on patients‟ age (18-65, 66+), gender, cancer diagnosis (lung, colorectal, breast, prostate, other), place of death, home-visits provided by GPs (0, 1, 2, 3, >3, and dichotomised into no/yes) in the period three months prior to the patients‟ death. The questionnaires included information on the patients‟ marital status (single, married or cohabiting), patients‟ children (no, yes at home, yes not at home), GPs‟ knowledge of the patient before the palliative period (dichotomized into poor (1,2 on a 1-5-point scale) and good (3, 4 and 5)), whether GPs provided their private phone number for emergency advise (no/yes), whether GPs had contact to patients‟ relatives (no/yes), whether community nurses (no/yes) or a specialist team (no, yes) were involved, and the duration of the palliative period at home and in total. The palliative period was defined as the last period of the patient‟s life during which all curative treatment had ceased and care and treatment were solely palliative. Analysis „Home-death‟ was defined as the outcome measure and associations with patient, GP and palliative pathway variables were calculated. Place of death is defined in The Danish Register of Causes of Death as „home‟, „nursing home‟, „hospital or hospice‟ and „death in other places‟. Since the patient‟s own GP provides care for the patient at home as well as in nursing homes, we defined „home-death‟ as death either at home or at a nursing home. GPs are rarely involved in palliative pathways at hospices, but the register data did not allow us to separate between hospital and hospice deaths.

Unadjusted and adjusted associations were calculated. Using robust variance estimates, the estimates were adjusted for clustering of patients in practices (10). Prevalence ratios (PRs) with 95% confidence intervals (95% CI) were used as a measure of association. Due to the high

75

Chapter 7. Paper 3

prevalence of the outcome measure (more than 20% home-deaths), odds ratios would overestimate the association (11;12). PRs were calculated with generalised linear models (GLM) with log link and the Bernoulli family, and when the model could not converge, we used the Poisson regression model (11;13).

The variables were assessed for collinearity (Pearson's correlation coefficient > 0.4) and multicollinearity (variance inflation factor < 10) (14;15). None of the included variables had to be excluded. We added the duration of the palliative period spent at home as a confounder, since it could be associated with GPs‟ possibility to provide palliative services. Data were analyzed using STATA 10 (16). 7.4. Results Among the 599 questionnaires sent to GPs, a total of 333 questionnaires from 231 general practices were filled in and 72 were excluded because the GP stated that no home care was provided during the palliative pathway (response rate 63.2%) (Figure 7.1). The general practices completed questionnaires for 1-6 patients (mean: 2). The non-responding practices were not statistically significantly different from participating practices with respect to practice organisation (single or partnership), number of patients per GP and number of questionnaires sent per general practice (data not shown).

The 194 cases from non-responding GPs were not statistically significantly different from included cases regarding patients‟ gender, place of death and number of GP home-visits (Table 7.1). However, the patients in the non-participating group were statistically significantly older than the patients in the included cases.

Association with home-death GP home-visits and involvement of homecare nurses was positively associated with home-death (Table 7.2). We also replaced home-visits (no, yes) with a categorical variable (0, 1, 2, 3 and 4+), referring to the actual number of visits, in the same model. When GPs made three or more homevisits, the patient‟s likelihood of home-death increased significantly (PR for 3 home-visits: 6.7 (95% CI: 2.0;22.9), PR for 4+ home-visits: 5.8 (95% CI: 1.7;19.6) with 0 home-visits as reference group). 7.5. Discussion Main findings

76

Chapter 7. Paper 3

In a consecutive group of cancer patients who died from cancer and had a palliative pathway at home (incl. nursing home), GPs were involved in 89% of the cases, knew the patient well before the palliative period (87%) and had contact with the relatives of the patient (87%). In nearly half of these cases, GPs paid at least four home-visits during the last three months of the patients‟ lives, and nearly one third of all home-visits were paid out-of-hours. In only 13% of the 599 cases, the GP did not pay home-visits at all. Likewise, in nearly half of the cases, GPs provided special services like unplanned home-visits and offering their private phone numbers.

In two thirds of the included cases, GPs were involved together with community nurses and in one third of the included cases a specialist team was involved. In 80 cases (24.0%), GPs were involved without the community nurses and the specialist team.

More than half of the patients died at home (one fifth at a nursing home). However, those who eventually died at institutions spent more than 75% of their time at home during the palliative pathway. This is significantly lower than the percentage for the home-death group (84%), but the differences between the actual duration of the palliative period at home and the total palliative period fell short of statistical significance in all place-of-death-groups.

Home-visits provided by GPs and involving homecare nurses were associated with home-death. Further, an increased likelihood of home-death was seen when GPs paid three or more homevisits during the last three months. Strengths and limitations of the study To eliminate differential misclassification, we used the standardised official health registers to identify the study population including the places and the reasons of death. However, in 25 of the 599 cancer death cases, GPs returned the questionnaires stating that the patient did not die from cancer. These cases were defined as non-response, which may have introduced a bias (incl. confounding by indication), but the direction and importance of this bias is hard to establish. However, we found no statistically significant differences between included and non-included cases regarding gender, place of death and number of GP home-visits, indicating that selection bias must be limited.

To minimise recall bias, the questionnaire was sent in January 2007 instead of waiting for the Danish Register of Causes of Death to update in 2008. Further, all Danish GPs can consult electronic patient files when completing a questionnaire (17).

77

Chapter 7. Paper 3

Our results are generalizable to the Danish national context of palliative care because we sampled patients who died from cancer and had some palliative period at home regardless of the involvement of specialised teams or hospital records and because Denmark is very homogenous with respect to primary care and social demography.

Approximately 1680 patients died from cancer in Aarhus County in 2006, and we included 599 cases recruited during a nine-month period (Figure 7.1). This lower amount of cases can be explained by the fact that we did not include persons less than 18 years, non-melanoma skin cancer, and the fact that we included only those with a cancer diagnose registered in a hospital in Aarhus County as the main diagnose of admittance in a 10-year period. Furthermore, the period where patients could die (1st March-30th November) was without the winter months of 2006, which may account for some of the missing cases, since winter months may have a higher average of deaths than the rest of the year. Comparison with existing literature Where cancer patients die is determined by a complex interplay of personal and cultural values, the presence of supportive networks, and a host of physical and medical factors as well as the organisation of the health care system. Contrary to our study on primary care, most previous papers on factors associated with place of death focus on socio-demographic factors and specialists‟ efforts (18).

The found association between home-death and GP home-visits fits well with findings of previous studies (19-22). Contrary to these studies, we adjusted the model by including time at home as a variable because dying at home often demands more home-visits and more home care, and the patient has to be at home to receive these visits. The association between the probability of dying at home and the home-visits made by GPs persisted in spite this adjustment. GPs‟ participation in palliative care may be seen as a package. Thus, if GPs are paying many home-visits during the palliative period, this would testify to their willingness to assume different tasks and provide different services in relation to palliative pathways. However, we lack insight into the precise contents of this informal, palliative health care package. More detailed studies should therefore be conducted to elucidate the specific effect of GP home-visits and other GP-related factors.

78

Chapter 7. Paper 3

In line with previous studies, a significant association was found between home-death and the involvement of community nurses (22-24). Addressing this association was not a separate aim, and the found association deserves further investigation.

The fact that patients who received palliative care at home and died at an institution still spent most of the palliative period at home definitely questions the use of home-death as measure of good quality in terminal care. A „successful palliative pathway‟ can be more than achieving home-death. The reasons for hospitalisation during the last days of life may be many, and hospital-death may represent successful care just as well as home-death. Studies are therefore warranted to establish what constitutes a successful palliative pathway for patients and relatives and how primary care professionals may be involved in that pathway. Implications for future research This study calls for further assessment of the predictive power of a more active approach to homevisits. Follow-up studies that include a needs assessment among patients with few and many visits are needed.

7.6. Conclusion In a consecutive group of cancer patients who died from cancer and had a palliative pathway at home (incl. nursing home), GPs paid at least four home-visits during the last three months of the patients‟ lives in more than 40% of the cases, and one third of all GP home-visits were paid out-ofhours. In only 13% of the 599 cases, the GP did not pay home-visits at all. The patients who eventually died at institutions spent more than 75% of their time at home during the palliative pathway, which emphasizes the need to focus on primary healthcare in palliative care. Furthermore, our study indicates that active involvement of GPs providing home-visits and the involvement of home nurses were independently associated with a higher likelihood of homedeath. The primary care team may be instrumental in allowing patients to die at home. Future research should examine the more precise mechanisms of their involvement. Funding body The study was funded by the Aarhus County Research Fund for the Clinical Development and Research in General Practice and across the Primary and Secondary Health Care Sectors (4-01-3-04), the Danish National Research Foundation for Primary Care (585-457808) and The Multipractice Study Committee (58504/2072). Ethical approval According to Scientific Committee for the County of Aarhus, the Biomedical Research Ethics Committee System Act does not apply here (j.no. 2005-2.0/14). The study was approved by the Danish Data Protection Agency (j.no. 2005-41-4967) and the Danish National Board of Health (j.no. 7-505-29-1007/1).

79

Chapter 7. Paper 3

Competing interests The authors have no conflicts of interest. Acknowledgements Profound gratitude is extended to participating GPs.

7.7. Tables Table 7.1. Characteristics of 333 included cases and 194 cases not included because the GP did not respond. Case data derived from GP questionnaires and registers. Data on cases with nonresponse from GPs is from registers. Cases in study (N=333) Patient’s age at time of death (mean, years (95% CI)) Patient’s gender (n (%)) Male Female Primary cancer diagnosis (n (%)) Bronchus/lung Colon/rectum Breast Prostate Other Patient’s marital status (n (%)) Single Married Having children (n (%)) No Yes, living at home Yes, adults Place of death (n (%)) Home Nursing home Hospital/hospice Other (e.g. other institution) GP involvement in palliation (n (%)) No Yes GP’s knowledge prior to palliative period (n (%)) Poor Well Number of home visits by GP during the last 3 months (median (IQI)) GP home visits in the last 3 months (n (%)) 0 home visits 1 home visit 2 home visits 3 home visits >3 home visits

Cases of nonresponders (N=194)

69.4 (68.0;70.8)

73.5 (71.7;75.3)*

181 (54.4) 152 (45.6)

113 (58.3) 81 (41.7)

65 (19.5) 50 (15.1) 34 (10.2) 39 (11.7) 145 (43.5)

37 (19.1) 32 (16.5) 22 (11.3) 31 (16.0) 72 (37.1)

125 (39.8) 189 (60.2)

-

32 (11.0) 34 (11.7) 226 (77.3)

-

120 (36.0) 69 (20.7) 140 (42.0) 4 (1.2)

70 (36.1) 51 (26.3) 70 (36.1) 3 (1.5)

37 (11.3) 290 (88.7)

-

44 (13.3) 279 (86.7)

-

3 (1;6)

3 (1;5)

44 (13.2) 48 (14.4) 40 (12.0) 49 (14.7) 152 (45.7)

26 (13.4) 30 (15.5) 30 (15.5) 24 (12.4) 84 (43.3)

80

Chapter 7. Paper 3

Time of GP home visits during the last 3 months (n (%)) Normal hours Out of hours Unplanned home visits by GP (n (%)) No Yes

945 (69.1) 422 (30.9)

751 (65.2) 400 (34.8)

134 (49.1) 138 (50.9)

-

GP gave private number to patient to use in out-of-office hours (n (%)) No 164 (58.4) Yes 116 (41.6) GP had made a plan with the patient for whom to contact in out-of-office hours No 180 (63.8) Yes 102 (36.2) GP had contact with relatives (n (%)) No 39 (13.5) Yes 243 (86.5) Community nurse involvement (n (%)) No 105 (31.1) Yes 228 (68.9) Specialist team involvement (n (%)) No 206 (61.9) Yes 127 (38.1) Common involvement of professionals (n (%)) GP and Community nurse 220 (66.1) GP and Specialist team 114 (34.2) GP, Community nurse and specialist team 96 (28.8) Duration of palliative period related to place of death (mean, weeks (95% CI)) Institution (hospital/hospice) 17.5 (14.6;20.3) Home (home, nursing home) 17.9 (15.4;20.3) Total 17.8 (15.9;19.6) Time spend at home related to place of death (mean, weeks (95% CI)) Institution (hospital/hospice) 14.6 (11.8;17.3) Home (home, nursing home) 14.5 (12.5;16.6) Total 14.6 (13.0;16.3) Percentage of time spend at home related to place of death (mean, % (95% CI)) Institution (hospital/hospice) 75.1 (71.0;79.1) Home (home, nursing home) 84.0 (80.5;87.6) Total 80.1 (77.4;82.8) * Statistically significantly different from cases in study. Not all sums of percentages are added to 100.0% because of round-offs.

-

-

-

-

-

-

-

-

-

81

Chapter 7. Paper 3

Table 7.2. Associations between home death and model variables. A total of 333 cases were included in the analyses. The unadjusted and the adjusted prevalence ratios (PRs) are shown with 95 % confidence intervals (95% CIs) Unadjusted prevalence ratio (95% CI)

Adjusted prevalence ratio (95% CI)

Age of patient 18 – 65 66+

1 a 1.3 (1.0;1.7)

1 1.1 (0.8;1.5)

Gender of patient Male Female

1 0.8 (0.7;1.0)

1 0.8 (0.6;1.1)

Primary cancer diagnose Bronchus/lung Colon/Rectum Breast Prostate Other

1 1.1 (0.8;1.5) 1.1 (0.8;1.6) 1.3 (0.9;1.7) 1.0 (0.8;1.3)

1 1.3 (0.9;1.8) 1.3 (0.9;2.1) a 1.3 (1.0;1.9) 1.2 (0.9;1.6)

Patient’s marital status Single Married

1 1.0 (0.9;1.3)

1 1.0 (0.8;1.3)

Children No Yes, living at home Yes, adults

1 0.9 (0.6;1.3) 1.0 (0.7;1.3)

1 0.9 (0.6;1.5) 1.0 (0.7;1.4)

GP knowledge prior to palliative period Poor Well

1 1.2 (0.9;1.7)

1 0.9 (0.6;1.3)

Home visits by GP No Yes

1 a 3.5 (1.9;6.4)

1 a 4.4 (1.3;15.2)

Unplanned home visits by GP No Yes

1 1.2 (0.9;1.4)

1 1.0 (0.8;1.2)

1 a 1.2 (1.0;1.5)

1 1.1 (0.8;1.3)

GP had made a plan with the patient for whom to contact in out-of-office hours No Yes

1 0.8 (0.7;1.0)

1 1.0 (0.8;1.3)

GP had contact with relatives No Yes

1 1.3 (0.9 ;1.9)

1 1.2 (0.8;1.8)

Community nurse involvement No Yes

1 a 1.8 (1.4;2.2)

1 a 1.4 (1.0;1.9)

GP gave private number to patient to use in out-of-office hours No Yes

82

Chapter 7. Paper 3

Specialist team involvement No Yes

1 a 1.2 (1.0;1.5)

1 1.1 (0.9;1.4)

a

: Statistically significant.

7.8. References of Chapter 7 (1) Hanratty B. Palliative care provided by GPs: the carer's viewpoint. Br J Gen Pract 2000 Aug;50(457):653-4. (2) Thomas C. The place of death of cancer patients: can qualitative data add to known factors? Soc Sci Med 2005 Jun;60(11):2597-607. (3) Thomas C, Morris SM, Clark D. Place of death: preferences among cancer patients and their carers. Soc Sci Med 2004 Jun;58(12):2431-44. (4) Grande GE, Farquhar MC, Barclay SI, Todd CJ. Valued aspects of primary palliative care: content analysis of bereaved carers' descriptions. Br J Gen Pract 2004 Oct;54(507):772-8. (5) Gomes B, Higginson IJ. Factors influencing death at home in terminally ill patients with cancer: systematic review. BMJ 2006 Mar 4;332(7540):515-21. (6) Neergaard MA, Olesen F, Jensen AB, Sondergaard J. Palliative care for cancer patients in a primary health care setting: Bereaved relatives' experience, a qualitative group interview study. BMC Palliat Care 2008 Jan 15;7(1):1. (7) The Danish National Board of Health. The Danish Register of Causes of Death [In Danish]. 2008. (8) Pedersen CB, Gotzsche H, Moller JO, Mortensen PB. The Danish Civil Registration System. A cohort of eight million persons. Dan Med Bull 2006 Nov;53(4):441-9. (9) de Fine Olivarius N, Hollnagel H, Krasnik A, Pedersen PA, Thorsen H. The Danish National Health Service Register. Dan Med Bull 1997;44(4):449-53. (10) Donner A, Klar N. Design and Analysis of Cluster Randomisation Trials in Health Research. 1 ed. London: Hodder Arnold; 2000. (11) Barros AJ, Hirakata VN. Alternatives for logistic regression in cross-sectional studies: an empirical comparison of models that directly estimate the prevalence ratio. BMC Med Res Methodol 2003 Oct 20;3(1):21. (12) Thompson ML, Myers JE, Kriebel D. Prevalence odds ratio or prevalence ratio in the analysis of cross sectional data: what is to be done? Occup Environ Med 1998;55(4):272-7. (13) Zou G. A modified poisson regression approach to prospective studies with binary data. Am J Epidemiol 2004 Apr 1;159(7):702-6. (14) Armitage P, Berry G, Matthews JNS. Statistical Methods in Medical Research. Fourth edition. 4 ed. Oxford: Blackwell Science; 2005.

83

Chapter 7. Paper 3

(15) O'Brien RM. A Caution Regarding Rules of Thumb for Variance Inflation Factors. Quality and Quantity 2007;41(5):673-90. (16) Stata Statistical Software: Release 9.0. College Station, TX: StataCorp LP; 2005. (17) Protti D. Comparison of information technology in general practice in 10 countries. Healthc Q 2007;10(2):107-16. (18) Tang ST, McCorkle R. Determinants of place of death for terminal cancer patients. Cancer Invest 2001;19(2):165-80. (19) Cantwell P, Turco S, Brenneis C, Hanson J, Neumann CM, Bruera E. Predictors of home death in palliative care cancer patients. J Palliat Care 2000;16(1):23-8. (20) Brazil K, Bedard M, Willison K. Factors associated with home death for individuals who receive home support services: a retrospective cohort study. BMC Palliat Care 2002 Mar 25;1(1):2-7. (21) Burge F, Lawson B, Johnston G, Cummings I. Primary care continuity and location of death for those with cancer. J Palliat Med 2003 Dec;6(6):911-8. (22) Aabom B, Kragstrup J, Vondeling H, Bakketeig LS, Stovring H. Population-based study of place of death of patients with cancer: implications for GPs. Br J Gen Pract 2005 Sep;55(518):684-9. (23) Fukui S, Fukui N, Kawagoe H. Predictors of place of death for Japanese patients with advanced-stage malignant disease in home care settings: a nationwide survey. Cancer 2004 Jul 15;101(2):421-9. (24) Howat A, Veitch C, Cairns W. A retrospective review of place of death of palliative care patients in regional north Queensland. Palliat Med 2007 Jan;21(1):41-7.

84

Chapter 8. Paper 4

CHAPTER 8 PAPER 4 Associations between successful palliative cancer pathways and general practitioner involvement Neergaard MA, Vedsted P, Olesen F, Sokolowski I, Jensen AB, Sondergaard J

Submitted to Scandinavian Journal of Primary Health Care

85

Chapter 8. Paper 4

8.1. Abstract Background: General practitioner (GP) involvement may be instrumental in obtaining successful palliative cancer pathways. Aim: To examine associations between bereaved relatives‟ evaluation of palliative cancer pathways and place-of-death and GP involvement. Design: Population-based, cross-sectional combined register and questionnaire study Setting: Aarhus County, Denmark. Method: Questionnaires seeking data on GPs‟ effort and relatives‟ evaluation of the palliative pathway were obtained for 153 cases of deceased cancer patients. Associations with a successful pathway were analysed. Results: Successful palliative pathways were statistically significantly associated with home-death. The variables „Relative living with patient‟ and „GP having contact with relatives‟ were also associated with a successful pathway. Conclusion: Home-death was statistically significantly associated with a successful palliative pathway according to the relatives. The relation between relative and patient and between GP and relative may also be of importance. Further research is needed to identify predictors associated with a successful pathway in primary care to optimise and focus the palliative primary health care effort. 8.2. Introduction Most terminally ill cancer patients and their relatives wish that it would be possible for the patient to die at home (1-4). Home-death, however, is not tantamount to a successful palliative pathway according to the relatives. Palliative home care often involves numerous health care workers with different professional backgrounds and their effort can affect relatives‟ view on the palliative pathway. The primary frontline professionals of palliative care in the patients‟ homes are general practitioners (GPs) and community nurses.

Two interview studies suggest that 24-hour back-up and involvement of the GP is an important element in bereaved relatives‟ evaluation of palliative pathways (5;6). Furthermore, relatives report CN and GP involvement as two of the most valuable services in the palliative pathway (7). However, it remains unclear what constitutes specifically important GP contributions in achieving a successful palliative pathway. The aim of the present study was to examine associations between the bereaved relatives‟ evaluation of palliative pathways, place of death and GPs‟ involvement.

86

Chapter 8. Paper 4

8.3. Materials and methods We conducted a survey in 2007 (January – July) including 231 general practices with 599 registered deceased cancer patients in Aarhus County (at that time approximately 640 000 inhabitants, equal to 12% of the Danish population and 1680 persons dying from cancer each year (2005)) (8). We subsequently conducted a postal questionnaire study targeting the closest relative of the deceased cancer patient. In Denmark, all citizens are registered with unique civil registration numbers(9) and by means of these numbers, questionnaire data were linked to health register data. Setting Danish health care system is tax-financed and more than 98% of Danes are registered with a GP and receive free medical care (10). Danish GPs act as gatekeepers for access to specialist treatment and are responsible for frontline care 24 hours a day. Large GP corporations provide out-of-hours care (4 pm. to 8 am and Sat/Sun), covering an entire county.

Denmark has no formal national agreement on task distribution in palliative care, but the Aarhus County provided a special fee for GPs‟ palliative care involvement. Community nurses employed by the municipalities are often involved in palliation and visit patients on a 24-hour basis. Palliative specialist outgoing teams based at the major hospitals are available during daytime hours, and specialist advice from these teams can be obtained by GPs or community nurses by telephone. Study population and sampling We included adults in Aarhus County who died from cancer from 1st March to 30th November 2006 and who had received some palliative home care either from GPs, community nurses or a palliative specialist team. Since no database on palliative patients was available and since the Danish Register of Causes of Death was not updated on 2006-deaths, we sampled the patients by combining official register data with questionnaire information obtained from GPs and relatives.

From the county hospital discharge register we identified 29,043 individuals above 18 years who were registered with at least one cancer diagnosis (ICD-10) (excluding non-melanoma skin cancers) during the period November 1996 to November 2006. In December 2006, using the Civil Registration System database, we identified 813 persons among the 29,042 who died from 1st March to 30th November 2006 (nine months). From the regional health authority‟s register we identified the GPs of these patients. Eight (1.0%) persons were not registered with a GP and 18

87

Chapter 8. Paper 4

(2.2%) had moved from the county after having been diagnosed, leaving 787 deceased cancer patients. A questionnaire was sent to their GPs who were asked about cause of death and if palliative care had been provided in the patients‟ homes. The GPs were also asked whether it was acceptable that we contacted the closest bereaved relative, asking them to fill in a questionnaire. The closest relative was identified by the GPs. When a GP stated that we could contact the relative, but without providing name or address, we used the following sequence: spouse, child above 18 years, oldest sibling and parent (data from the Civil Registration System database). We asked the person contacted to give the questionnaire to the deceased cancer patient‟s closest relative.

In late 2008, data from the Danish Register of Causes of Death for deaths in 2006 was available. Merging the information of cancer deaths with our database, we excluded 188 patients who were not registered with cancer as the death reason which reduced our study population to 599 deceased cancer patients (Figure 8.1). Figure 8.1. Flow-chart of sampling and GP questionnaire. Responders and non-responders.

Flow-chart GP questionnaires 29043 Patients with a cancer diagnosis in Aarhus County within 10 years before17 November 2006 28230 Alive or dead before 1st March 2006 or < 18 years old 813 Cancer patients in Aarhus County died between 1 March and 30 November 2006 8 Were not registered with GP and 18 moved out of the area 787 Cancer patients in Aarhus County died between 1 March and 30 November 2006 registered with a GP

350 GP questionnaires were filled in

437 Questionnaires were not filled in 2008: Data from The Danish Register of Causes of Death



599 Patients registered as cancer deaths in Aarhus County died between 1 March and 30 November 2006 registered with a GP 333 GP questionnaires were filled in


194 Non-responders 9 GP retired, were ill or had died 73 GP did not want to participate 87 GP did not respond 25 GP stated :“Not cancer death”

72 Cases excluded No palliative home care

88

Chapter 8. Paper 4

Data collection GP questionnaires included themes identified through literature studies, clinical experience and group-interview studies with bereaved relatives (5) and involved professionals (GPs, community nurses and hospital consultants). The 72-item GP questionnaire was pilot-tested among 30 GPs in another Danish county. Important GP questionnaire themes were: the GP‟s knowledge of the patient prior to the palliative period, whether the GP provided a plan on whom to contact during out-of-office hours, the GP‟s contact to the patient‟s relatives and the duration of the palliative period. In partnership practices, we asked for the GP who was supposedly most familiar with the patient to answer the questionnaire. GPs received a small economic compensation for their efforts, and GP non-responders were sent reminders four and seven weeks following the first questionnaire. Questionnaire data from the GP questionnaires included items on the patient‟s marital status (single, married or cohabiting), the patient‟s children (no, yes at home, yes not at home), the GP‟s knowledge of the patient before the palliative period (dichotomized into poor (1,2 on a 1-5-point scale) and good (3, 4 and 5)), whether the GP provided a private phone number for direct contact out-of-hours (no/yes), whether the GP had contact to the patient‟s relatives (no/yes), whether community nurses (no/yes) or a specialist team (no, yes) were involved and the duration of the palliative period at home and in total. The palliative period was defined as the last period of the patient‟s life during which all curative treatment had been stopped and care and treatment were solely palliative.

Important themes in the questionnaire for relatives were demographic characteristics of the relatives and their evaluation of the palliative pathway. The relatives were asked: „How, in your own words, was the entire period at home during which the deceased was dying compared with how you feel it should have been?‟ with the options: „Very well‟, „Well‟, „Fairly well‟, „Bad‟, „Very bad‟. The 65-item questionnaire to relatives was pilot-tested among 14 bereaved relatives not included in this study. Non-responders of relatives were sent one reminder four weeks after the first questionnaire.

From the relatives we obtained data on their age (18-65, 66+), gender, relation to deceased (nonspouse, spouse), living with the patient (no, yes), vocational educational level (3 years or less, > 3 years) and their evaluation of the palliative pathway of disease (dichotomized into „Not successful‟ („Fairly well‟, „Bad‟, „Very bad‟) and „Successful‟ („Very well‟, „Well‟)).

89

Chapter 8. Paper 4

We retrieved register data on the patient‟s age (18-65, 66+), gender, cancer diagnosis (lung, colorectal, breast, prostate, other), place of death (institution (hospital and hospice), nursing home, home), home-visits provided by GPs (0, 1, 2, 3, >3, and dichotomised into no/yes) in the period three months prior to the patient‟s death. Analysis ‟A successful palliative pathway‟ was defined as the outcome measure and associations with patient, GP and palliative pathway variables were calculated.

Unadjusted and adjusted associations were calculated. Using robust variance estimates, the estimates were adjusted for clustering of patients in practices (11). Prevalence ratios (PRs) with 95% confidence intervals (95% CI) were used as a measure of association. Due to the high prevalence of the outcome measure (more than 20% „successful palliative pathways‟), odds ratios would overestimate the association (12;13). PRs were calculated with generalised linear models (GLM) with log link and the Bernoulli family, and when the model could not converge, we used the Poisson regression model (12;14).

The variables were assessed for collinearity (Pearson's correlation coefficient > 0.4) and multicollinearity (variance inflation factor < 10) (15;16). Due to collinearity, „Relative relation to diseased‟ was not included in the model. We added the duration of the palliative period spent at home as a confounder, since it could be associated with the GPs‟ possibility to provide palliative services. Data were analyzed using STATA 10 (17). Ethics According to Scientific Committee for the County of Aarhus, the Biomedical Research Ethics Committee System Act does not apply here as no active intervention was involved in the study. The study was approved by the Danish Data Protection Agency and the Danish National Board of Health. 8.4. Results Among the 599 questionnaires sent to GPs, a total of 333 questionnaires from 231 general practices were filled in and 72 were excluded because the GP stated that no home care was provided during the palliative pathway (response rate 63.2%) (Figure 8.1). The general practices completed questionnaires for 1-6 patients (mean: 1.9 (SD:1.1)). The non-responding practices were not statistically significantly different from the participating practices with respect to practice

90

Chapter 8. Paper 4

organisation (single or partnership), number of patients per GP and number of questionnaires sent per general practice (data not shown). The 194 cases from non-responding GPs were not statistically significantly different from the included cases in terms of the patients‟ gender, place of death and number of GP home-visits, but the patients in the non-responder cases tended to be significantly older (Table 8.1).

In 106 (31.8%) of the 333 cases, the GPs advised against sending the relative a questionnaire (Figure 8.2). Among the 227 questionnaires sent to bereaved relatives, four cases were excluded since the relatives stated that the questionnaire was not relevant to the pathway at all and a total of 153 relative questionnaires were filled in (response rate 68.6%) (Figure 8.2). Figure 8.2. Flow-chart of questionnaire to bereaved relatives. Responders and non-responders.

Flow-chart Questionnaires for bereaved relatives 333 GPs were asked if the relative could receive a questionnaire 106 cases excluded by the GP for the following reasons: 24: Relative mentally disabled 2: Relative linguistically disabled 3: Relative dead 2: Relative too sick 1: No name or address of the relative could be found 1: Relative living abroad 1: Relative did not participate at all 1: Patient not registered with the GP 71: No reason stated

227 Questionnaires for relatives were sent out 70 Non-responders

4 Excluded

8: Questionnaires were returned unanswered 2: Relatives too sick to answer 55: Relatives did not respond 5: Postal services did not find relative on the given address

Relative did not find the course of disease relevant for our study

153 Questionnaires for relatives were filled in

Comparison of the 153 included cases with the 70 cases where the relatives did not respond showed that relatives of male patients tended to answer the questionnaire statistically significantly more frequently than relatives of female patients (Table 8.2).

91

Chapter 8. Paper 4

Comparing the 153 included cases with the 106 cases where GPs advised against sending the relative a questionnaire showed that patients in the cases included comprised more males, more home-deaths and were characterised by more involvement by GP (on several variables), community nurses and a palliative specialist team (Table 8.2). Associations with evaluation of palliative pathway „A successful palliative pathway‟ was statistically significantly associated with home-death compared with institutional-death (PR: 1.5 (95%CI: 1.0; 2.1)) (Table 8.3). A successful palliative pathway was also associated with the patient living with the relative (PR: 1.7 (0.8; 3.5)) and with the GP having contact with the relatives (PR: 1.6 (0.5; 5.4)). However, these associations did not reach statistically significance. 8.5. Discussion Main findings In a group of patients who died from cancer and had a palliative pathway at home, we found that the relatives‟ positive evaluation of the palliative pathway was associated with home-death. However, we identified no statistically significant associations between the evaluations and the ways in which the GPs were involved or specific factors in relation to the GPs, but a relation between the GP and the relative tended to increase the likelihood of a positive evaluation of the palliative pathway. Strengths and limitations of the study The strengths of this study are its sampling procedure and its comprehensive data collection. To eliminate differential misclassification, we used the standardised official health registers to identify the study population, including the places and the reasons of death. However, in 25 of the 599 cases, GPs returned the questionnaires stating that the patient did not die from cancer. These cases were registered as non-responders, which may have introduced bias (incl. confounding by indication), but the direction and importance of this bias is hard to establish. To minimise recall bias, the questionnaire was sent in January 2007 instead of awaiting the update of the Danish Register of Causes of Death in 2008. Furthermore, all Danish GPs are able to consult electronic patient files when completing a questionnaire (18).

The major weakness of these analyses was the selection bias. We found many differences between included cases and the cases whose relatives were excluded by GPs (e.g. cases included in the study had statistically significantly more home-deaths, more involvement by GP, community

92

Chapter 8. Paper 4

nurses and a palliative specialist team). The differences meant that we would tend to exclude those cases where the GP was not as involved and did not know the patient as well as in those cases that were included. In these excluded cases the relative may evaluate the pathway as less successful. Thus, we would tend to overestimate the associations between a successful pathway and GP-related variables. This would also be the case if GPs excluded cases they knew had been unsuccessful (which they might have done for several reasons) despite a high level of involvement and efforts to support the palliative pathway. As home-death is associated with GP involvement (19-22), such selection bias, therefore, tends to produce a higher association between home-death and a successful pathway in this study.

Approximately 1680 patients died from cancer in Aarhus County in 2006, and we included 599 cases recruited during a nine-month period (Figure 8.1). This lower amount of cases can be explained by the fact that we did not include persons less than 18 years, non-melanoma skin cancer, and the fact that we included only those with a cancer diagnose registered in a hospital in Aarhus County as the main diagnose of admittance in a 10-year period. Furthermore, the period where patients could die (1st March-30th November) was without the winter months of 2006, which may account for some of the missing cases, since winter months may have a higher average of deaths than the rest of the year. Comparison with existing literature Quality in dying is much debated and there is no doubt that what constitutes a good death is determined by a complex interplay of personal and cultural values, supportive network characteristics, and physical and medical factors, as well as the efforts displayed by the health care system. To our knowledge, no other quantitative studies have examined the association between relatives‟ or patients‟ general evaluation of the palliative pathway and GP-related factors. Surprisingly, we found that none of the GP-related factors in our model were statistically significantly associated with the relatives‟ evaluation of the palliative pathway, not even GP home-visits even though prior studies have reported that such home-visits are strongly associated with home-death(19;21;22), and that the GPs‟ effort in the palliative pathway were important to the relatives (5). Previous studies show that GP providing many home-visits (>20) or willingness to perform home-visits in the palliative pathway are positively associated with bereaved relatives‟ satisfaction with the GP involved(23;24), but in our study no association with the evaluation of the palliative pathway in general was found.

93

Chapter 8. Paper 4

We found that home-death was statistically significantly associated with the relatives‟ evaluation of the palliative pathway as successful. Other studies have found that bereaved relatives‟ retrospect satisfaction with the GP involved was statistically higher when the GP visited the patient 20 times or more than if he visited less (23). Furthermore, other studies found that home-death was associated with better bereavement response (19;21;22) and overall satisfaction of the palliative pathway (25). We did not include any variables of symptom severity. The difference between bereaved relatives‟ evaluation of pathways with home-death and institutional-death may therefore be due to the fact that patients dying at a hospital or hospice may suffer worse symptoms, e.g. pain, than patients dying at home, which, indeed, would affect the relatives‟ evaluation. However, patients with severe symptoms may also have more contact with a palliative specialist team, and adjusting for this we may have adjusted for some of the confounding rooted in symptom severity. Eliminating „specialist team involvement‟ from the model accordingly weakens the association between a successful pathway and home-death. Further research is needed to investigate the implications for the relatives of the patients‟ death at home.

We also found indications that a successful palliative pathway was associated with the patient living with the relative and the GP‟s contact with the relatives. These associations did not reach statistically significance in the present study which may be due to the fact that only 153 cases were included in the analysis, and that the GPs and relatives did not answer all the questions: only 100 cases were used in the adjusted analysis. If we could have included more cases, we would have obtained more power. However, it seems that the relatives‟ relations to patient and to the GP are important to achieving a successful palliative pathway.

Implications for future research The factors of GP involvement that we included in our model showed no statistical significance in relation to relatives‟ evaluation of the palliative pathway and there is a need for further research of the predictive power of a more active approach in primary health services for a successful palliative pathway and our results also necessitate further research into how to measure bereaved relatives‟ evaluation of the palliative pathway. Furthermore, the fact that the association between a successful pathway and home-death may partly be due to selection bias questions the use of home-death as a solid variable measuring a good death in future research.

94

Chapter 8. Paper 4

8.6. Conclusion Our study indicates that dying at home is positively associated with a higher likelihood that the bereaved relative will evaluate the palliative pathway as successful. The relation between the bereaved relative and the deceased cancer patient and between the GP and the relative also seems to be of importance. Further research is needed of predictors associated with a successful pathway in primary care to optimise and focus the palliative primary health care effort. Acknowledgements Profound gratitude is extended to participating GPs. The study was funded by the Aarhus County Research Fund for the Clinical Development and Research in General Practice and across the Primary and Secondary Health Care Sectors (4-01-3-04), the Danish National Research Foundation for Primary Care (585-457808) and The Multipractice Study Committee (585-04/2072). Conflicts of interest The authors have no conflicts of interest.

8.7. Tables Table 8.1. Characteristics of 333 included cases and 194 cases not included because the GP did not respond. Case data obtained from formal health registers. Cases of GP responders (N=333)

Cases of GP nonresponders (N=194)

Patient’s age at time of death (mean (SD))

69.4 (12.7)

73.5 (12.7)*

Patient’s gender (n (%)) Male Female

181 (54.4) 152 (45.6)

113 (58.3) 81 (41.7)

Primary cancer diagnosis (n (%)) Bronchus/lung 65 (19.5) Colon/rectum 50 (15.1) Breast 34 (10.2) Prostate 39 (11.7) Other 145 (43.5) Place of death (n (%)) Home 120 (36.0) Nursing home 69 (20.7) Hospital/hospice 140 (42.0) Other (e.g. other institution) 4 (1.2) Number of home visits by GP during the last 3 months 3 (1;6) (median (IQI) 3.9 (3.2) (mean (SD)) * Statistically significantly different from cases of GP responders (p-value < 0.05) Not all sums of percentages are added to 100.0% because of round-offs.

37 (19.1) 32 (16.5) 22 (11.3) 31 (16.0) 72 (37.1) 70 (36.1) 51 (26.3) 70 (36.1) 3 (1.5) 3 (1;5) 3.6 (3.1)

95

Chapter 8. Paper 4

Table 8.2. Characteristics of 153 included cases, the 70 cases not included because the relative did not respond and the 106 cases where the GP advised against sending the relative a questionnaire. Case data in study arise from the GP and relative-questionnaires and from formal health registers. Case data of relative-non-responders and of the group where the GP advised against sending the relative a questionnaire stem from GP questionnaires and formal health registers.

Patient’s age at time of death (mean years (SD)) Patient’s gender (n (%)) Male Female Primary cancer diagnosis (n (%)) Bronchus/lung Colon/Rectum Breast Prostate Other Place of death (n (%)) Home Nursing home Hospital / Hospice Other(e.g. other institution) Gender of relative (n (%)) Male Female Relation to deceased (n (%)) Spouse Girlfriend or Boyfriend Daughter or son Sister or brother Parent Friend Daughter-in-law Relative lived with patient (n (%)) No Yes Relative’s vocational education (n (%)) 3 years or less > 3 years GP involvement (n (%)) No Yes GP knowledge of patient prior to palliative period (n (%)) Poor Well

Cases in the study (N=153)

Cases of relative nonresponders (N=70)

Cases where the GP advised against sending the relative a questionnaire (N=106)

68.2 (12.8)

69.6 (12.2)

70.9 (12.5)

95 (62.1) 58 (37.9)

33 (47.1)* 37 (52.9)*

50 (47.2)* 56 (52.8)*

25 (16.3) 22 (14.4) 17 (11.1) 22 (14.4) 67 (43.8)

18 (25.7) 12 (17.1) 7 (10.0) 6 (8.6) 27 (38.6)

22 (20.8) 15 (14.2) 10 (9.4) 11 (10.4) 48 (45.2)

76 (49.7) 23 (15.0) 53 (34.6) 1 (0.7)

25 (35.7) 11 (15.7) 33 (47.2) 1 (1.4)

18 (17.0)* 34 (32.0)* 52 (49.1)* 2 (1.9)

42 (27.4) 111 (72.6)

-

-

-

-

27 (18.2) 121 (81.8)

-

-

94 (64.4) 52 (35.6)

-

-

11 (7.3) 140 (92.7)

5 (7.3) 64 (92.7)

20 (19.4)* 83 (80.6)*

14 (9.5) 133 (90.5)

7 (10.0) 63 (90.0)

23 (22.6)* 79 (77.4)*

106 (69.3) 1 (0.7) 39 (25.4) 2 (1.3) 2 (1.3) 1 (0.7) 2 (1.3)

96

Chapter 8. Paper 4

Number of GP home visits during the last 3 months (Median (IQI)) 3 (2;6) 4 (2;7) (mean (SD)) 4.2 (3.2) 4.4 (3.0) GP home visits during the last 3 months (n (%)) No 11 (7.2) 5 (7.1) Yes 142 (92.8) 65 (92.9) Unplanned home visits by GP (n (%)) No 58 (43.9) 27 (43.6) Yes 74 (56.1) 35 (56.4) GP gave private number to patient to use in outof-office hours (n (%)) No 68 (50.7) 33 (53.2) Yes 66 (49.3) 29 (46.8) GP had made a plan with the patient for whom to contact in out-of-office hours (n (%)) No 92 (68.2) 43 (68.3) Yes 43 (31.8) 20 (31.7) GP had contact with relatives (n (%)) No 8 (5.8) 5 (7.8) Yes 131 (94.2) 59 (92.2) Home care nurse involvement (n (%)) No 42 (27.5) 18 (25.7) Yes 111 (72.5) 52 (74.3) Specialist team involvement (n (%)) No 85 (55.6) 43 (61.4) Yes 68 (44.4) 27 (38.6) * Statistically significantly different from the 153 cases in the study (p-value < 0.05)

2 (0;5) 3.2 (3.3)* 27 (25.5)* 79 (74.5)* 46 (61.3)* 29 (38.7)*

61 (75.3)* 20 (24.7)*

43 (52.4)* 39 (47.6)* 26 (34.2)* 50 (65.8)* 48 (45.3)* 58 (54.7)* 74 (69.8)* 32 (30.2)*

97

Chapter 8. Paper 4

Table 8.3. Associations between a successful palliative course and model variables. A total of 153 cases were included in the analyses. The unadjusted and the adjusted prevalence ratios (PRs) are shown with 95% confidence intervals (95%CIs) Unadjusted Prevalence ratio (95 % CI)

Adjusted pvalue

Gender of relative Male 1 Female 1.1 (0.8;1.4) 0.559 Age of relative 18 - 64 1 0.013 65+ 1.3 (1.1;1.6) Relative living with patient No 1 0.011 Yes 1.9 (1.2; 3.1) Relative’s relation to diseased Not spouse 1 0.005 Spouse 1.6 (1.2;2.2) Relative’s vocational education 3 years or less 1 > 3 years 0.8 (0.6:1.0) 0.102 GP knowledge prior to palliative period Poor 1 Well 1.4 (0.8;2.4) 0.211 GP home visits No 1 Yes 1.1 (0.6;1.9) 0.719 Unplanned home visits by GP No 1 Yes 1.1 (0.9;1.5) 0.335 GP gave private number to patient to use in out-of-office hours No 1 Yes 1.0 (0.8;1.3) 0.800 GP had made a plan with the patient for whom to contact in out-of-office hours No 1 Yes 1.0 (0.7;1.3) 0.767 GP had contact with relatives No 1 Yes 1.1 (0.6;2.4) 0.712 Community nurse involvement No 1 Yes 0.9 (0.7;1.2) 0.470 Specialist team involvement No 1 Yes 1.0 (0.8;1.3) 0.917 Place of death Institution (hospital or hospice) 1 Nursing home 0.9 (0.5;1.4) 0.636 0.021 Home 1.4 (1.0;1.8) Note: Significant correlations with a p-value < 0.05 are in bold text.

Prevalence ratio (95 % CI)

p-value

1 1.0 (0.7;1.5)

0.882

1 1.0 (0.8;1.3)

0.988

1 1.7 (0.8;3.5) 0.119 Not included because of collinarity with „Relative living with patient‟ 1 0.9 (0.7;1.2)

0.397

1 1.2 (0.7;2.1)

0.405

1 1.0 (0.4;2.2)

0.951

1 1.2 (0.9;1.5)

0.238

1 0.8 (0.6;1.1)

0.205

1 1.0 (0.7;1.5)

0.917

1 1.7 (0.5;5.2)

0.360

1 0.9 (0.6;1.3)

0.524

1 1.0 (0.8;1.4)

0.808

1 1.0 (0.5;2.0) 1.5 (1.0;2.1)

0.940 0.031

98

Chapter 8. Paper 4

8.8. References of Chapter 8 (1) Hanratty B. Palliative care provided by GPs: the carer's viewpoint. Br J Gen Pract 2000 Aug;50(457):653-4. (2) Thomas C. The place of death of cancer patients: can qualitative data add to known factors? Soc Sci Med 2005 Jun;60(11):2597-607. (3) Thomas C, Morris SM, Clark D. Place of death: preferences among cancer patients and their carers. Soc Sci Med 2004 Jun;58(12):2431-44. (4) Grande GE, Farquhar MC, Barclay SI, Todd CJ. Valued aspects of primary palliative care: content analysis of bereaved carers' descriptions. Br J Gen Pract 2004 Oct;54(507):772-8. (5) Neergaard MA, Olesen F, Jensen AB, Sondergaard J. Palliative care for cancer patients in a primary health care setting: Bereaved relatives' experience, a qualitative group interview study. BMC Palliat Care 2008 Jan 15;7(1):1. (6) Weibull A, Olesen F, Neergaard MA. Caregivers' active role in palliative home care - to encourage or to dissuade? A qualitative descriptive study. BMC Palliat Care 2008 Sep 16;7(1):15. (7) Brazil K, Bedard M, Krueger P, Abernathy T, Lohfeld L, Willison K. Service preferences among family caregivers of the terminally ill. J Palliat Med 2005 Feb;8(1):69-78. (8) The Danish National Board of Health. The Danish Register of Causes of Death [In Danish]. 2008. (9) Pedersen CB, Gotzsche H, Moller JO, Mortensen PB. The Danish Civil Registration System. A cohort of eight million persons. Dan Med Bull 2006 Nov;53(4):441-9. (10) de Fine Olivarius N, Hollnagel H, Krasnik A, Pedersen PA, Thorsen H. The Danish National Health Service Register. Dan Med Bull 1997;44(4):449-53. (11) Donner A, Klar N. Design and Analysis of Cluster Randomisation Trials in Health Research. 1 ed. London: Hodder Arnold; 2000. (12) Barros AJ, Hirakata VN. Alternatives for logistic regression in cross-sectional studies: an empirical comparison of models that directly estimate the prevalence ratio. BMC Med Res Methodol 2003 Oct 20;3(1):21. (13) Thompson ML, Myers JE, Kriebel D. Prevalence odds ratio or prevalence ratio in the analysis of cross sectional data: what is to be done? Occup Environ Med 1998;55(4):272-7. (14) Zou G. A modified poisson regression approach to prospective studies with binary data. Am J Epidemiol 2004 Apr 1;159(7):702-6. (15) Armitage P, Berry G, Matthews JNS. Statistical Methods in Medical Research. Fourth edition. 4 ed. Oxford: Blackwell Science; 2005. (16) O'Brien RM. A Caution Regarding Rules of Thumb for Variance Inflation Factors. Quality and Quantity 2007;41(5):673-90.

99

Chapter 8. Paper 4

(17) Stata Statistical Software: Release 9.0. College Station, TX: StataCorp LP; 2005. (18) Protti D. Comparison of information technology in general practice in 10 countries. Healthc Q 2007;10(2):107-16. (19) Aabom B, Kragstrup J, Vondeling H, Bakketeig LS, Stovring H. Population-based study of place of death of patients with cancer: implications for GPs. Br J Gen Pract 2005 Sep;55(518):684-9. (20) Gomes B, Higginson IJ. Factors influencing death at home in terminally ill patients with cancer: systematic review. BMJ 2006 Mar 4;332(7540):515-21. (21) Fukui S, Fukui N, Kawagoe H. Predictors of place of death for Japanese patients with advanced-stage malignant disease in home care settings: a nationwide survey. Cancer 2004 Jul 15;101(2):421-9. (22) Howat A, Veitch C, Cairns W. A retrospective review of place of death of palliative care patients in regional north Queensland. Palliat Med 2007 Jan;21(1):41-7. (23) Fakhoury W, McCarthy M, Addington-Hall J. Determinants of informal caregivers' satisfaction with services for dying cancer patients. Soc Sci Med 1996 Mar;42(5):721-31. (24) Lecouturier J, Jacoby A, Bradshaw C, Lovel T, Eccles M. Lay carers' satisfaction with community palliative care: results of a postal survey. South Tyneside MAAG Palliative Care Study Group. Palliat Med 1999 Jul;13(4):275-83. (25) Ringdal GI, Jordhoy MS, Kaasa S. Family satisfaction with end-of-life care for cancer patients in a cluster randomized trial. J Pain Symptom Manage 2002 Jul;24(1):53-63.

100

Chapter 9. Paper 5

CHAPTER 9 PAPER 5 Associations between successful palliative cancer pathways and community nurse involvement Neergaard MA, Vedsted P, Olesen F, Sokolowski I, Jensen AB, Sondergaard J Submitted to BMC Palliative Care

101

Chapter 9. Paper 5

9.1. Abstract Background: Most terminally ill cancer patients and their relatives wish that the patient dies at home. Community nurses (CNs) are often frontline workers in the patients‟ homes and CN involvement may be important in attaining successful palliative pathways. The aim of the present study was to examine associations between bereaved relatives‟ evaluation of palliative pathways and place of death and CN involvement. Methods: The study is a population-based, cross-sectional combined register and questionnaire study performed in Aarhus County, Denmark, including CN questionnaires obtaining data on CNs‟ efforts, GP-questionnaires obtaining data on pathway characteristics and questionnaires to relatives evaluating the palliative pathway were sent in relation to 599 deceased cancer patients. Associations between bereaved relatives‟ evaluation of palliative pathways and place of death and CN involvement were analysed. Results: „A successful palliative pathway‟ was positively associated with home-death and death at a nursing home compared with institutional death. Conclusions: Our study indicates that dying at home is positively associated with a higher likelihood that the bereaved relative will evaluate the palliative pathway as successful. Symptom severity may be an important confounder unadjusted for in the present study. No significant associations were identified between the evaluations of the palliative pathway and how the CNs were involved and the degree to which they were involved. 9.2. Introduction Most terminally ill cancer patients and their relatives wish that the patient dies at home (1-4). However, home-death is not necessarily tantamount to a successful palliative pathway, since many factors and the professionals involved can influence the degree of successfulness in the patients‟ or relatives‟ views. Palliative home care often involves many different health-professionals, and the primary care professionals, community nurses (CNs) and general practitioners (GPs) are often frontline workers in the patients‟ homes. Previous studies suggest that 24-hour back-up and involvement of the CN are count as important factors in bereaved relatives‟ evaluation of palliative pathways (5-7). Furthermore, research has also shown that involvement of a CN is positively associated with home-death (8-10). However, knowledge on the importance of their involvement in relation to achieving a successful palliative pathway is lacking. The aim of the present study was to examine the association between the bereaved relatives‟ evaluation of palliative pathways and the place of death and the CNs‟ involvement.

102

Chapter 9. Paper 5

9.3. Method We conducted a survey in 2007 (January – July) of 231 general practices with 599 registered deceased cancer patients in Aarhus County (at that time approximately 640 000 inhabitants, equal to 12% of the Danish population and with 43 municipalities and 1680 persons die from cancer each year (2005)) (11). We subsequently conducted a postal questionnaire study targeting the CNs and the closest relative of the deceased cancer patient. In Denmark, all citizens are registered with unique civil registration numbers (12) and by means of these numbers, questionnaire data were linked to health register information.

Setting Danish health care system is tax-financed and provides free home care by CNs and home carers for those in need of help at home. CNs are employed by the municipalities are often involved in palliation and visit patients on a 24-hour basis.

Denmark has no formal national agreement on task distribution in palliative care, but the Aarhus County provided a special fee for GPs‟ palliative care involvement. Palliative specialist outgoing teams based at the major hospitals are available during daytime hours, and specialist advice can be obtained from these specialist teams by CNs or GPs by telephone. Study population and sampling We included adults in Aarhus County who died from cancer from 1st March to 30th November 2006 and who had received some palliative home care either from CNs, GPs or a palliative specialist team. Since no database on palliative patients was available and since the Danish Register of Causes of Death was not updated on 2006-deaths, we sampled the patients by combining official register data with questionnaires information.

From the county hospital discharge register we identified 29,043 individuals above 18 years who were registered with at least one cancer diagnosis (ICD-10) (excluding non-melanoma skin cancers) during the period November 2006 and ten years back. In December 2006, using the Civil Registration System database, we identified 813 patients among the 29,042 who died from 1st March to 30th November 2006 (nine months). From the regional health authority‟s register we identified their GPs. Eight (1.0%) patients were not registered with a GP and 18 (2.2%) had moved from the county after having been diagnosed, leaving 787 deceased cancer patients for analysis. A CN questionnaire was sent to the local health centre of the patient‟s address and a GP questionnaire was sent to the patient‟s GP. They were asked about cause of death, if palliative

103

Chapter 9. Paper 5

care had been provided in the patient‟s home and if the closest bereaved relative could receive a questionnaire. If one of the two (CN or GP) advised against sending the relative a questionnaire, it was not sent. The closest relative was defined by the professionals, but if they stated that we could contact the relative but without providing the name or the address, we sent the questionnaire to the relative in the following order: Spouse, child above 18 years, oldest sibling and parent (data from the Civil Registration System database). We asked the person contacted to give the questionnaire to the closest relative involved in the palliative pathway.

In late 2008, data from the Danish Register of Causes of Death for deaths in 2006 was available. Merging the information of cancer deaths with our database, we excluded 188 patients who were not registered with cancer as the death reason, which reduced our study population to 599 deceased cancer patients (Figure 9.1). Figure 9.1. Flow-chart of sampling of study population and data collection. Community nurses (CNs) questionnaires. Responders and non-responders.

Flow-chart Community nurse (CN) questionnaire 29043 Patients with a cancer diagnosis in Aarhus County within 10 years before17th November 2006 28230 Alive or dead before 1st March 2006 or < 18 years old 813 Cancer patients in Aarhus County died between 1 March and 30 November 2006 8 Were not registered with GP and 18 moved out of the area 787 Cancer patients in Aarhus County died between 1 March and 30 November 2006 registered with a GP 72 Excluded by GPs because of no palliative home care 507 Questionnaires were not filled in

208 CN questionnaires were filled in

2008: Data from The Danish Register of Causes of Death 7 Not registered as cancer deaths


599 Patients registered as cancer deaths in Aarhus County died between 1 March and 30 November 2006 registered with a GP 201 CN questionnaires were filled in


220 Cases with nurses non-responders 17: CN not working at the nursing centre anymore 11: Nursing centre leader not wanting nurse to participate 7: CN did not have the time to participate 20: CN did not recollect the patient 163: CN did not respond 2: CN stated that patient did not die from cancer

178 Cases excluded 165: No palliative home care 13: Nursery home resident

104

Chapter 9. Paper 5

Data collection The questionnaires included themes identified through literature studies, clinical experience and group interview studies with bereaved relatives (5) and involved professionals (CNs, GPs and hospital consultants). The 46-item CN questionnaire was pilot-tested among 14 CNs who had also participated in a prior interview study about palliative care. Important themes were: nurse characteristics, nurse‟s knowledge of the patient prior to the palliative period, contact to relatives and nurse home-visits. If the CNs did not respond, two reminders were sent four and seven weeks following the first questionnaire, respectively. Thus, questionnaire data from the CNs included information on the nurse‟s age (min-39, 40+), number of years as CN (0-5, 6+), amount of extra education or classes in palliative care (no, yes), knowledge of the patient before the palliative period (dichotomized into poor (1,2 on a 1-5-point scale) and good (3, 4 and 5)), whether the nurse had contact to the patient‟s relatives (no/yes) and how often the nurse or community services paid the patient a home-visit (less than once a day, once a day or more).

The 72-item GP questionnaire was pilot-tested among 30 GPs in another Danish county. In this study, GP questionnaires were mainly used to locate relatives, to inform that a palliative pathway had taken place at home or not, and to determine the duration of the palliative period. GPs received a small economic compensation for their efforts, and GP non-responders were sent reminders four and seven weeks following the first questionnaire.

Data from the GP questionnaires included information on the involvement of a GP (no, yes) or a specialist team (no, yes) and the duration of the palliative period at home. The palliative period was defined as the last period of the patient‟s life during which all curative treatment had stopped and care and treatment served purely palliative purposes. Important themes in the questionnaire for relatives were relatives‟ characteristics and evaluation of the palliative pathway. The relatives were asked: „How, in your own words, was the entire period at home during which the deceased was dying compared with how you feel it should have been?‟ with the options: „Very well‟, „Well‟, „Fairly well‟, „Bad‟, „Very bad‟. The 65-item questionnaire to relatives was pilot-tested among 14 bereaved relatives not included in this study. Non-responders of relatives were sent one reminder four weeks after the first questionnaire.

105

Chapter 9. Paper 5

From the relatives we obtained data on their age (18-65, 66+), gender, relation to the deceased (non-spouse, spouse), living with the patient (no, yes), vocational educational level (3 years or less, > 3 years) and their evaluation of the palliative pathway (dichotomized into unsuccessful („Fairly well‟, „Bad‟, „Very bad‟) and successful („Very well‟, „Well‟)). We retrieved register data on the patient‟s age (18-65, 66+), gender, cancer diagnosis (lung, colorectal, breast, prostate, other) and place of death (institution (hospital and hospice), nursing home, home and other).

Analysis ‟A successful palliative pathway‟ was defined as the outcome measure and associations with place of death and nurse involvement were calculated.

Unadjusted and adjusted associations were calculated. Using robust variance estimates, the estimates were adjusted for clustering of patients of the same CN (13). Prevalence ratios (PRs) with 95% confidence intervals (95% CI) were used as a measure of association. Due to the high prevalence of the outcome measure (more than 20% „successful palliative pathways‟), odds ratios would overestimate the association (14;15). PRs were calculated with generalised linear models (GLM) with log link and the Bernoulli family, and when the model could not converge, we used the Poisson regression model (14;16).

The variables were assessed for collinearity (Pearson's correlation coefficient > 0.4) and multicollinearity (variance inflation factor < 10) (17;18). Due to collinearity, „Relatives living with the patient‟ and „CNs‟ age‟ was not included in the model. We added the duration of the palliative period spent at home as a confounder, since it could be associated with the CNs‟ possibility to provide palliative care. Data were analyzed using STATA 10 (19).

Ethics According to Scientific Committee for the County of Aarhus, the Biomedical Research Ethics Committee System Act does not apply here. The study was approved by the Danish Data Protection Agency and the Danish National Board of Health. 9.4. Results A total of 201 questionnaires from 129 CNs were filled in. For 178 cases, the CN or GP stated that there had been no home care in the palliative pathway or the patient were a nursing home resident

106

Chapter 9. Paper 5

from the beginning of the palliative pathway and therefore had no contact with CNs. For 220 cases the nurse did not respond, leaving a response rate of 47.7% (Figure 9.2). The 220 cases from nonresponding CNs were not statistically significantly different from the included cases in terms of the patients‟ gender and number of GP home-visits, but the non-responder cases tended to be significantly older and died more often at a nursing home (Table 9.1).

For 52 cases (25.9% of the 201 cases with nurse response), the CN or GP advised against sending the relative a questionnaire (Figure 9.2). A total of 101 relative questionnaires were filled in and two were excluded since the relative stated that the questionnaire was not relevant to the pathway at all, leaving 46 relatives non-responders (response rate 68.7%) (Figure 9.2). Figure 9.2. Flow-chart of questionnaires to bereaved relatives. Responders and non-responders

Flow-chart Questionnaire of bereaved relative 201 Community nurses (CNs) and general practitioners (GPs) were asked if the relative could receive a questionnaire 52 cases excluded by CN or GP for the following reasons: 10: Relative is mentally disabled 2: Relative is psychologically disabled 1: Relative is linguistically disabled 4: Relative is dead 2: Relative is too sick 1: Relative did not participate at all 32: No reason stated

149 Questionnaires sent to bereaved relatives 46 Non-responders

2 Excluded

5: Returned unanswered 1: Relative too sick to answer 1: Relative did not want to participate 37: Relative did not respond 2: Postal service did not find relative on address

2: Relative did not find palliative pathway relevant

101 Questionnaires filled in by bereaved relatives

Comparison of the 101 included cases with the 46 cases where the relatives did not respond showed that relatives answered the questionnaire significantly more frequently when the questionnaires concerned a male patient (p-value < 0.05) than when it concerned a female patient (Table 9.2).

107

Chapter 9. Paper 5

Comparing the 101 included cases with the 52 cases where CNs or GPs advised against sending the relative a questionnaire showed that cases included comprised more home-deaths, less institutional and nursing-home-deaths and were characterised by the CNs having more contact to the relatives (p-value < 0.05) (Table 9.2). Associations with evaluation of palliative pathway „A successful palliative pathway‟ was statistically significantly associated with home-death (2.3 (95% CI: 1.2;4.4)) (Table 9.3). „A successful palliative pathway‟ was also associated with nursing home-death compared with hospital-death (1.8 (95% CI: 0.9;3.7), even if the association fell short of significance. None of the variables concerning CN involvement were statistically significantly associated with a successful palliative pathway. 9.5. Discussion Main findings In a group of patients who died from cancer and had a palliative pathway at home, we found that the relatives‟ positive evaluation of the palliative pathway was associated with home-death and nursing-home-death. However, the latter association was not significant. Surprisingly, we identified no significant associations between this evaluations and the involvement of CNs. Strengths and limitations of the study The strengths of this study are its sampling and its comprehensive data collection. To eliminate differential misclassification, we used the standardised official health registers to identify the study population, including the places and the reasons of death. To minimise recall bias, the questionnaire was sent in January 2007 instead of awaiting the update of the Danish Register of Causes of Death in 2008.

The major weakness of these analyses was the selection bias. We found differences between included cases and the cases where the relatives had been excluded by CNs or GPs. The differences meant that we would tend to exclude those cases where the CN was not as involved and did not know the patient as well as in those cases that were included. In these excluded cases the relative may evaluate the pathway as less successful or tend to see the nurses involvement as less important. Thus, we would tend to overestimate the associations between a successful pathway and CN-related variables. This would also be the case if the CNs or the GPs had excluded cases that they knew had been unsuccessful (which they might have chosen to do for a

108

Chapter 9. Paper 5

number of reasons) despite an otherwise high CN level of involvement to assist the palliative pathway. As home-death is associated with CN involvement (8-10), such selection bias, therefore, tends to strengthen the association between home-death and a successful pathway in this study.

Approximately 1680 patients died from cancer in Aarhus County in 2006, and we included 599 cases recruited during a nine-month period (Figure 9.1). This lower amount of cases can be explained by the fact that we did not include persons less than 18 years, non-melanoma skin cancer, and the fact that we included only those with a cancer diagnose registered in a hospital in Aarhus County as the main diagnose of admittance in a 10-year period. Furthermore, the period where patients could die (1st March-30th November) was without the winter months of 2006, which may account for some of the missing cases, since winter months may have a higher average of deaths than the rest of the year. Comparison with existing literature We found that home-death was statistically significantly associated with the relatives‟ evaluation of the palliative pathway as being more successful compared with pathways ending with institutional death. In line with this, previous studies showed that home-death was associated with better bereavement response (20-22) and overall satisfaction with the palliative pathway (23). The fact that a successful palliative pathway was associated with home-death in this study may partly be attributable to selection bias, which, in turn, questions the use of home-death as a strong variable measuring a good death.

Some of the association between a successful pathway and home-death in this study may also be rooted in the fact that patients dying at a hospital or hospice often have worse symptoms and problems, e.g. pain, than patients dying at home, which, indeed, would affect the relatives‟ evaluation. We included no variables that could describe symptom and problem severity. However, patients with severe symptoms may also have more contact with a palliative specialist team, and adjusting for this we may have minimised some of the confounding from this factor. However, eliminating „specialist team involvement‟ from the model weakens the association between a successful pathway and home-death, but only slightly, which indicates that symptoms may be a confounder for which control was not fully achieved.

We also found that dying at a nursing home was associated with the relatives‟ evaluation of the palliative pathway compared with institutional death. But this association fell short of statistical significance in our study with only 101 included cases. The found could be attributable to the safe,

109

Chapter 9. Paper 5

but still homely environment created at a nursing home with possibility of 24-hour professional help. However, another reason for this result could again be that patients dying at a hospital or hospice presumably have worse symptoms than those who die at home and at nursing homes. Further research is needed to investigate the implications of death at home and at nursing homes for patients and relatives.

Quality in dying is much debated and there is no doubt that what makes a good death is determined by a complex interplay of many factors, e.g. personal and cultural values, supportive network characteristics, physical and medical factors and the services offered by the health care systems. Surprisingly, we found that none of the CN-related factors in our model were statistically significantly associated with the relatives‟ evaluation of the palliative pathway. To our knowledge, no previous studies have explored the association between relatives‟ or patients‟ evaluation of the entire palliative pathway and CN-related factors. However, studies show that CN providing frequent CN home-visits, visiting at night, knowing enough and spending enough time in the home in the palliative pathway are positively associated with bereaved relatives‟ satisfaction with the CN involved (24;25), but in our study no association with the relatives‟ evaluation of the palliative pathway in general was found. Our results highlight the need for further research on how to measure bereaved relatives‟ evaluation of palliative pathways and the factors concerning CNs‟ involvement affecting this evaluation. Implications for future research The elements of CN involvement included in our model showed no statistical significance with successfulness in bereaved relatives‟ evaluation of the palliative pathway. Further research into the constitutive elements of a well–performed palliative effort by the CN is therefore required. Furthermore, studies of the predictive power of a more active approach in primary health services in achieving a successful palliative pathway are needed. 9.6. Conclusion Our study indicates that dying at home is positively associated with an increase in the likelihood of that the bereaved relative will evaluate the palliative pathway as successful. Symptom severity may be an important confounder; one that was not adjusted for in the present study. No significant associations between the evaluations and way the CNs were involved could be identified. There is

110

Chapter 9. Paper 5

a need for studies exploring predictors of the primary care effort associated with a “good death” to improve and ensure more focus in the palliative primary health care effort. Competing interests The author(s) declare that they have no competing interests. Authors' contributions MAN participated in the design of the study, in the development of the questionnaires, handled the data, drafted the manuscript and performed the statistical analysis. PV helped to draft the manuscript and IS helped to perform the statistical analysis FO, ABJ and JS participated in the design of the study and in the development of the questionnaires. JS also helped to draft the manuscript. All authors read and approved the final manuscript. Acknowledgements Profound gratitude is extended to participating bereaved relatives, CNs and GPs. The study was funded by the Aarhus County Research Fund for the Clinical Development and Research in General Practice and across the Primary and Secondary Health Care Sectors (4-01-3-04), the Danish National Research Foundation for Primary Care (585-457808) and The Multipractice Study Committee (585-04/2072).

111

Chapter 9. Paper 5

9.7. Tables Table 9.1. Characteristics of 201 included cases and 220 cases not included because the community nurses (CNs) did not respond. Data on nurses obtained from CN questionnaires. Case data are obtained from formal health registers. Cases of community nurse responders

Cases of community nurse nonresponders

(N=201, 129

(N=220)

nurses) CN’s gender (n (%)) Male Female CN’s age (Mean (95 % CI)) Years as nurse (Mean (95 % CI)) Years as CN (Mean (95 % CI)) Number of questionnaires pr CN (Median (IQI))

-

0 (0.0) 129 (100.0) 45.3 (43.9;46.7)

-

19.9 (18.2;21.5)

-

10.7 (9.5;12.0)

-

1.4 (1.3;1.5)

-

CN’s extra education or courses in palliative care (n (%)) No

107 (84.9)

-

19 (15.1)

Yes Patient’s age at time of death (mean (95 % CI))*

70.0 (68.4;71.6)

73.2 (71.5;74.9)

111 (55.2)

136 (58.4)

90 (44.8)

97 (41.6)

Patient’s gender (n (%)) Male Female Primary cancer diagnosis (n (%)) Bronchus/lung

34 (16.9)

52 (22.3)

Colon/rectum

26 (12.9)

40 (17.2)

Breast

23 (11.4)

20 (8.6)

Prostate

31 (15.4)

28 (12.0)

Other

87 (43.4)

93 (39.9)

Home

97 (48.3)

79 (33.9)

Nursing home

35 (17.4)

69 (29.6)

Hospital/hospice

67 (33.3)

82 (35.2)

2 (1.0)

3 (1.3)

Place of death (n (%))*

Other (e.g. other institution)

* Statistically significantly different from 101 cases included in study with p-value < 0.05

112

Chapter 9. Paper 5

Table 9.2. Characteristics of 101 included cases, the 46 cases not included because the relative did not respond and the 52 cases where the general practitioners (GP) or community nurses (CN) advised against sending the relative a questionnaire. Case data in study stem from GP, CN and relative-questionnaires and from formal health registers. Case data of relative-non-responders and of the group where the GP advised against sending the relative a questionnaire stem from GP and CN questionnaires and formal health registers.

Patient’s age at time of death (mean (95% CI)) Patient’s gender (n (%)) Male Female Primary cancer diagnosis (n (%)) Bronchus/lung Colon/Rectum Breast Prostate Other Place of death (n (%)) Home Nursing home Hospital / Hospice Other(e.g. other institution) Relative’s age at time of filling in questionnaire (mean (95% CI)) Gender of relative (n (%)) Male Female Relative’s relation to deceased (n (%)) Spouse Girlfriend or boyfriend Daughter or son Sister or brother Parent Daughter-in-law Relative lived with patient (n (%)) No Yes Relative’s vocational education (n (%)) 3 years or less > 3 years

Cases in study (N= 101)

Cases of relative-nonresponders (N=46)

Cases where the GP or CN advised against sending the relative a questionnaire (N=52)

69.6 (67.3;71.9)

68.1 (64.8;71.4)

73.0 (69.9;76.2)

61 (60.4) 40 (39.6)

18 (39.1)* 28 (60.9)*

30 (57.7) 22 (42.3)

16 (15.8) 13 (12.9) 11 (10.9) 17 (16.8) 44 (43.6)

10 (21.7) 8 (17.4) 9 (19.6) 3 (6.5) 16 (34.8)

8 (15.4) 5 (9.6) 3 (5.8) 11 (21.2) 25 (48.1)

62 (61.4) 13 (12.9) 25 (24.8) 1 (1.0)

25 (54.4) 3 (6.5) 18 (39.1) 0 (0.0)

10 (19.2)* 18 (34.6)* 23 (44.2)* 1 (1.9)

58.3 (55.4; 61.2)

-

-

28 (27.7) 73 (72.3)

-

-

-

-

-

-

-

-

63 (62.4) 1 (1.0) 32 (31.7) 1 (1.0) 1 (1.0) 3 (3.0) 23 (23.7) 74 (76.3)

62 (63.3) 36 (36.7)

CN’s age (mean (95 % CI)

44.4 ( 42.7;46.2)

46.2 (43.8;48.6)

46.1 (43.9;48.4)

CN’s years as CN (n (%)) Five and less More than five

36 (40.0) 54 (60.0)

16 (40.0) 24 (60.0)

12 (25.0) 36 (75.0)

113

Chapter 9. Paper 5

CN’s extra education or courses in palliative care (n (%)) No 70 (82.4) 34 (85.0) Yes 15 (17.6) 6 (15.0) CN’s and community services’ homevisits pr day (n (%)) Less than one 14 (14.7) 5 (12.2) One or more 81 (85.3) 36 (87.8) CN’s knowledge prior to palliative period (n (%)) Poor 57 (60.6) 22 (51.2) Well 37 (39.4) 21 (48.8) CN’s contact with relatives (n (%)) No 3 (3.1) 2 (4.7) Yes 94 (96.9) 41 (95.3) GP involvement (n (%)) No 14 (15.2) 4 (9.3) Yes 78 (84.8) 39 (90.7) Specialist team involvement (n (%)) No 39 (55.7) 21 (58.3) Yes 31 (44.3) 15 (41.7) * Statistically significantly different from the 101 cases in study with p-value < 0.05. Not all sums of percentages are added to 100.0% because of round-offs.

39 (86.7) 6 (13.3)

9 (18.4) 40 (81.6)

29 (56.9) 22 (43.1) 15 (30.6)* 34 (69.4)* 10 (20.0) 40 (80.0) 21 (56.8) 16 (43.2)

114

Chapter 9. Paper 5

Table 9.3. Associations between a successful palliative pathway and model variables. A total of 101 cases were included in the analyses. The unadjusted and the adjusted prevalence ratios (PRs) are shown with 95% confidence intervals (95%CIs). CN is the Community nurse involved.

Gender of relative Male Female Age of relative 18 - 64 65+ Living with patient No Yes Relative’s relation to diseased Not spouse Spouse Relative’s vocational education 3 years or less > 3 years CN’s age Less than 40 40 and above CN’s years as CN Five and less More than five CN’s extra education or courses in palliative care (n (%)) No Yes CN’s knowledge prior to palliative period Poor Well CN’s contact with relatives No Yes CN’s and community services’ home-visits pr day Less than one One or more GP involvement No Yes Specialist team involvement No Yes Place of death Institution (Hospital or Hospice) Nursing home Home

Unadjusted prevalence ratio (95% CI)

Adjusted prevalence ratio (95% CI)

1 1.0 (0.8;1.4)

1 1.1 (0.7;1.7)

1 1.3 (1.0;1.7)

1 1.0 (0.8;1.6) Not included because of collinarity with „Relatives‟ relation to deceased‟

1 1.6 (1.0;2.6)

1 1.4 (1.0;2.0)

1 0.9 (0.6;1.5)

1 0.8 (0.6;1.1) 1 1.1 (0.8;1.4)

1 0.7 (0.3;1.3) Not included because of collinarity with „CN‟s years as CN‟

1 1.2 (0.9;1.5)

1 0.9 (0.6;1.3)

1 0.7 (0.5;1.2)

1 0.8 (0.4;1.7)

1 1.2 (0.9;1.5)

1 1.3 (0.9;1.8)

1 0.6 (0.6;0.7)

1 1.1 (0.7;1.8)

1 0.9 (0.6;1.2)

1 0.7 (0.4;1.1)

1 1.0 (0.7;1.3)

1 1.0 (0.5;1.7)

1 0.9 (0.7;1.3)

1 1.0 (0.7;1.4)

1 1.3 (0.7;2.5) 1.7 (1.1;2.7)

1 1.8 (0.9;3.7) 2.3 (1.2;4.4)

115

Chapter 9. Paper 5

9.8. References of Chapter 9 (1) Hanratty B. Palliative care provided by GPs: the carer's viewpoint. Br J Gen Pract 2000 Aug;50(457):653-4. (2) Thomas C. The place of death of cancer patients: can qualitative data add to known factors? Soc Sci Med 2005 Jun;60(11):2597-607. (3) Thomas C, Morris SM, Clark D. Place of death: preferences among cancer patients and their carers. Soc Sci Med 2004 Jun;58(12):2431-44. (4) Grande GE, Farquhar MC, Barclay SI, Todd CJ. Valued aspects of primary palliative care: content analysis of bereaved carers' descriptions. Br J Gen Pract 2004 Oct;54(507):772-8. (5) Neergaard MA, Olesen F, Jensen AB, Sondergaard J. Palliative care for cancer patients in a primary health care setting: Bereaved relatives' experience, a qualitative group interview study. BMC Palliat Care 2008 Jan 15;7(1):1. (6) Weibull A, Olesen F, Neergaard MA. Caregivers' active role in palliative home care - to encourage or to dissuade? A qualitative descriptive study. BMC Palliat Care 2008 Sep 16;7(1):15. (7) Brazil K, Bedard M, Krueger P, Abernathy T, Lohfeld L, Willison K. Service preferences among family caregivers of the terminally ill. J Palliat Med 2005 Feb;8(1):69-78. (8) Aabom B, Kragstrup J, Vondeling H, Bakketeig LS, Stovring H. Population-based study of place of death of patients with cancer: implications for GPs. Br J Gen Pract 2005 Sep;55(518):684-9. (9) Fukui S, Fukui N, Kawagoe H. Predictors of place of death for Japanese patients with advanced-stage malignant disease in home care settings: a nationwide survey. Cancer 2004 Jul 15;101(2):421-9. (10) Howat A, Veitch C, Cairns W. A retrospective review of place of death of palliative care patients in regional north Queensland. Palliat Med 2007 Jan;21(1):41-7. (11) The Danish National Board of Health. The Danish Register of Causes of Death [In Danish]. 2008. (12) Pedersen CB, Gotzsche H, Moller JO, Mortensen PB. The Danish Civil Registration System. A cohort of eight million persons. Dan Med Bull 2006 Nov;53(4):441-9. (13) Donner A, Klar N. Design and Analysis of Cluster Randomisation Trials in Health Research. 1 ed. London: Hodder Arnold; 2000. (14) Barros AJ, Hirakata VN. Alternatives for logistic regression in cross-sectional studies: an empirical comparison of models that directly estimate the prevalence ratio. BMC Med Res Methodol 2003 Oct 20;3(1):21. (15) Thompson ML, Myers JE, Kriebel D. Prevalence odds ratio or prevalence ratio in the analysis of cross sectional data: what is to be done? Occup Environ Med 1998;55(4):272-7.

116

Chapter 9. Paper 5

(16) Zou G. A modified poisson regression approach to prospective studies with binary data. Am J Epidemiol 2004 Apr 1;159(7):702-6. (17) Armitage P, Berry G, Matthews JNS. Statistical Methods in Medical Research. Fourth edition. 4 ed. Oxford: Blackwell Science; 2005. (18) O'Brien RM. A Caution Regarding Rules of Thumb for Variance Inflation Factors. Quality and Quantity 2007;41(5):673-90. (19) Stata Statistical Software: Release 9.0. College Station, TX: StataCorp LP; 2005. (20) Grande GE, Farquhar MC, Barclay SI, Todd CJ. Caregiver bereavement outcome: relationship with hospice at home, satisfaction with care, and home death. J Palliat Care 2004;20(2):69-77. (21) Akiyama A, Numata K, Mikami H. Factors enabling home death of the elderly in an institution specializing in home medical care: analysis of apprehension of the bereaved family. Geriatr Gerontol Int 2008 Jun;8(2):73-9. (22) van der Heide A, de Vogel-Voogt E, Visser AP, van der Rijt CC, van der Maas PJ. Dying at home or in an institution: perspectives of Dutch physicians and bereaved relatives. Support Care Cancer 2007 Dec;15(12):1413-21. (23) Ringdal GI, Jordhoy MS, Kaasa S. Family satisfaction with end-of-life care for cancer patients in a cluster randomized trial. J Pain Symptom Manage 2002 Jul;24(1):53-63. (24) Fakhoury W, McCarthy M, Addington-Hall J. Determinants of informal caregivers' satisfaction with services for dying cancer patients. Soc Sci Med 1996 Mar;42(5):721-31. (25) Lecouturier J, Jacoby A, Bradshaw C, Lovel T, Eccles M. Lay carers' satisfaction with community palliative care: results of a postal survey. South Tyneside MAAG Palliative Care Study Group. Palliat Med 1999 Jul;13(4):275-83.

117

118

Chapter 10. Discussion of methods

CHAPTER 10 DISCUSSION OF METHODS

119


In the following sections the methods of the studies are reviewed. Strengths and weakness of the chosen methods, selection of participants, design of questionnaires, analysis and validity in both the quantitative and qualitative studies are discussed, 10.1. Mixed method design Research-combining methods, multiple methods research, is the third major research paradigm, adding a complementary dimension to the quantitative and qualitative research methods (1). The literature is not unanimous in its use of the different terms in research-combining methods. Figure 10.1 explains how the terms are used in this thesis (2;3). Figure 10.1. Terms in Multiple methods research. (QUAL: Qualitative and QUAN: Quantitative)

Terms in Multiple methods research Multiple methods research

Multimethods research

Multimethods QUAL design

Multimethods QUAN design

Mixed methods research

Mixed model design

Mixed method design

Concurrent

Sequential

QUAN QUAL + QUAL

QUAN + QUAN

QUAL/QUAN mixed

QUAL + QUAN

QUAL

QUAL or

QUAN

In a mixed method design the strengths of an additional study can be used to overcome the weaknesses of another study (the principle of complementarity) and to provide stronger evidence for a conclusion through convergence and corroboration of findings (the principle of triangulation) (1). Hence, this design affords the opportunity of answering a broader and more complete range of

120


research questions and in a two phase sequential design, the result of the first phase can be used to develop and inform the purpose and design of the two.

In our study, the interviews with bereaved relatives identified the importance of certain themes which we explored in the interviews with professionals (e.g. shared care), and both interview studies added themes and questions to the questionnaire study and served to broaden its issues: - The issues of „transferring professional responsibility for the care when discharging a terminally ill patient‟, of „cooperation between professionals‟ and „supporting the relative in the pathway and in bereavement‟ were all themes emerged from the interviews, used in the questionnaires. - Some of the variables used in the analysis were directly taken from the interviews: e.g. „GPs making unplanned visits‟, „GP giving his/her telephone number‟ and „GP and CN knowledge before the palliative pathway‟.

In this way, the interviews added useful information to the questionnaires and analytic models in the surveys (4).

A huge disadvantage of mixed method is that it is more expensive, both in economical and human resources (time consuming). Furthermore, a team approach to mixed methods would be much preferred as it can be difficult for a single researcher to carry out both quantitative and qualitative research (1). Some even contend that one should always work with either a quantitative or a qualitative paradigm (1). However, the advantages of the mixed method approach in this study are described above, and the conclusion must be a thorough discussion of strengths and weaknesses before applying a mixed method approach to a study.

10.1.1. Choice of informants / participants Terminally ill cancer patients are the recipients of palliative care and their opinions are very important when describing and analysing associated factors in palliative pathways. In the initial planning phase of the study it was also contemplated to do interviews and to send questionnaires to terminally ill patients as well as their relatives. However, as the aims of the study took form we realised that we wanted to evaluate completed pathways where the last days of the patients‟ life were also included, since important information about the care provided the last days of the palliative phase otherwise may have been overlooked. With this in mind, we decided to use only bereaved relatives as spokespersons of the recipients of the care given. Furthermore, arranging

121


interviews or doing questionnaires with terminally ill patients can be difficult since the patients often weakens fast and finding the right time in their palliative period to interview them or send them a questionnaire is difficult (5). These practical difficulties were considered disproportionate to the added value of data from patients in our study.

According to the professionals participating in the studies, GPs and CNs were obvious participants because of our focus on the primary care professionals. We added hospital consultants as participants in the interviews since the discharge from hospitals to the patients‟ homes seemed to be a big issue in the palliative pathway at home and since the hospital doctors are important collaborators in palliative home care whenever the patient has an open admittance agreement with their regular hospital ward. 10.2. Qualitative study The aim of qualitative research is to describe or analyse the informant‟s perception and experience of the world and the phenomena and how we interpret own and others‟ experiences. The strength of the qualitative research design lies in its ability to broaden the framework of understanding and it is well-suited for answering “why”, “how” and “what” questions about human behaviour, motives, views and barriers (4). The qualitative approach was therefore suitable in answering research question A (Chapter 3).

10.2.1. The focus group interviews In our study with bereaved relatives, the use of focus group interviews (FGI) gave the informants a feeling of shared experience and of being able to support each other and it made new themes emerge. The group setting in the interviews with the professionals also raised new themes owing to the different experiences of the professionals (6). Furthermore, the professionals in the interviews learned from each other, e.g. when a hospital consultant voiced his surprise: „Would you (The GPs) really like me to call you before discharge of a terminally ill patient?‟ If we had chosen another qualitative method, e.g. individual interviews or participant observation, we would not have had the ability to benefit from this group interaction. However, in FGIs one has to consider the social control within the group which may hinder different individual aspects from emerging. If we wanted a more detailed and individual information about palliative care individual interviews would have been more useful. One must also consider the time limit in a PhD study, and gathering the informants in groups gave us an opportunity to obtain experiences from many more informants than we could have obtained doing an individual interview or a participant observational study.

122


Furthermore, we found that this method was the better method for qualifying generally applicable themes for our questionnaire study.

10.2.2. Selection of informants We invited a total of 18 relatives to participate in the interviews. The inclusion criteria are seen in Table 4.1 in Chapter 4. We examined the relatives‟ gender, age, relation to patient, patients‟ age, gender, diagnosis and time since death of patients and agreed that the 18 relatives proposed to us ensured a wide range of demographic characteristics and in time since death of the patient (3), and all 18 were invited by mail to participate in a FGI. Among these, 14 agreed to be interviewed and four relatives were not able to participate on the proposed dates, which gave us the problem that only three informants participated in the first interview. However, this did not influence the interaction in the group which was just as spirited as the two other groups. The inclusion criteria that the patient should have died less than one year before the inclusion of the relatives was used to minimise memory decay (7). When one tries to obtain a wide range of demographic characteristics it is also meant as a proxy to a wide range of experiences and views on the subject examined. However, we do not know if this was the case in our interviews. The results of the interview studies are almost certainly not exhaustive on the issues raised, but this does not diminish the importance of the themes emerged.

10.2.3. Analysis QD is often criticised for being too simple and for lacking rigour (8). However, QD is a welldescribed method of attaining data close description, i.e. a rich, straight description of an experience or an event in a language similar to the informants‟ own language. It entails lowinference interpretation compared with for example phenomenological or grounded theory studies and QD ties nicely with quantitative data and is useful for a mixed method inquiry (9). This was exactly what we aimed for, e.g. identifying themes for our questionnaire study and for possible improvements in palliative home care.

We tried to enhance rigour in the analysis by using on the following strategies: 1) authenticity, i.e. the focus in the study was on the participants‟ own experiences, 2) credibility, i.e. we reflected on how believable our results were, 3) criticality, i.e. we were critical of every decision made throughout the research process and 4) integrity, i.e. we had an on-going reflection and selfcriticism during the study (10). Using QD no theoretical strings are attached. This may make the analytical process somewhat subjective as descriptions will always depend on the researcher‟s perceptions, inclinations, sensitivities, and sensibilities (8). We therefore involved a group of

123


researchers in the analytic process to reduce the subjective element. Also being aware of the fact that the moderator (MAN) and the observers (JS, and FO) of the interviews were all GPs and being clear on our role as researchers during the interviews we tried to minimise the subjective element (11). However, one can never set oneself free of ones‟ pre-understanding. This may have introduced subjectivity, but it may also have been a strength that we knew the system from within and had tried to participate as GPs in palliative pathways ourselves.

10.2.4. Validity The validity terms in qualitative research are credibility, dependability, conformability and transferability, as described by Kvale (11).

10.2.4.1. Credibility The GPs and CNs participating were undoubtedly more interested in palliative care and in research than the average GPs and CNs. Our results from the FGIs may therefore not reflect the majority of the GPs‟ and CNs‟ willingness to perform palliative care in the patients‟ homes. On the other hand the informants were professionals who take an active role and know exactly where things can go wrong and where improvements are needed, and this may have allowed us to identify more relevant aspects concerning palliative home care than would have been possible with a representative GP and CN group.

Likewise, the participating hospital physicians may be members of a group of hospital physicians who are more committed to palliative home care than the average hospital physician. However, it was striking that only the oncologist had a written procedure at the hospital ward of what to do when discharging a terminally ill patient to the patient‟s home. We chose not to let GPs find potential bereaved relatives for our interviews because the GP‟s role was a major issue in the interviews with relatives. However, letting CNs and the palliative specialist team find potential informants could also have distorted the experiences of the participating relatives compared to the average relatives‟ experiences. Thus, the nurses could, unconsciously, have chosen relatives from pathways were the contact to CNs was good and the contact to GPs was not as good. From an ethical point of view, this procedure remained much better than finding relatives in a register because it minimizes the risk of inviting people who were not able or not ready to talk about their experiences.

124


Furthermore, the relatives who agreed to participate may be expected to be well-functioning and their experiences may therefore not represent the types of problems that weak relatives experience. On the other hand, relatives who want to participate in an interview may be those who have experienced major problems and need to point their criticism at someone.

To minimize the problems stated above, we chose the informants purposefully trying to ensure a wide range of demographic characteristics (3).

10.2.4.2. Dependability and conformability To ensure dependability, we searched for consistent statements regarding the issues that emerged in the interviews. As mentioned above, we sought to ensure conformability by researcher triangulation and by being conscious of our pre-understandings during the interviews and analysis.

10.2.4.3. Transferability Using qualitative data from FGIs does not provide „objective‟ data, but it provides a broad insight into barriers and facilitators of successful palliative home care. Not all issues and problems can be addressed in the interviews and not all issues are important to all palliative pathways. Furthermore, using the low-inference analytic approach, QD, reduces the ability to make generalizations. However, our rich descriptive summary yielded working hypotheses and key categories for future theory-based research, just as it provided us with appropriate results for the development of the questionnaires. 10.3. Quantitative surveys The aim of quantitative research is to obtain an overview of the extent of and mapping of a phenomenon and with the deductive approach quantitative research is well-suited for asking “when”, “how much” and “how many” questions and is therefore suitable for problem quantification and testing of theories, interventions and new treatments (4). The quantitative approach was therefore suitable in answering research question B (Chapter 3).

10.3.1. Study design The quantitative study was a combined questionnaire and cross-sectional register study with retrospective data which was followed by analysis of associations.

The nature of our survey does not allow us to make causal inferences but merely to suggest associations between outcomes and included variables and hence to raise hypotheses about

125


predictors. Predictors would have been found in a follow-up or randomised intervention design. However, a study design like ours has advantages in relation to economy and duration of the study.

10.3.2. Questionnaires Since no validated questionnaires that met our aims were available (see Appendix F) (12;13) we constructed ad hoc questions for all three questionnaires. The phrasing of questions about the patient-GP relation and questions considering demographic data were inspired by existing questionnaires (14-18). However, in our study the relatives were to answer the questions and instead of the patients, which forced us to change some of the phrasings and the validation of the existing questionnaires can not be transferred to our questionnaires. As to the important issues of the relatives‟ evaluation of the palliative pathway, which we used in Papers 4 and 5 as the main outcome, it would, indeed, have strengthened the study if a validated questionnaire or scale could have been used.

We did not include questions about the patients‟ socioeconomic status other than marital status, having children, educational level and job status. We could have included e.g. household income and type of housing (e.g. rented / owned, flat / house / farm and so on). However, the questionnaires were already very long and we had to prioritize the questions included since too many questions may have reduced the response rates (19).

We also omitted questions about symptom severity since all validated questionnaires that ask respondents to rate symptoms concern present symptoms (e.g. EORTC QLQ-C15-PAL) (20) and have not been validated in retrospective studies. Furthermore, symptom control has not seen to effect bereaved relatives‟ evaluation of the GP and CN (21) and their evaluation of their GP is not affected by the degree of symptom control obtained (22). In retrospect, we should have included questions about symptoms of some kind since this factor might be an important confounder in our studies (See 10.3.5.9).

10.3.3. Variables in the models

In the models of how variables were associated with the main outcomes, we drew on hypotheses about each variable as presented in the literature and through interviews and our own experience. In Paper 3 the overall hypothesis was that the terminally ill cancer patients wish to die at home. The reasons for including the variables in the model of associations with home death (Table 7.2, Chapter 7) are shown in Table 10.1.

126


Table 10.1. Variables included in the model of associations with home death in Paper 3. Variable „Age and gender of patient‟

„Primary cancer diagnose‟

Patient‟s marital status

Having children

GP knowledge prior to palliative period

Home visits by GP

Unplanned home visits by GP

Literature, interview results and hypotheses prior to study In previous studies, gender has been shown to be of importance in univariate analysis (23) Hypothesis: Older people have a stronger wish to stay at home than younger patients during the terminal stages of illness. However, one could also have argued that older patients have more comorbidity and therefore have a higher change of being admitted to hospital during their last days of life. Men have wives to take care of them and it is therefore likely that more men than women will die at home. Different cancer diagnoses have been shown to have an effect on place of death (24) Hypothesis: Different cancer diagnoses give different palliative pathways because problems and symptoms vary Previous studies have shown that patient‟s marital status is of importance (23-26) Hypothesis: Patients having a relative at home 24 hours a day will be more likely to die at home than patients who have no relative at home Hypothesis: Patients having adult children have a relative who can take care of them, which may give them a better possibility of dying at home than patients who have no adult children, whereas having small children living at home may complicate the palliative pathway at home and cause admittance during the last days. Continuity is known to be of importance to patients and relatives (27) and to reduce hospital deaths (28) Hypothesis: If the GP knew the patient well from before the palliative pathway began the possibility of dying at home would be higher than if this was not the case because of the safety this knowledge would give the patient and thefamily and because the GP could be more involved. Home visits are shown to be of importance in achieving home death (23;29;30) Hypothesis: GP home visits enhance home death because treatment may take place in the patient‟s home and because of the safety it gives the patient and his/her family. In our interviews many of the bereaved relatives mentioned unplanned home visits by the GP as something that had made a difference in the palliative pathway. Hypothesis: Extra involvement of the GP may enhance home death because treatment may take place in the patient‟s home and because of the safety it gives the patient and his/her family.

127


GP gave private number to patient to use in out-of-office hours

GP had made a plan with the patient for whom to contact in out-of-office hours‟ GP had contact with relatives‟ Community nurse involvement‟ Specialist team involvement‟

Duration of palliative period at home‟

In our interviews many of the bereaved relatives and CNs mentioned that whenever the GP gave a private phone number to patients that had made a difference in the palliative pathway. It was also found in a previous study that 24-hours back-up is important to the relatives (31). Hypothesis: The feeling of being able to reach the GP 24-hours a day may enhance home death because treatment may take place in the patient‟s home and because of the safety it gives the patient and his/her family. As above

Hypothesis: Whenever the GP collaborate with the relatives around the palliative pathway, the possibility of home death may be strengthened. Hypothesis: The CN is often involved together with the GP, and CN involvement could be an important confounder in achieving home death. Hypothesis: The specialist team is often involved together with the GP, and their involvement could be an important confounder in the study. The direction of this confounder‟s influence was difficult to predict since the specialist team would more often be involved in pathways with severe symptoms and problems which may lead to a higher possibility of institutional death, but on the other hand, the specialists may have enhance a higher proportion of home deaths in the pathways they are involved in than other pathways. Hypothesis: The longer the palliative pathway is at home the more time is there to plan a home death.

The reasons for including the variables in Paper 4 on GP involvement and many of the confounders are the same as in Paper 3 (Table 10.1.). Our hypothesis was that relatives wish to fulfil the patient‟s wish for home death and that their evaluation of the palliative pathway may be more positive in the presence than in the absence of one or more of the factors facilitating home death. The variables in the model in Paper 4 (Table 8.3, Chapter 8) different from the variables in the model in Paper 3 are shown in Table 10.2.

The reasons for including many of the variables in the model in Paper 5 are the same as those in Table 10.2. The variables in the model in Paper 5 (Table 9.3, Chapter 9) different from the variables in the model in Paper 4 are shown in Table 10.3.

128


Table 10.2. Variables included in the model of associations with a successful palliative pathway evaluated by the bereaved relatives concerning GP involvement in Paper 4. Variable Gender of relative

Age of relative

Relative living with patient

Relative‟s relation to diseased

Relative‟s education

Place of death

Literature, interview results and hypotheses prior to study Hypothesis: If the relative is a man, he may have more difficulties coping with the difficult situation and a lower possibility of evaluating the pathway as successful as a woman Younger relatives are statistically significantly more prone to feel the burden of caring for a dying person than older relatives (32) Hypothesis: If the relative is young, he or she may have more difficulties coping with the difficult situation and a lower possibility of evaluation the pathway as successful as an old relative Living with a relative has been shown to have an impact on place of death (24). Hypothesis: If the relative is living with the patient, he or she would be at home 24 hours a day which may give the patient a better possibility of caring for the patient and a higher possibility of evaluation the pathway as successful than if the relative were not living at home with the patient The relation with the relative has been shown to have an impact on place of death (24). Hypothesis: If the relative is a spouse, he or she would be at home 24 hours a day, which may improve the possibility that the pathway will be evaluated as successful than if the relative were not a spouse Hypothesis: If the relative has a higher education, he or she would also be better at coping with the difficult situation than if the relative had no higher education and this, in turn, would increase the likelihood of a positive evaluation of the pathway Bereaved relatives are seen to be more satisfied with the care pathway when home death is accomplished than when it is not. Relatives wish to fulfil the patient‟s wish of home death (31;33;34) Hypothesis: If home death is achieved, the relative would be more satisfied with the palliative pathway than if it was not achieved

Table 10.3. Variables included in the model of associations with a successful palliative pathway evaluated by the bereaved relatives concerning CN involvement in Paper 5. Variable Nurse‟s age Nurse‟s years as community nurse Extra education or courses in palliative care (n (%) Nurse‟s knowledge prior to palliative period

Literature, interview results and hypotheses prior to study Hypothesis: The more life experience and professional experience a nurse has, the better her effort in palliative care will be and the more probable it is that the palliative pathway will be successful Hypothesis: The more experience a nurse has as a CN, the better her effort in palliative care will be and the more probable it is that the palliative pathway will be successful Hypothesis: The more knowledge and education a nurse has in the palliative field, the better her effort in palliative care will be and the more probable it is that the palliative pathway will be successful Continuity has been shown to be of importance to patients and relatives (27). Hypothesis: If the CN knew the patient well from before the palliative

129


Nurse‟s contact with relatives Nurse‟s or home care services‟ home visits pr day

GP involvement

pathway, the possibility of dying at home would be higher because of the safety this knowledge would give the patient and the family and because the CN may be more involved than if this was not the case. Hypothesis: When the CN collaborates with the relatives around the palliative course the possibility rises that home death may be achieved. Involvement of CNs is shown to be of importance in achieving home death (23;35;36). Hypothesis: CN home visits enhance home death because treatment may be given in the patient‟s home and because of the safety it gives the patient and the family. Hypothesis: The GP is often involved together with the CN, and GP involvement could be an important confounder in achieving a successful palliative pathway.

10.3.4. Response rates Response rates of 63.2% (GPs in Papers 3 and 4), 68.6% (relatives in Paper 4), 47.7% (CNs in Paper 5), 68.7% (relatives in Paper 5) must be considered as good, especially considering the relatives, where similar studies with bereaved relatives and postal questionnaires or scales all had response rates between 28 and 59% (22;37-45) (See Appendix F). However, as we asked the GP and CN to inform us if the relative would be able to receive a questionnaire, this probably have excluded relatives who were too weak, in any way, to fill out a questionnaire, leading to a higher response rate concerning the relatives. The impact of non-response on the survey is discussed later in this chapter.

10.3.5. Internal validity In every survey biases occur, and the internal validity of a study depends upon the severity of these biases (46;47). The following issues and possible biases of this survey will be discussed: Validity of data from registers, sampling errors, selection bias because of non-responders, selection bias because of cases excluded by GPs and CNs, social desirability bias, recall biases and confounding.

10.3.5.1. Validity of data from registers We used several official registers to sample our population of deceased cancer patients (the county hospital discharge registry (HDR), The Civil Registration System database, the regional health authority‟s registry and The Danish Register of Causes of Death (RCD)) and we used several variables from the registers in our analysis (e.g. patient demographics, place of death, GP home visits) (The Civil Registration System database, The Danish Register of Causes of Death (RCD) and The Registry of Health Care Providers) as described in Chapter 4. The validity of our

130


survey firmly lies with the validity of the register data. The validity of some of the registers has been examined.

We used the HDR to locate cancer patients in the former Aarhus County. This database has many advantages, as data were readily available, costless and data could be linked to other register data by means of unique civil registration numbers (48). The completeness of the HDR for e.g. haematological and ovarian malignancies has been found to be 91.5% and 96%, respectively, compared with data from the Danish Cancer Registry (49;50) but others have found a more severe incompleteness in diagnosis registrations (51). Hence, we must bear in mind that we may not have captured the entire number of cancer patients in the given period, which may account for some of the missing cases (See 10.3.5.2.) We were not able to use the Danish Cancer Registry in this study as the registry was not sufficiently updated at the time of the study.

The Civil Registration System database was used to locate deceased patients out of a studydatabase of more than 29 000 cancer patients and their relatives (52), and the regional health authority‟s registry was used to locate their GPs. We have no reason to doubt the validity of The Civil Registration System database, and in the regional health authority‟s registry we were able to locate GPs to all patients registered with a GP. However, this database was not fully updated since several of the stated GPs had either retired, were dead or had left the practice. Also, we learned from 15 GPs that the patient in question had changed their GP either before or at the beginning of the palliative period. We then sent a questionnaire to the new GP instead. The RCD was used as a “golden standard” of cancer deaths and to locate data of place of death. The RCD has since 1943 been classifying and coding the diagnoses on all Danish death certificates according to the international classification of diseases, the ICD. Diagnoses on death certificates are not the results of a scientific process with formal and uniform diagnostic criteria and strict requirements for documentation of the cause of death, but is the opinion of the doctor who fill in the death certificate. Maybe most important, the RCD is not updated until approximately two years after a citizens‟ death and therefore data was not available in our survey to locate patients who had died from cancer. Instead, we had to locate the study-database by merging registers as described in Chapter 4 and then, after the survey had taken place, we used data from the RCD as the “golden standard”. In 17 cases, the GP had filled in questionnaires, stating that the patient died from cancer, but these patients were not registered as such in the RCD and were therefore not part of our final study-database. Furthermore, in 25 cases of the cases where the RCD had registered a

131


cancer death, the GP had not filled in the questionnaire and had stating that the patient did not die from cancer. These case were categorised as „non-responding GP‟.

Furthermore, on the death certificate until 2005 hospice deaths have been registered as death at nursing homes. In the 2006 RCD data, hospice deaths are registered as hospital deaths, but from 2007 it should be possible to examine hospice deaths alone. The reason for this confusion about registration of hospice deaths is apparently the transition to electronic death certificates and a new national database to capture data from these electronically reported data. However, the validity of the reported hospice deaths in 2006 seems uncertain. The above stated arguments question the validity of the RCD and a validation study of both diagnosis and place of death in the RCD would be highly relevant.

We used the Registry of Health Care Providers to locate GP home visits in the palliative period. The registry provides data on the GPs‟ payment of services provided, and the date of this payment. This gave us a problem since we wanted to include all home visits provided by the GPs during a period from three months prior to death until death. Since the payment for services falls due at the earliest the following month, we enclosed all home visits for which payment had been made for two months prior to the patient‟s death and forward.

10.3.5.2. Sampling errors In 2006, 1680 patients died from cancer in Aarhus County (53), but we included only 599 during a nine-month period where we ought to have included approximately 1260 cases. We excluded patients younger than 18 years old and patients with non-melanoma skin cancers which probably account for some difference in the number of cases.

In search of the reasons for the rest of the missing cases the following issues must be considered: We included all patients in the HDR in a 10-year period who were admitted to hospital or had an ambulant control or treatment with a cancer diagnosis (excl non-melanoma skin cancer) as their primary diagnosis (see Figure 4.1., Chapter 4). This procedure may have excluded the following cases: Patients who had had a cancer more than 10 years ago, and who had not been admitted since then Patients who had immigrated to Aarhus County after the registration of their diagnosis and treated in another county before the terminal phase of the disease

132


Cancer patients who had been admitted because of another diagnosis than cancer (i.e. pneumonia, respiratory distress, acute abdomen, etc) and who had either died at hospital or had been discharged without cancer treatment. In this case the cancer, if registered at all, would be registered as a secondary diagnosis Patients who had been admitted to hospitals and died at the hospital without a cancer diagnosis, but subsequently the autopsy revealed that they had a disseminated cancer.

Also season variation has to be considered. Our 9-months study-period excluded the winter months (December – February), and if any season variation in cancer deaths exists, one could imagine that cancer patients would die more frequently in wintertime than in the rest of the year. The monthly distribution of deaths in our study-database is shown in Figure 10.2. It is seen that 96 patients died in March with is far above the average of 67 deaths pr. month in our study-database. If a high amount of patients died in December, January and February 2006, we would have less than expected deaths in the 9-months period we were looking at. Figure 10.2. Monthly distribution of deaths in our study-database.

133


Another fact is that deaths take a few days to register in The Civil Registration System database, and there are no deaths in the last three days of the period (28-30 November 2006), which may account for 5-10 missing cases in late November.

We have no reason to believe that the above mentioned cases, the cases with death in the winter months and in the end of November would have altered the results since these patient do not represent a special group of patients in concern of palliative home care. In this case, the bias is a non-differentiated misclassification (information bias), which will reduce the power in our study but not bias the estimates.

However, the missing cases‟ impact on the results in the survey may be further elaborated from an examination of the places of death of the 599 included cases. It appears from Table 10.4. that there were more home deaths and nursing home deaths and fewer hospital deaths in the studydatabase than should be expected from data on places of deaths in 2005 in Denmark. We would therefore expect that among the missing cases there would have been more hospital deaths than among the included cases, thus opening for bias due to too many home deaths in our studydatabase. Table 10.4. Places of death in our study-database and for all cancer deaths in Denmark in 2005. Expected percentage distribution of place of death in Aarhus county (53) Place of death

Our study-database (2006)

Denmark 2005**

% n % Home 195 32.6 25.8 Nursing home and hospice* 17.9 Nursing home 139 23.2 Hospital 55.1 Hospital and hospice* 258 43.1 Other Place 7 1.2 0.8 Not registered 0 0 0.5 In total 599 100.1 100.1 * In 2005 deaths at hospices were registered as deaths at nursing homes, but in 2006 as deaths at hospitals. ** Data on places of deaths of all cancer deaths in DK in 2006 are not yet available from the RCD. If the excluded cases, which must have a high percentage of hospital deaths, have had a palliative period at home with GP visits we would tend to overestimate the association between home deaths and GP home visits. On the other hand, if the excluded cases with a higher percentage of hospital deaths did not have GP visits at home due to at short period at home we would tend to underestimate the association between home deaths and GP home visits. Likewise, concerning

134


the relatives‟ evaluation of the palliative pathway, we have no opportunity in our study-database to figure out which way the sample errors would alter the result as we do not know if the relatives to the missing cases were more or less satisfied than the average.

Only new data on the 9-month period in Aarhus County extracted from the updated RCD combined with data on hospital admittances and GP services may help interpret to which extent the missing data alter our results.

10.3.5.3. Selection bias because of non-responding GPs Selection biases are distortions that result from procedures used to select subjects and from factors that influence study participation (47).

In 194 cases out of 599 questionnaires sent, the GP were classified as non-responders and the stated reasons for their lack of response are given in Figure 4.2., Chapter 4.

As seen in Figure 4.2., in many of the cases, we do not know their genuine reasons for declining participation, but may offer some possible explanations: The GP never received the questionnaire (e.g. the postal services failed, GP had either died, was sick or on leave, had retired), the GP did not find the pathway relevant for the survey (e.g. the GP did not think that the patient died from cancer or there was no palliative care at home) or the GP did not have the time to fill in the questionnaire.

When interpreting the results it is important to consider if some GPs chose not to answer because they felt that they had been involved in an unsuccessful palliative pathway or that they did not succeed in achieving home death. The association between a positive evaluation of the palliative pathway and the GP‟s effort would have been overestimated in this case.

However, in Paper 3 (Table 7.1, Chapter 7) and Paper 4 (Table 8.1, Chapter 8), we found no statistically significant differences between included and non-included cases regarding the patient‟s gender, place of death and number of GP home visits. Only the patient‟s age was showed with significant differences between the two groups indicating that selection bias must be limited as far as GP non-responders is concerned.

135


10.3.5.4. Selection bias because of non on-responding CNs In 220 cases out of 527 questionnaires sent, the CNs were classified as non-responders and the reasons are given in Figure 4.2., Chapter 4.

As seen in Figure 4.2., in many of the cases, we do not know the genuine reasons for their decision to decline participation, but may offer some possible explanations: the CN never received the questionnaire (e.g. the postal services failed, the CN had either died, was sick or on leave, had retired or the head of the nursing centre did not pass on the questionnaire to the nurse), the CN did not find the pathway relevant for the survey (e.g. the CN did not think that the patient died from cancer or there was no palliative care at home) or the CN did not have the time to fill in the questionnaire. The low response rate of CNs may also in part be due to the fact that the home care services are not used to participating in research and the fact that 1st January 2007 the new regions in Denmark were a reality with new community boarders. This resulted in re-organising of many community nursing centres and many CNs reported that they were even busier in the beginning of 2007 than usually because of administrative tasks.

Important when interpreting the results on CN participation is if some CNs declined to answer because they felt that they had participated in an unsuccessful palliative pathway or that they did not succeed in achieving home death. In this case, we would have overestimated the association between the positive evaluation of the palliative pathway or home death, respectively, and the CN‟s effort.

In Paper 5 (Table 9.1, Chapter 9) we found no statistically significant differences between included and non-included cases in terms of the patient‟s age, gender, diagnosis and place of death indicating that selection bias must be limited as far as the CN non-responders is concerned.

10.3.5.5. Selection bias because of non-responding relatives In 80 cases out of 268 questionnaires sent, the relatives were classified as non-responders and the reasons are given in Figure 4.2, Chapter 4.

As seen in Figure 4.2., in many of the cases, we do not know the genuine reasons why they decided not to participate, but their reasons could be: the relative were sick or had died, did not have the energy, or time or the postal services did not find relative at the given address.

136


If the main reason why they declined invitation was because the palliative pathway had been either unsuccessful or successful, this may have altered the associations, but the direction and importance of this bias is difficult to establish. Furthermore, the professionals‟ efforts do not necessarily amount to successfulness of the palliative pathway.

In Paper 4 (Table 8.2., Chapter 8) and Paper 5 (Table 9.2., Chapter 9), we found no relevant differences between included and non-included cases. Only statistically significant difference was found in terms of the patients‟ gender: the relatives of male patients tended to answer the questionnaire statistically significantly more frequently than the relatives of female patients. The overall similarity of these two groups indicates that selection bias must be limited as far as bereaved relative non-responders are concerned.

10.3.5.6. Selection bias because of relatives excluded by GPs or CNs In 121 cases out of 389 cases the GP or CN advised against sending the relative a questionnaire. The reasons given can be seen in Table 4.2., Chapter 4, but in 82 cases no reasons were stated.

In Paper 4 we found many differences between included cases and the cases whose relatives were excluded by GPs (e.g. cases included in the study had statistically significantly more home deaths, more involvement and services by GP, CNs and a palliative specialist team) (Tables 8.2, Chapter 8). If GPs excluded cases they knew had been unsuccessful (which they might have done for several reasons) despite a high level of involvement and efforts to support the palliative pathway we may have overestimated the association between a successful pathway and GPs services. Furthermore, since the proportion of home deaths was smaller among the excluded than among the included cases, the association between a successful pathway and home death tends to be overestimated.

In Paper 5 we found many differences between included cases and the cases whose relatives were excluded by GPs or CNs (e.g. the cases included had statistically significantly more home deaths, less institutional and nursing-home deaths and were characterised by the CNs having more contact to the relatives) (Table 9.2, Chapter 9). This indicates that we may have overestimated the associations between a successful pathway and CNs services. However, we do not know if the relatives of excluded cases would evaluate the palliative pathway as successful or unsuccessful. Furthermore, the proportion of home deaths was smaller among excluded than among included cases, and the association between a successful pathway and home death therefore tends to be overestimated, also in this survey.

137


10.3.5.7. Social desirability bias (Information bias) It may be a problem in our survey that the relatives do not want to criticise the GP or the CN because they know them and some are receiving care or treatment themselves. To minimize this social desirability bias, we explicitly explained the anonymous nature of the study in the cover letter to relatives: “The filled in questionnaires will be anonymised so that GPs, CNs and other health care professionals will not be able to see, what you have answered. Only the research group stated below will be able to see the individual response.” We believe that this phrasing minimised this bias.

11.3.5.8. Recall bias (Information bias) To minimise recall bias, the questionnaires were sent in the beginning of 2007 instead of waiting till late 2008 when the Danish Register of Causes of Death would have been updated on deaths in 2006. All Danish GPs can consult electronic patient files when completing a questionnaire (54) and CNs have their own journals to consult. However, these journals does not include all information requested for in the questionnaires and not remembering the specific palliative pathway may account for some of the non-responding GPs and CNs, but the questionnaires which were filled in were to a surprisingly extent filled in entirely, which indicate that the professionals remember these pathways well. The last reminder to relatives was sent in July 2007, which, in the worst case would be 16 months after the patient‟s death but research has shown that an event like loosing a loved one is likely to be well recalled (55). All in all, recall bias among participants is not considered a great problem in the present survey.

10.3.5.9. Confounding A confounder is a factor that is associated with both the exposure and the outcome factor. It is associated with, but not a consequence of the exposure. Consequently, confounding results from a mix of different influential factors (47). It can either be reduced in study design or in analysis. Randomisation, restriction and matching are means of controlling for both known and unknown confounders, but do not apply to the method approach of this study. Instead, we used statistical methods to adjust for the influence of known confounders on the estimates in the multivariate analysis (56). The known possible confounders included were drawn from literature studies, information from preceding interviews and the clinical and personal experiences of the research group (See Tabel 10.2-4, Chapter 10).

138


In Paper 3, we adjusted for the patient‟s gender, age, marital status, diagnosis, CN and for specialist team involvement and the duration of the period spent at home during the palliative period. In Paper 4, we adjusted for the relative‟s age, gender, educational level, relation to deceased, relative living with patient, and for CN and specialist team involvement and the duration of the period spent at home during the palliative period. In Paper 5, we adjusted for the relative‟s age, gender, educational level, relation to deceased, relative living with patient, CNs age, years as CN and extra education or pathways in palliative care, GP and specialist team involvement and the duration of the period spent at home during the palliative period.

However, we did not adjust for symptom severity which may have been an important confounder as indicated in several results in the surveys. The positive association between bereaved relatives‟ positive evaluation of pathways and home death in Papers 4 and 5 may therefore partly have been obtained because patients dying at a hospital or hospice may be suffering worse symptoms, e.g. pain, than patients dying at home, which, indeed, would affect the relatives‟ evaluation. However, patients with severe symptoms may also have more contact with a palliative specialist team, and by adjusting for this, we may have adjusted for some of the confounding rooted in symptom severity. If symptom severity is a significant confounder and the variable „specialist team involvement‟ is also adjusting for symptom severity, limiting this variable from the model would weaken the association between a successful pathway and home death markedly. However, this is not the case. Further, one may also argue against the hypothesis that the variable „specialist team involvement‟ also adjust for symptom severity because it is more likely that home death may be accomplished is if a team is involved than if it is not, and this will positively affect the relatives‟ evaluation of the palliative pathway. Looking back, we probably should have included a variable in our models specifically addressing symptom severity.

10.3.6. Generalisability Optimal generalisability would imply that findings from this study could be extrapolated to all patients in basic level palliative home care, nationally and internationally.

However, the sampling errors mentioned above with a higher percentage of home-deaths in our study-database compared with the average percentage of home-death in Denmark question the generalisability of our survey, but we have no opportunity from the available data of this thesis to figure out which way the sample errors would alter the results.

139


It is possible that we have selected a special group of GPs, CNs and bereaved relatives for participation in our survey as described earlier in this chapter. Concerning the selection of GPs and CNs, only minor differences were demonstrated between cases included and those not included. The selection of bereaved relatives is slightly more problematic as several significant differences were demonstrated between the included cases and the cases excluded by the GP or CN.

The above mentioned biases may alter the results in both ways, and we still consider the findings generalisable to other Danish patients in basic level palliative home care in spite of the different biases in the study.

The generalisability of the survey to the Danish national context of palliative care are strengthened by the sampling of patients who died from cancer and had some palliative period at home regardless of the involvement of specialised teams or hospital records. Furthermore, Denmark is quite homogenous with respect to the organisation of primary care and social demography and Aarhus County was a county of both rural and urban citizens, and are considered as being quite representative of the Danish population.

All in all, we consider that the results may be largely generalised to Denmark as a whole. Extrapolation to other countries, however, requires due consideration of differences in health care systems (e.g. publicly financed, organising of palliative care) and cultural factors.

10.4. Ethical considerations In this study the deceased patient is the centre of care and thus the quality of care provided in the palliative pathway. However, doing a retrospective study after-death we have no opportunity to ask the patient of participation. Since we focus mainly on the palliative pathway and not on the patients themselves, we consider the study ethically justifiable concerning the patients included in the study.

In relation to contacting bereaved relatives, one may object that the family who has been burdened with care giving may feel overwhelmed, may be in acute grief, or may find the requirements of participation too much (57). However, findings from other studies suggest that it is pertinent to invite family caregivers to be involved in palliative care research and that many participants actually derive benefits (58). This is in line with the fact that many of the participating relatives in our survey wrote on the questionnaire, that if they could be of any other help to us, they could be contacted again and the participating relatives in the interviews said the same thing. The

140


willingness to help others in the same difficult situation is important to remember when one meets reluctance against doing research with participation of newly bereaved relatives. However, ethical issues should always be considered when planning the study design and we asked the GPs and CNs if we could contact the bereaved relatives for this reason.

10.5. References of Chapter 10 (1) Johnson B, Christenen L. Chapter 14. Mixed Research: Mixed Method and Mixed Model Research. Educational Research Quantitative, Qualitative, and Mixed Approaches. 2 ed. Sage; 2007. (2) Stange KC, Gotler RS. Mixed methods and diverse perspectives. Ann Fam Med 2006 Jul;4(4):290-1. (3) Crabtree BF, Miller WL. Doing qualitative research. Newbury Park: Sage Publications; 1992. (4) Malterud K. Shared understanding of the qualitative research process. Guidelines for the medical researcher. Fam Pract 1993;10(2):201-6. (5) Kaasa S, De CF. Palliative care research. Eur J Cancer 2001 Oct;37 Suppl 8:S153-S159. (6) Kitzinger J. Qualitative research. Introducing focus groups. BMJ 1995;311(7000):299-302. (7) Hinton J. How reliable are relatives' retrospective reports of terminal illness? patients' and relatives' accounts compared. Soc Sci Med 1996;43(8):1229-36. (8) Sandelowski M. Whatever happened to qualitative description? Res Nurs Health 2000 Aug;23(4):334-40. (9) Sullivan-Bolyai S, Bova C, Harper D. Developing and refining interventions in persons with health disparities: the use of qualitative description. Nurs Outlook 2005 May;53(3):127-33. (10) Milne J, Oberle K. Enhancing rigor in qualitative description: a case study. J Wound Ostomy Continence Nurs 2005 Nov;32(6):413-20. (11) Kvale S. Interviews: An Introduction to Qualitative Research Interviewing. 1 ed. Thousand Oaks: SAGE Publications; 1996. (12) Aspinal F, Addington-Hall J, Hughes R, Higginson IJ. Using satisfaction to measure the quality of palliative care: a review of the literature. J Adv Nurs 2003 May;42(4):324-39. (13) Mayland C, Williams E, Ellershaw J. How well do current instruments using bereaved relatives' views evaluate care for dying patients? Palliat Med 2008 Mar;22(2):133-44. (14) Heje HN. Patient evaluation in general practice. Methodological aspects, influence of patient and GP characteristics and the GPs' experiences with the evaluations [thesis]. 1 ed. Aarhus: Research Unit and Department of General Practice, Faculty of Health Sciences, University of Aarhus; 2006.

141


(15) Vedsted P, Sokolowski I, Heje HN. Data quality and confirmatory factor analysis of the Danish EUROPEP questionnaire on patient evaluation of general practice. Scand J Prim Health Care 2008;26(3):174-80. (16) Rasmussen NK. Modelspørgeskema til undersøgelse af sundhed og sygelighed i befolkningen [In Danish]. København: DIKE; 1994. (17) Hansen RP, Olesen F, Sorensen HT, Sokolowski I, Sondergaard J. Socioeconomic patient characteristics predict delay in cancer diagnosis: a Danish cohort study. BMC Health Serv Res 2008;8:49. (18) Mikkelsen TH, Søndergaard J, Sokolowski I, Jensen AB, Olesen F. Cancer survivors' rehabilitation needs in a primary healthcare context. submitted 2008. (19) Leungh WC. How to design a questionnaire. Student BMJ 2001;9:143-5. (20) Groenvold M, Petersen MA, Aaronson NK, Arraras JI, Blazeby JM, Bottomley A, et al. The development of the EORTC QLQ-C15-PAL: a shortened questionnaire for cancer patients in palliative care. Eur J Cancer 2006 Jan;42(1):55-64. (21) Grande GE, Farquhar MC, Barclay SI, Todd CJ. Valued aspects of primary palliative care: content analysis of bereaved carers' descriptions. Br J Gen Pract 2004 Oct;54(507):772-8. (22) Hanratty B. Palliative care provided by GPs: the carer's viewpoint. Br J Gen Pract 2000 Aug;50(457):653-4. (23) Aabom B, Kragstrup J, Vondeling H, Bakketeig LS, Stovring H. Population-based study of place of death of patients with cancer: implications for GPs. Br J Gen Pract 2005 Sep;55(518):684-9. (24) Gomes B, Higginson IJ. Factors influencing death at home in terminally ill patients with cancer: systematic review. BMJ 2006 Mar 4;332(7540):515-21. (25) Higginson IJ, Astin P, Dolan S. Where do cancer patients die? Ten-year trends in the place of death of cancer patients in England. Palliat Med 1998 Sep;12(5):353-63. (26) Grande GE, Addington-Hall JM, Todd CJ. Place of death and access to home care services: are certain patient groups at a disadvantage? Soc Sci Med 1998 Sep;47(5):56579. (27) Neergaard MA, Olesen F, Jensen AB, Sondergaard J. Palliative care for cancer patients in a primary health care setting: Bereaved relatives' experience, a qualitative group interview study. BMC Palliat Care 2008 Jan 15;7(1):1. (28) Burge F, Lawson B, Johnston G, Cummings I. Primary care continuity and location of death for those with cancer. J Palliat Med 2003 Dec;6(6):911-8. (29) Brazil K, Bedard M, Willison K. Factors associated with home death for individuals who receive home support services: a retrospective cohort study. BMC Palliat Care 2002 Mar 25;1(1):2-7. (30) Cantwell P, Turco S, Brenneis C, Hanson J, Neumann CM, Bruera E. Predictors of home death in palliative care cancer patients. J Palliat Care 2000;16(1):23-8.

142


(31) Weibull A, Olesen F, Neergaard MA. Caregivers' active role in palliative home care - to encourage or to dissuade? A qualitative descriptive study. BMC Palliat Care 2008 Sep 16;7(1):15. (32) Brazil K, Bedard M, Krueger P, Abernathy T, Lohfeld L, Willison K. Service preferences among family caregivers of the terminally ill. J Palliat Med 2005 Feb;8(1):69-78. (33) Grbich C, Parker D, Maddocks I. The emotions and coping strategies of caregivers of family members with a terminal cancer. J Palliat Care 2001;17(1):30-6. (34) Mok E, Chan F, Chan V, Yeung E. Family experience caring for terminally ill patients with cancer in Hong Kong. Cancer Nurs 2003 Aug;26(4):267-75. (35) Fukui S, Fukui N, Kawagoe H. Predictors of place of death for Japanese patients with advanced-stage malignant disease in home care settings: a nationwide survey. Cancer 2004 Jul 15;101(2):421-9. (36) Howat A, Veitch C, Cairns W. A retrospective review of place of death of palliative care patients in regional north Queensland. Palliat Med 2007 Jan;21(1):41-7. (37) Ringdal GI, Jordhoy MS, Kaasa S. Family satisfaction with end-of-life care for cancer patients in a cluster randomized trial. J Pain Symptom Manage 2002 Jul;24(1):53-63. (38) Medigovich K, Porock D, Kristjanson LJ, Smith M. Predictors of family satisfaction with an Australian palliative home care service: a test of discrepancy theory. J Palliat Care 1999;15(4):48-56. (39) Addington-Hall J, Walker L, ones C, arlsen S, cCarthy M. A randomised controlled trial of postal versus interviewer administration of a questionnaire measuring satisfaction with, and use of, services received in the year before death. J Epidemiol Community Health 1998;52:802-7. (40) Jacoby A, Lecouturier J, Bradshaw C, Lovel T, Eccles M. Feasibility of using postal questionnaires to examine carer satisfaction with palliative care: a methodological assessment. South Tyneside MAAG Palliative Care Study Group. Palliat Med 1999 Jul;13(4):285-98. (41) Lecouturier J, Jacoby A, Bradshaw C, Lovel T, Eccles M. Lay carers' satisfaction with community palliative care: results of a postal survey. South Tyneside MAAG Palliative Care Study Group. Palliat Med 1999 Jul;13(4):275-83. (42) Malacrida R, Bettelini CM, Degrate A, Martinez M, Badia F, Piazza J, et al. Reasons for dissatisfaction: a survey of relatives of intensive care patients who died. Crit Care Med 1998 Jul;26(7):1187-93. (43) Rogers A, Karlsen S, Addington-Hall J. 'All the services were excellent. It is when the human element comes in that things go wrong': dissatisfaction with hospital care in the last year of life. J Adv Nurs 2000 Apr;31(4):768-74. (44) Wilkinson HJ. Assessment of patient satisfaction and hospice: a review and an investigation. Hosp J 1986;2(4):69-94.

143


(45) Miyashita M, Morita T, Sato K, Hirai K, Shima Y, Uchitomi Y. Good death inventory: a measure for evaluating good death from the bereaved family member's perspective. J Pain Symptom Manage 2008 May;35(5):486-98. (46) Mainz J, Olesen F. Krav til spørgeskemaer I - validitet, reliabitet og diskrimationsevne. Månedsskr Prakt Lægegern 1994;72:1367-73. (47) Rothman KJ, Greenland S. Modern Epidemiology. Second Edition. Philadelphia: LippicottRaven Publishers; 1998. (48) Andersen TF, Madsen M, Jørgensen J, Mellemkjær M, Olsen JH. The Danish National Hospital Register. A valuable source of data for modern health sciences. Dan Med Bull 1999;46:263-8. (49) Nørgaard M, Skriver MV, Gregersen H, Pedersen G, Schønheyder HC, Sørensen HT. The data quality of haematological malignancy ICD-10 diagnoses in a population-based hospital discharge registry. Eur J Cancer Prev 2005;14:201-6. (50) Tetsche M, Nørgaard M, Skriver MV, Andersen ES, Lash TL, Sørensen HT. Accuracy of ovarian cancer ICD-10 diagnosis in a Danish population-based hospital discharge registry. Eur J Gynaecol Oncol 2005;26(3):266-70. (51) Nickelsen TN. [Data validity and coverage in the Danish National Health Registry. A literature review]. Ugeskr Laeger 2001 Dec 31;164(1):33-7. (52) Pedersen CB, Gotzsche H, Moller JO, Mortensen PB. The Danish Civil Registration System. A cohort of eight million persons. Dan Med Bull 2006 Nov;53(4):441-9. (53) The Danish National Board of Health. The Danish Register of Causes of Death [In Danish]. 2008. (54) Protti D. Comparison of information technology in general practice in 10 countries. Healthc Q 2007;10(2):107-16. (55) Addington-Hall J, McPherson C. After-death interviews with surrogates/bereaved family members: some issues of validity. J Pain Symptom Manage 2001 Sep;22(3):784-90. (56) Rothman KJ. Epidemiology. An introduction. New York: Oxford University Press; 2002. (57) Kirchhof KT, Kehl KA. Recruiting Participants in End-of-Life Research. Am J Hosp Palliat Care 2008;24:515-21. (58) Hudson P. The experience of research participation for family caregivers of palliative care cancer patients. Int J Palliat Nurs 2003 Mar;9(3):120-3.

144

Chapter 11. Discussion of results

CHAPTER 11 DISCUSSION OF RESULTS

145


In the following chapter, two overall issues are discussed: Quality of care at the end of life and relatives‟ versus patients‟ evaluation. Then, a discussion of each of the main results in Papers 1 – 5 follows. 11.1. Quality of care at the end of life Three main end-of-life quality constructs exist: Quality of life (at the end of life), quality of death and dying and quality of care at the end of life (1). Quality of life at the end of life includes the domains of physical, psychological, social and spiritual or existential experiences (2). Quality of death and dying incorporates the domains of quality of life, but also includes the nature of health care, life closure and circumstances of death, whereas, quality of care at the end of life incorporates all of the end-of-life experiences, but are specially focussed on the extent to which health care structure and processes affect these outcomes (1). In this thesis we have focused on quality of care but the three end-of-life quality constructs are overlapping and equally important.

As mentioned in Chapter 2 international studies show that most patients and their relatives wish terminal care and death to take place at home (3-7). Hence, home death is often used as a measure of a good death in lack of better outcome measures. But quality in dying is much debated and there is no doubt that what constitutes a good death is determined by a complex interplay of personal and cultural values, supportive network characteristics, and physical and medical factors, as well as the efforts displayed by the health care system (8-10). Using plus / minus home death as a single outcome measure may therefore be a somewhat simplified way of evaluating a palliative pathway. In or study (Paper 3) the patients, who eventually died at institutions, spent more than 75% of their time at home during the palliative pathway (84% for the home death group).This definitely questions the use of home death as measure of good quality of terminal care. A „successful palliative pathway‟ can be more than achieving home death. The reasons for hospitalisation during the last days of life may be many, and hospital-death may represent successful care just as well as home death. Studies are therefore warranted to establish what constitutes a successful palliative pathway for patients and relatives and how primary care professionals may be involved in that pathway. Finally the results emphasise the need for research that addresses the quality in the total palliative pathway from day one until death.

In examining quality of care in death and dying it seems obvious to use a qualitative research approach because of the complexity of the issue, and many studies have found themes important to achieve a good death (10). However, several studies have tried to operationalise bereaved relative‟s views of quality of care retrospectively (11-16;16-19), but not with a focus on the primary

146


health care professionals‟ involvement. Some questionnaires have included services provides from primary care professionals (20-26), but they have only been used at a descriptive level. The findings of these studies are reported in the following, when relevant to our main results. 11.2. Relatives’ versus patients’ evaluation Generally, the patient is regarded as the most valid source of collecting data in the palliative pathway (2). An alternative approach is to use relatives to assess the quality of care. A retrospective after-death approach can give information about the care the patient received, as well as give information about the experience of the relative (27).

When we asked bereaved relatives of their experiences and views on the care provided in the palliative pathway, we implied that relatives were more or less a proxy or an indicator of the patient‟s voice (28). Research has shown that agreement between patients‟ and relatives‟ reports is poor as far as subjective aspects are concerned, e.g. pain, anxiety and depression where the relatives retrospectively score these symptoms higher than the terminally ill patients did, but in retrospect relatives can provide reliable information on the quality of services (29;30). We therefore hope that the patients‟ views on the care in the palliative pathway have been sufficiently covered by the relatives‟ views since terminally ill patients are the main recipient in palliative care and the achievement of a successful palliative pathway and a good death for these patients is the primary reason for doing this study.

The practical and ethical advantages of doing research on bereaved relatives and not on the dying patients themselves are obvious, but both approaches are needed to evaluate palliative pathways in relation to primary care and thereby, hopefully, identify factors of the services provided that can enhance a good palliative pathway and death to patients and relatives.

11.3. Discussion of main results from Papers 1 - 5 In the following, the main results from the papers are highlighted in relation to the research questions. The findings are discussed in relation to existing literature on the subject and the consequences of the findings to clinical and administrative decisions are considered.

147


11.3.1. Research question A In research question A, suggestions, wishes and experiences coincident with successfulness of palliative pathways at home were explored based on input from bereaved relatives, GPs, CNs and hospital consultants. The results from Papers 1 and 2 can be summarized as follows: The bereaved relatives asked for health professionals who had the necessary knowledge, were available and actively sought to establish personal contact with them (Paper 1). In earlier studies of the professionals‟ management of palliative care, some relatives were reported not to be satisfied with symptom control in primary care, but they still rated the overall GP care as very good (6). The overall satisfaction with the GPs was significantly associated with symptom control, time to listen, accessibility and, whenever the patient died, death at the preferred place (6;23). In our study, continuity of care and good knowledge of the whole course of disease were among the reasons why the GPs were valued despite their at times poor management of palliative care. Furthermore, in line with other interview studies the accessibility and the attitude of the GP, and the relationship with the primary care professionals were important aspects (5). The relatives described both organizational and cultural problems in the health care system (Paper 1) and the professionals suggested improved cooperation in palliative homecare through organizational and working culture improvements (Paper 2)

The importance of communication and shared care between primary care professionals, hospital professionals and specialists in palliative care is obvious from our interview studies. In line with our study, extant research emphasizes that continuous contact with the GP can be maintained by referring the patients back to their GPs after hospitalization and by ensuring good communication between specialists and GPs (31;32). This may also include steps to ensure a re-established contact and relation between the GP and the patient when they leave hospital and the taking of active steps to ensure professional collaboration (33). Part of this can be achieved by developing better IT and common web-based information systems and by providing a clear definition of the GP‟s role (9;34).

It seems from our study that important barriers to optimizing shared care and co-operation may lie in the lack of a home-care coordinator in palliative care, which has also been seen in other studies

148


(9;35), and a lack of a shared care culture among health professionals. Professionals must acknowledge the strengths of other health-professionals (36), a theme also raised in the present study. It is thought-provoking that not only the professionals but also the bereaved relatives in Paper 1 saw poor cooperation as a barrier to successful palliative care. It has been shown, that coordinated care and team meetings improved interprofessional communication in palliative pathways using the Gold Standards Framework, which is a British tool to improve the delivery of palliative care in general practice (37). The professionals in Paper 2 also described that whenever the health-professionals met one another in person, the chance of mutual respect increased (38) and nurtured a feeling of team-spirit promoting responsibility. In Denmark the Danish College of General Practitioners has published a guideline to palliative care for GPs but a more overall national guideline to improve interprofessional co-operation and thus improving quality of palliative care is needed (39) and it should be followed by health authorities‟ decision on incentives to participate in this co-operation by all professionals involved. The professionals described a division between basic and specialist level of palliative care that seemed to be characterized by arbitrariness (Paper 2).

In Denmark, palliative care and palliative education has been given a low priority compared to surrounding countries (40). Even though the Danish National Cancer plan recommend a division between basic and specialist level of palliative care (41) no initiative has been taken from the Danish health authorities to support this division in practice. The consequence is an organisation with specialised palliative services (outgoing specialist teams, hospices, specialist palliative ward) but no national guideline specify the criteria of admission to these services. We found indications that during the planning phase, the likelihood that a terminally ill patient be referred to a specialist team when discharged to palliative care at home, depends on the hospital professionals‟ knowledge of GPs providing palliative care and not as much on the patient‟s symptoms or a prediction which problems would arise during the palliative period as ought to be the case deciding on the division between basic and specialist levels of palliative care (41). There is a lack of educated specialised palliative professionals in Denmark (42), and if the tendency of arbitrariness in the division between basic and specialist levels holds true, this is counterproductive to the use of specialised resources. More research and education is needed in relation to the division between basic and specialist levels of palliative care and there seems to be a need for more precise agreements on collaboration and knowledge transfer between professionals involved in palliative home care.

149


The bereaved relatives felt an unspoken pressure from the professionals to be “semiprofessionals” themselves (Paper 1).

We found that the relatives feel emotionally and physically exhausted in the palliative pathway, as also found in other studies (7;43), which may be due to their “professionalization” and to the absence of support from health care professionals, particularly after bereavement. It is important that the professionals help the relatives strike a balance between their needs, wishes and resources (43;44), but this balance can be difficult to find in end-of-life care situations, where the relatives feel obliged to fulfil the wishes of their dying relative. In another interview study we found that prerequisites for a positive outcome when the relative were actively involved in the palliative pathway were 24-hour back-up, professionals' confidence in the relatives' abilities and that relatives had influence on decision-making and were asked about needs and wishes (45). There are much evidence that the relatives of terminally ill patients have unmet needs (46;47), and it seems that professionals need to have a more active approach to the relatives and, most importantly, to have a dialogue with the relatives concerning their needs, wishes and resources in the palliative pathway.

11.3.2. Research question B Research question B explored associations between home death and evaluations of palliative pathways, respectively, in a series of newly finished palliative pathways at home. The findings in Papers 3, 4 and 5 can be divided in the following issues: In a consecutive group of cancer patients who died from cancer and had a palliative pathway at home (incl. nursing home), the following results were obtained (Paper 3): -

GPs were involved in 89% of the pathways, knew the patient well before the palliative period

and had contact with their relatives -

In nearly half of these cases, GPs paid at least four home visits during the last three months of

the patients‟ lives and nearly one third of the home visits during the last three months of the patients‟ lives were out-of-hours visits -

In approximately 40% of the cases, GPs provided special services like unplanned home visits

and offered their private phone numbers for contact if acute problems in out-of-hours would occur -

In two thirds of the cases, a GP and a CN were both involved, and in one third a GP and a

specialist team were both involved. In 80 cases (24.0%), GPs were involved without the CNs and the specialist team

150


The GPs stated that they were involved in nearly nine out of ten of the cases included in the study, but they paid unplanned home visits and offered their private phone numbers in only approximately 40% of the cases. It is interesting to note what amounts to „being involved‟ according to the GPs. We did not explore the different pathways to determine if there was an association between the duration of the palliative period at home and the GPs services, which would be interesting, because it seems that there are many ways a GP can be involved. It would therefore be interesting to examine these “packages” of services provided by the GPs, and to identify what is required to achieve home death or a successful pathway.

The findings of the involvement of GPs, CNs and specialist teams provide new knowledge of palliative pathways‟ use of health care services. In one third of the cases in our study-database the GP and the specialist team were both involved and in one quarter of the cases the GPs were involved without the CNs and the specialist team. These findings point out the importance of focus on the basic level palliative care and the need of registration of all palliative patients to do further research and quality assurance in the effort of achievement of the best quality in palliative care. If one focuses only on specialist level palliative care in future research the majority of palliative pathways and their characteristics (e.g. symptom severity, need of care and treatment, quality of services provided, wish for and actual place of death) , which may be a lot different from the characteristics of specialist level palliative pathways, are not even represented.

•

We found a positive association between home death and GPs making home visits

during the last three months and an increased likelihood of home death when GPs paid three or more home visits in this period (Paper 3).

This finding fits well with the results of other studies (48-50). This is hardly a surprising association because dying at home often demands more home visits and more home care, and the patient has to be at home to receive these visits. In our study, we attempted to adjust for this confounder by including time at home as a variable, but the association persisted.

If GPs are paying many home visits per time unit during the palliative period, this would testify to their willingness to assume different tasks and provide different services in relation to palliative pathways. We found no correlations between the different GP variables in the model, but it would be interesting to further explore what different GPs do in the palliative pathway and if certain „packages‟ of services are a more useful measure of the GP‟s involvement than just an

151


assessment of home visits. Our findings suggest that a home visit is a proxy for a whole quality package of palliative care from the GP, and future research should address the precise contents of this package. Unfortunately, in our study we were unable to analyse data at the level of the individual GP, since most of the GPs did not state their name on the questionnaire. This was also the reason for adjusting for clustering at practice level in the models in Papers 3 and 4. •

Bereaved relatives’ positive evaluation of the palliative pathway was positively

associated with home death. However, we identified no associations between the evaluations and the ways in which the GPs (Paper 4) or the CNs were involved (Paper 5). In line with our findings that home death was statistically significantly associated with the relatives‟ evaluation of the palliative pathway as successful, other studies found that home death was associated with better bereavement response (51-53) and a higher satisfaction with the services provided (54). The fact that some of the association between a successful pathway and home death in our study may partly be due to selection bias questions the use of home death as the only indicator of a good death. To our knowledge, no other quantitative studies have examined the association between relatives‟ or patients‟ evaluation of the entire palliative pathway and GP or CN-related factors. However, Hanratty found that bereaved relatives of patients who died in their preferred place were significantly more likely to rate GP care as good or excellent (6). Lecoutrier et al found that bereaved relatives‟ views about the quality of the GP care overall were positively significantly associated with perceived ease of getting an appointment, willingness to make home visits, time to listen and to discuss matters fully and with efforts in regard of symptom relief. Furthermore, they found that the quality of the CN care overall were positively significantly associated with e.g. number of CN home visits a day, degree and timing of help received and whether the nurse spent enough

time, knew enough and provided enough support (23). This may indicate that the

variables in our study assessing the characteristics and quality of the GP and CN services have not been sufficiently sensitive or that the power of our study has been to low.

In Paper 3 we found an association between home death and involvement of CNs. We concluded that addressing this association was not a separate aim in Paper 3, and that the found association deserved further investigation. However, other studies found a significant association between home death and the involvement of CNs (48;55;56), and further research into the services in palliative care provided by the CNs is needed.

152


11.4. References of Chapter 11 (1) Hales S, Zimmermann C, Rodin G. The quality of dying and death. Arch Intern Med 2008 May 12;168(9):912-8. (2) Kaasa S, Loge JH. Quality of life in palliative care: principles and practice. Palliat Med 2003 Jan;17(1):11-20. (3) Thomas C, Morris SM, Clark D. Place of death: preferences among cancer patients and their carers. Soc Sci Med 2004 Jun;58(12):2431-44. (4) Thomas C. The place of death of cancer patients: can qualitative data add to known factors? Soc Sci Med 2005 Jun;60(11):2597-607. (5) Grande GE, Farquhar MC, Barclay SI, Todd CJ. Valued aspects of primary palliative care: content analysis of bereaved carers' descriptions. Br J Gen Pract 2004 Oct;54(507):772-8. (6) Hanratty B. Palliative care provided by GPs: the carer's viewpoint. Br J Gen Pract 2000 Aug;50(457):653-4. (7) Brazil K, Howell D, Bedard M, Krueger P, Heidebrecht C. Preferences for place of care and place of death among informal caregivers of the terminally ill. Palliat Med 2005 Sep;19(6):492-9. (8) Tang ST, McCorkle R. Determinants of place of death for terminal cancer patients. Cancer Invest 2001;19(2):165-80. (9) Steinhauser KE, Christakis NA, Clipp EC, McNeilly M, McIntyre L, Tulsky JA. Factors considered important at the end of life by patients, family, physicians, and other care providers. JAMA 2000 Nov 15;284(19):2476-82. (10) Steinhauser KE, Clipp EC, McNeilly M, Christakis NA, McIntyre LM, Tulsky JA. In search of a good death: observations of patients, families, and providers. Ann Intern Med 2000 May 16;132(10):825-32. (11) Wilkinson HJ. Assessment of patient satisfaction and hospice: a review and an investigation. Hosp J 1986;2(4):69-94. (12) Malacrida R, Bettelini CM, Degrate A, Martinez M, Badia F, Piazza J, et al. Reasons for dissatisfaction: a survey of relatives of intensive care patients who died. Crit Care Med 1998 Jul;26(7):1187-93. (13) Archer KC, Boyle DP. Toward a measure of caregiver satisfaction with hospice social services. Hosp J 1999;14(2):1-15. (14) Baker R, Wu AW, Teno JM, Kreling B, Damiano AM, Rubin HR, et al. Family satisfaction with end-of-life care in seriously ill hospitalized adults. J Am Geriatr Soc 2000 May;48(5 Suppl):S61-S69. (15) Patrick DL, Engelberg RA, Curtis JR. Evaluating the quality of dying and death. J Pain Symptom Manage 2001 Sep;22(3):717-26.

153


(16) Teno JM, Clarridge B, Casey V, Edgman-Levitan S, Fowler J. Validation of Toolkit AfterDeath Bereaved Family Member Interview. J Pain Symptom Manage 2001 Sep;22(3):7528. (17) Curtis JR, Patrick DL, Engelberg RA, Norris K, Asp C, Byock I. A measure of the quality of dying and death. Initial validation using after-death interviews with family members. J Pain Symptom Manage 2002 Jul;24(1):17-31. (18) Sulmasy DP, McIlvane JM, Pasley PM, Rahn M. A scale for measuring patient perceptions of the quality of end-of-life care and satisfaction with treatment: the reliability and validity of QUEST. J Pain Symptom Manage 2002 Jun;23(6):458-70. (19) Miyashita M, Morita T, Sato K, Hirai K, Shima Y, Uchitomi Y. Good death inventory: a measure for evaluating good death from the bereaved family member's perspective. J Pain Symptom Manage 2008 May;35(5):486-98. (20) Fakhoury W, McCarthy M, Addington-Hall J. Determinants of informal caregivers' satisfaction with services for dying cancer patients. Soc Sci Med 1996 Mar;42(5):721-31. (21) Fakhoury WK, McCarthy M, Addington-Hall J. The effects of the clinical characteristics of dying cancer patients on informal caregivers' satisfaction with palliative care. Palliat Med 1997 Mar;11(2):107-15. (22) Jacoby A, Lecouturier J, Bradshaw C, Lovel T, Eccles M. Feasibility of using postal questionnaires to examine carer satisfaction with palliative care: a methodological assessment. South Tyneside MAAG Palliative Care Study Group. Palliat Med 1999 Jul;13(4):285-98. (23) Lecouturier J, Jacoby A, Bradshaw C, Lovel T, Eccles M. Lay carers' satisfaction with community palliative care: results of a postal survey. South Tyneside MAAG Palliative Care Study Group. Palliat Med 1999 Jul;13(4):275-83. (24) Addington-Hall J, Walker L, ones C, arlsen S, cCarthy M. A randomised controlled trial of postal versus interviewer administration of a questionnaire measuring satisfaction with, and use of, services received in the year before death. J Epidemiol Community Health 1998;52:802-7. (25) Addington-Hall J, McCarthy M. Regional Study of Care for the Dying: methods and sample characteristics. Palliat Med 1995 Jan;9(1):27-35. (26) Rogers A, Karlsen S, Addington-Hall J. 'All the services were excellent. It is when the human element comes in that things go wrong': dissatisfaction with hospital care in the last year of life. J Adv Nurs 2000 Apr;31(4):768-74. (27) Mayland C, Williams E, Ellershaw J. How well do current instruments using bereaved relatives' views evaluate care for dying patients? Palliat Med 2008 Mar;22(2):133-44. (28) Hinton J. How reliable are relatives' retrospective reports of terminal illness? patients' and relatives' accounts compared. Soc Sci Med 1996;43(8):1229-36. (29) Addington-Hall J, McPherson C. After-death interviews with surrogates/bereaved family members: some issues of validity. J Pain Symptom Manage 2001 Sep;22(3):784-90.

154


(30) McPherson CJ, Addington-Hall JM. Judging the quality of care at the end of life: can proxies provide reliable information? Soc Sci Med 2003 Jan;56(1):95-109. (31) Norman A, Sisler J, Hack T, Harlos M. Family physicians and cancer care. Palliative care patients' perspectives. Can Fam Physician 2001 Oct;47:2009-12,;%2015-6.:2009-12. (32) Borgsteede SD, Graafland-Riedstra C, Deliens L, Francke AL, van Eijk JT, Willems DL. Good end-of-life care according to patients and their GPs. Br J Gen Pract 2006 Jan;56(522):20-6. (33) Nielsen JD, Palshof T, Mainz J, Jensen AB, Olesen F. Randomised controlled trial of a shared care programme for newly referred cancer patients: bridging the gap between general practice and hospital. Qual Saf Health Care 2003 Aug;12(4):263-72. (34) Evans R, Stone D, Elwyn G. Organizing palliative care for rural populations: a systematic review of the evidence. Fam Pract 2003 Jun;20(3):304-10. (35) Norris K, Merriman MP, Curtis JR, Asp C, Tuholske L, Byock IR. Next of kin perspectives on the experience of end-of-life care in a community setting. J Palliat Med 2007 Oct;10(5):1101-15. (36) Bliss J, Cowley S, While A. Interprofessional working in palliative care in the community: a review of the literature. Journal of interprofessional care 2000;14(3):281-90. (37) Thomas K, Noble B. Improving the delivery of palliative care in general practice: an evaluation of the first phase of the Gold Standards Framework. Palliat Med 2007 Jan;21(1):49-53. (38) Mitchell G, Cherry M, Kennedy R, Weeden K, Burridge L, Clavarino A, et al. General practitioner, specialist providers case conferences in palliative care--lessons learned from 56 case conferences. Aust Fam Physician 2005 May;34(5):389-92. (39) Ferris FD, Gomez-Batiste X, Furst CJ, Connor S. Implementing quality palliative care. J Pain Symptom Manage 2007 May;33(5):533-41. (40) Centeno C, Clark D, Lynch T, Racafort J, Praill D, De Lima L, et al. Facts and indicators on palliative care development in 52 countries of the WHO European region: results of an EAPC Task Force. Palliat Med 2007 Sep;21(6):463-71. (41) The National Board of Health. National Cancer Plan II • Denmark National Board of Health recommendations for improving cancer healthcare services. Copenhagen: The National Board of Health; 2005. (42) Sjogren P. [Palliative medicine--an independent medical specialty?]. Ugeskr Laeger 2007 Oct 29;169(44):3735. (43) Grbich C, Parker D, Maddocks I. The emotions and coping strategies of caregivers of family members with a terminal cancer. J Palliat Care 2001;17(1):30-6. (44) Proot IM, Abu-Saad HH, Crebolder HF, Goldsteen M, Luker KA, Widdershoven GA. Vulnerability of family caregivers in terminal palliative care at home; balancing between burden and capacity. Scand J Caring Sci 2003 Jun;17(2):113-21.

155


(45) Weibull A, Olesen F, Neergaard MA. Caregivers' active role in palliative home care - to encourage or to dissuade? A qualitative descriptive study. BMC Palliat Care 2008 Sep 16;7(1):15. (46) Harding R, Higginson IJ. What is the best way to help caregivers in cancer and palliative care? A systematic literature review of interventions and their effectiveness. Palliat Med 2003 Jan;17(1):63-74. (47) Bee PE, Barnes P, Luker KA. A systematic review of informal caregivers' needs in providing home-based end-of-life care to people with cancer. J Clin Nurs 2008 Jul 1. (48) Aabom B, Kragstrup J, Vondeling H, Bakketeig LS, Stovring H. Population-based study of place of death of patients with cancer: implications for GPs. Br J Gen Pract 2005 Sep;55(518):684-9. (49) Brazil K, Bedard M, Willison K. Factors associated with home death for individuals who receive home support services: a retrospective cohort study. BMC Palliat Care 2002 Mar 25;1(1):2-7. (50) Cantwell P, Turco S, Brenneis C, Hanson J, Neumann CM, Bruera E. Predictors of home death in palliative care cancer patients. J Palliat Care 2000;16(1):23-8. (51) Grande GE, Farquhar MC, Barclay SI, Todd CJ. Caregiver bereavement outcome: relationship with hospice at home, satisfaction with care, and home death. J Palliat Care 2004;20(2):69-77. (52) Akiyama A, Numata K, Mikami H. Factors enabling home death of the elderly in an institution specializing in home medical care: analysis of apprehension of the bereaved family. Geriatr Gerontol Int 2008 Jun;8(2):73-9. (53) van der Heide A, de Vogel-Voogt E, Visser AP, van der Rijt CC, van der Maas PJ. Dying at home or in an institution: perspectives of Dutch physicians and bereaved relatives. Support Care Cancer 2007 Dec;15(12):1413-21. (54) Ringdal GI, Jordhoy MS, Kaasa S. Family satisfaction with end-of-life care for cancer patients in a cluster randomized trial. J Pain Symptom Manage 2002 Jul;24(1):53-63. (55) Fukui S, Fukui N, Kawagoe H. Predictors of place of death for Japanese patients with advanced-stage malignant disease in home care settings: a nationwide survey. Cancer 2004 Jul 15;101(2):421-9. (56) Howat A, Veitch C, Cairns W. A retrospective review of place of death of palliative care patients in regional north Queensland. Palliat Med 2007 Jan;21(1):41-7.

156

Chapter 12. Conclusion thesis

CHAPTER 12 CONCLUSION THESIS

157


12.1. Conclusion in relation to the overall aim of the thesis We obtained knowledge about bereaved relatives‟, GPs‟, CNs‟ and hospital consultants‟ experiences, wishes and suggestions for improvement of the palliative effort through interviews. Special attention was given to the wishes for the functioning of general practice and the interaction and co-operation between the primary and the secondary health care sectors.

We found that the relatives appreciated the effort of GPs and CNs, but also that improvement was needed (see Conclusion in relation to aim 1). The professionals who participated in the interviews were all committed to palliative care, but they disclosed the existence of problems with organisation and cooperation among professionals (see Conclusion in relation to aim 2).

Through the combined register and questionnaire study, we obtained knowledge of how bereaved relatives evaluated a series of newly finished palliative pathways at home, and found factors associated with home death and experiences of successful palliative pathways (see Conclusion in relation to aims 3 – 5).

12.2. Conclusion in relation to aim 1 The interview study with bereaved relatives (Paper 1) indicated that relatives often experienced insufficient palliative care, mainly due to organizational and cultural problems among professionals and the lack of a shared care scheme. Palliative care in primary care in general needs improvement, both in terms of professional knowledge, attitude, availability and communication skills. Furthermore, attention was drawn to the "professionalization" of the relatives and the need to strike a balance between their needs, wishes and resources in end-of-life care and bereavement. 12.3. Conclusion in relation to aim 2 The interview study with GPs, CNs and hospital consultants (Paper 2) indicated problems with both organization among health-professionals and working culture. The informants drew attention to problems with task distribution, information exchange, availability, respect and personal acquaintance among professionals. 12.4. Conclusion in relation to aim 3 In a consecutive group of cancer patients who died from cancer and had a palliative pathway at home (incl. nursing home) we found that even though GPs were involved in 89% of the pathways, knew the patient well before the palliative period, had contact with their relatives and paid nearly one third of the home visits during the last three months of the patients‟ lives out-of-hours, they

158


only provided special services like unplanned home visits and offered their private phone numbers in about 40% of the cases. Furthermore, the patients who eventually died at institutions spent more than 75% of their time at home during the palliative pathway (84% for the home death group) Active involvement of GPs providing home visits and the use of home nurses were independently associated with a higher likelihood of home death. 12.5. Conclusion in relation to aim 4 The study (Paper 4) showed that home death was positively associated with relatives evaluating the palliative pathway as successful. The relation between relative and patient and between GP and relative may also of importance. 12.6. Conclusion in relation to aim 5 The study (Paper 5) showed that home death was positively associated with relatives evaluating the palliative pathway as successful. 12.7. Conclusion of thesis To sum up, the PhD thesis has shown that there is a scope for developing and improving the basic level palliative care effort in the patients‟ homes. Particularly important is the foci on the relatives‟ role, the co-operation of professionals, both in terms of organisation and working culture, and the primary care professionals‟ services and further involvement in the effort to achieve as successful palliative

pathways

for

patients

and

relatives

as

possible.

159

160

Chapter 13. Perspectives and implications for health care

CHAPTER 13 PERSPECTIVES AND IMPLICATIONS FOR HEALTH CARE

161

Chapter 13. Perspectives and implications for health care

Perspectives and Implication for health care The present study leaves the question of how to establish a better and more solid foundation of organisation of palliative home care, to guide GPs and CNs in what way to be involved and to guide staff in hospitals in ways of supporting palliative home care. Next the studies raise the question of how to improve co-operation among professionals involved in both primary and secondary health care sectors.

Further education in palliative care is needed, both on pre- and post graduation level, and on basic and specialist level care. The education should focus on the importance of co-operation both across and within the health care sectors and on symptom control and other patient-specific factors. In this context it seems important both to give education to primary care professionals and to professionals in secondary care on ways to improve co-operation and connection between the health care sectors in the palliative pathways.

From Paper 2 it is seen that for cooperation among professionals to be improved, it is necessary to explore the organising of the palliative care effort and the nature and mechanisms of interprofessional relations. The following issues should be addressed: Planning of the palliative pathway from the beginning, clear distribution of tasks and team leadership, information exchange and availability of professionals involved and respect, shared care culture and personal acquaintance among professionals (Tabel 6.4, Chapter 6). A main challenge to the health care system is certainly to create a shared care culture characterised by cooperation and mutual respect among professionals within the health care system. This challenge starts within the educational institutions of doctors, nurses and other health care professionals with information and education in interdisciplinary co-operation. And it extends to focus on interdisciplinary co-operation of political and health care authorities and of the single head of hospital wards and units. At the individual level it seems that personal acquaintance is important and spending time and resources on meetings or telephone calls may be a good investment in breaking down barriers to interdisciplinary co-operation.

The problems outlined in the interview studies indicate that national and local guidelines in palliative care are needed, not only on care and treatment, but also on organisation (e.g. division in basal and specialised palliative care) and co-operation in palliative care to ensure consistent, successful palliative pathways. Further improvement on actual performance should be in focus and this will normally demand a well-planned strategy for implementation of guidelines and new knowledge.

162

Chapter 14. Future research

CHAPTER 14 FUTURE RESEARCH

163


Future research Research in palliative care in Denmark has so far been sparse. Research in palliative care is needed on many issues but particularly limited has been health services research into how to organise the effort.

From this thesis the following foci for future research in organisation of palliative home care seem important: 

The is a need to analyse consequences for patients, relatives and professionals of

implementing shared care as a GP-supporting method of organising palliative care at home compared to a specialist controlled patient-supporting method of organising palliative care at home. An intervention study of this nature is currently being conducted. 

There is a need to improve communication between professionals; to develop better IT

systems and common web-based information platforms. A study on tele-home care and a patientspecific it-information-platform for patient, relatives, GPs, CNs, hospital staff and a palliative specialist team is currently being prepared. 

Research that adopts a broad perspective into professionals‟ barriers to co-operation and

shared care with a focus on educational, cultural, personal and organisational barriers is needed

Furthermore, it is important in future research to address the following issues on how to achieve quality in palliative home care: 

Predictors in the primary care effort associated with a “good death” with a view to optimising

and focusing the palliative primary health care effort 

Danish terminally ill cancer patients‟ preferences for place of palliative care and death and

the change of this wish over time in the palliative period. Furthermore, associations of the fulfilment of wish for place of death with factors related to primary health care 

Relative‟s role in palliative care. Their need of support and the degree of achievement of

support

164


Also further research with a focus on palliative pathways‟ use of health services is needed: 

Research is needed into description of palliative pathways, including their total use of health

services, both in the primary and secondary health care sectors. The associations between the services used and place of death or evaluation of the palliative pathways is required Finally the primary health care professionals‟ roles in palliative home care are needed to be further investigated by addressing the following issues: 

GPs‟ and CN‟s views on involvement in palliative care in general, and especially personal

and organisational barriers towards involvement in palliative care in quantitative surveys. 

Patients‟, relatives‟, GP‟s and CNs views on the gate-keeper role in palliative care in

quantitative surveys.

165

166

Chapter 15. English summary

CHAPTER 15 ENGLISH SUMMARY

167


English summary This thesis is based on the project: “Palliative home care for cancer patients in Denmark - with a particular focus on the present and future role of general practice” – a mixed method study conducted from 2005 to 2008 in the former Aarhus County, Denmark. The thesis is based on five papers and focuses on palliative care for cancer patients at home. Introduction Approximately forty percent of all cancer deaths in Denmark occur in the patients‟ homes (incl. nursing homes), but most patients and their relatives wish terminal care and death to take place at home. The present organisation of palliative care does not succeed in achieving home death to the extent seemingly requested by patients and relatives. There is a need for a deeper insight into experiences, barriers and facilitators of home death or even better, delivery of good palliative home care. Aims The overall research questions of this thesis were: Research question A: Which suggestions for improvement and which wishes to the function and organisation of palliative care at home do bereaved relatives, GPs, CNs and consultants at relevant hospital wards have, and which factors are related to experiences of successful and unsuccessful palliative pathways? Research question B: How do bereaved relatives, general practitioners and community nurses evaluate a series of newly finished palliative pathways at home, and which factors are associated with experiences of successful and unsuccessful palliative pathways? The specific aim of this thesis was divided in the following: 1: To analyse experiences and preferences of bereaved relatives of terminally ill cancer patients in a primary care setting to explore barriers and facilitators for delivery of good palliative home care (Paper 1). 2: To illuminate involved health-professionals‟ experiences and views on interprofessional cooperation in basic level palliative home-care (Paper 2). 3: To describe cancer patients in palliative home care in relation to demographic characteristics, palliative pathway, degree of general practitioner involvement, and to examine the association between home death and general practitioner involvement (Paper 3). 4: To examine associations between the bereaved relatives‟ evaluation of palliative pathways, place of death and general practitioners‟ involvement (Paper 4). 5: To examine the association between the bereaved relatives‟ evaluation of palliative pathways and the place of death and the community nurses‟ involvement (Paper 5). Methods The study was a mixed method study firstly with focus group interviews with participation of bereaved relatives and health care professionals involved and secondly a combined register and questionnaire study on general practitioners, community nurses and bereaved relatives in relation to a consecutive population of 599 cases of deceased cancer patients.

168


Results Interviews with bereaved relatives indicated that relatives experience insufficient palliative care, mainly due to organizational and cultural problems among professionals. Palliative care in primary care in general needs improvement, both in terms of professional knowledge, communication skills, attitude and availability to patients, relatives and other professionals involved. Furthermore, attention should be drawn to the "professionalization" of the relatives (Paper 1). Interviews with general practitioners, community nurses and hospital consultants describe problems with both organization among health-professionals and working culture. The main issues were: distribution of tasks, information exchange, availability, interprofessional respect and personal acquaintance (Paper 2). In the first questionnaire survey it is shown that active involvement of general practitioners providing home visits and the involvement of home nurses were independently associated with a higher likelihood of home death (Paper 3). The second and third surveys indicate that dying at home is positively associated with a higher likelihood of the bereaved relative to evaluate the palliative pathway as successful, but no associations between „a successful pathway‟ and the general practitioners and community nurses‟ efforts were found (Paper 4 and 5). The relation between the relative and the cancer patient and between the general practitioner and the relative may also be of importance (Paper 4). Conclusion and perspectives The PhD thesis has shown that there is a scope for developing and improving the basic level palliative care effort in the patients‟ homes. Particularly important is to further explore the relatives‟ role, the co-operation of professionals, both in terms of organisation and working culture, and the primary care professionals‟ services and involvement in the effort to achieve as successful palliative pathways as possible. Finally, education of both primary and secondary health care professionals and further research into the organisation of the palliative care effort should be strengthened.

169

170

Chapter 16. Danish Summary

CHAPTER 16 DANISH SUMMARY

171


Dansk resumé af ph.d. tesen med den danske titel: Den palliative indsats for cancer patienter i Danmark – med et særligt fokus på primærsektoren, praktiserende læger og hjemmesygeplejersker. Et mixed-method-studie blev foretaget fra 2005 til 2008 i det tidligere Århus Amt omhandlende den palliative indsats for cancer patienter i patienternes eget hjem. Ph.d. tesen er baseret på fem artikler fra dette studie. Baggrund Omkring en fjerdedel af alle cancer dødsfald foregår i patientens eget hjem, men langt de fleste patienter og deres pårørende ønsker, at patienten plejes og dør hjemme. Den nuværende organisation af den palliative indsats i Danmark kan åbenbart ikke opfylde patienternes og de pårørendes ønske om dødssted. Der er et behov for at opnå viden om erfaringer, barrierer og facilitatorer for hjemmedød eller endnu vigtigere, for et så godt palliativt forløb som muligt. Formål De overordnede forskningsspørgsmål for ph.d. tesen var: Forskningsspørgsmål A: Hvilke forslag til forbedringer og ønsker til funktion og organisation af palliativ hjemmebehandling har efterladte pårørende, praktiserende læger, hjemmesygeplejersker og overlæger på relevante hospitalsafdelinger og hvilke faktorer er relateret til oplevelser af et succesfuldt og et ikke succesfuldt palliativt forløb? Forskningsspørgsmål B: Hvordan evalueres en serie nyligt afsluttede palliative forløb i hjemmet og hvilke faktorer er associerede med oplevelsen af et succesfuldt og et ikke succesfuldt palliativt forløb? Det specifikke formål med ph.d. tesen var inddelt i de følgende formål: 1: At analysere erfaringer og ønsker fra efterladte pårørende til cancer patienter for at undersøge barrierer og facilitatorer for at opnå et godt palliativt forløb i forhold til primærsektoren (Artikel 1). 2: At analysere involverede sundhedsprofessionelles erfaringer og syn på tværfaglig samarbejde på basis niveau af palliativ hjemmepleje (Artikel 2). 3: At beskrive døende cancer patienters demografi, det palliative forløb og involvering af den praktiserende læge, samt at analysere associationen mellem hjemmedød og involvering af den praktiserende læge (Artikel 3). 4: At analysere associationen mellem den efterladte pårørendes evaluering af forløbet og involvering af den praktiserende læge og dødssted (Artikel 4). 5: At analysere associationen mellem den efterladte pårørendes evaluering af forløbet og dødssted og involvering af hjemmesygeplejersken (Artikel 5). Metoder Vi foretog ti fokusgruppe interviews med deltagelse af pårørende til afdøde cancer patienter og involverede sundhedsprofessionelle. Derefter foretog vi et kombineret register og spørgeskema studie med deltagelse af pårørende til afdøde cancer patienter, praktiserende læger og hjemmesygeplejersker i relation til en konsekutiv population af 559 patienter, døde af cancer.

172


Resultater Interviewene med efterladte pårørende påviste problemer med organisation og samarbejdskultur blandt de professionelle. Ønskede forbedringer blev påpeget i forhold til de professionelles viden, evne til at kommunikere om emnet, attitude og tilgængelighed for patienter, pårørende og andre involverede professionelle. Derudover var det et problem for de pårørende, at de følte, at de blev ”professionaliseret” i forløbet (Artikel 1). Interviews med praktiserende læger, hjemmesygeplejersker og hospitals overlæger beskrev problemer med både organisation af indsatsen og med samarbejdskulturen blandt professionelle. Hovedområder i interviewene var: Opgavefordeling, informationsdeling, tilgængelighed, interprofessionel respekt og personlig kendskab (Artikel 2). Det første spørgeskema-studie viste, at en aktiv involvering med hjemmebesøg af den praktiserende læge og involvering af hjemmesygeplejerskerne var associerede med en højere sandsynlighed for hjemmedød (Artikel 3) Andet og tredje spørgeskema-studie viste, at hjemmedød var positivt associeret med, at den efterladte pårørende evaluerer forløbet som godt, men ingen faktorer angående den praktiserende læges eller hjemmesygeplejerskens indsats blev fundet statistisk signifikante med et positivt evalueret forløb (Artikel 4 og 5). Relationen mellem den pårørende og patienten og mellem den praktiserende læge og den pårørende ser også ud til at være vigtig for et godt palliativt forløb (Artikel 4). Konklusion og perspektiver Ph.d.-studiet viser, at der er brug for optimering af den palliative indsats i patientens hjem på basis niveau, især i relation til den pårørendes rolle, samarbejde mellem professionelle både m.h.t. organisation og samarbejdskultur, samt i relation til primærsektorens professionelles involvering og indsats for at opnå så gode palliative forløb som muligt for patienter og pårørende. Der er desuden brug for uddannelse af både primær og sekunder sektorernes professionelle og yderligere forskning i organiseringen af den palliative indsats.

173

174

Chapter 17. References listed alphabetically by author

CHAPTER 17 REFERENCES LISTED ALPHABETICALLY BY AUTHOR

175


References listed alphabetically by author Aabom B Kragstrup J Vondeling H Bakketeig L S Stovring H. Population-based study of place of death of patients with cancer: implications for GPs. Br J Gen Pract 2005; 55: 684-689. Aabom B Kragstrup J Vondeling H Bakketeig L S Stovring H. Does persistent involvement by the GP improve palliative care at home for end-stage cancer patients? Palliat Med 2006; 20: 507-512. Abom BM Obling N J Rasmussen H Kragstrup J. [Unplanned emergency admission of dying patients. Causes elucidated by focus group interviews with general practitioners]. Ugeskr Laeger 2000; 162: 5768-5771. Addington-Hall J McCarthy M. Regional Study of Care for the Dying: methods and sample characteristics. Palliat.Med. 1995; 9: 27-35. Addington-Hall J McPherson C. After-death interviews with surrogates/bereaved family members: some issues of validity. J Pain Symptom.Manage. 2001; 22: 784-790. Addington-Hall J Walker L ones C arlsen S cCarthy M. A randomised controlled trial of postal versus interviewer administration of a questionnaire measuring satisfaction with, and use of, services received in the year before death. J Epidemiol Community Health 1998; 52: 802-807. Addington-Hall JM MacDonald L D Anderson H R Freeling P. Dying from cancer: the views of bereaved family and friends about the experiences of terminally ill patients. Palliat Med 1991; 5: 207-214. Agar M Currow D C Shelby-James T M Plummer J Sanderson C Abernethy A P. Preference for place of care and place of death in palliative care: are these different questions? Palliat.Med. 2008; 22: 787-795. Akiyama A Numata K Mikami H. Factors enabling home death of the elderly in an institution specializing in home medical care: analysis of apprehension of the bereaved family. Geriatr.Gerontol.Int. 2008; 8: 73-79. Allen J Gay B Crebolder H Heyrman J Svab I Ram P. The European Definitions of the Key Features of the Discipline of General Practice: the role of the GP and core competencies. British Journal of General Practice 2002; 52: 526-527. Amtsrådsforeningen, Sundhedsministeriet, Kommunernes Landsforening, 2001Hjælp til at leve til man dør. Rapport fra arbejdsgruppe om palliativ indsats i amter og kommuner. Andersen TF Madsen M Jørgensen J Mellemkjær M Olsen J H. The Danish National Hospital Register. A valuable source of data for modern health sciences. Dan Med Bull 1999; 46: 263-268. Archer KC Boyle D P. Toward a measure of caregiver satisfaction with hospice social services. Hosp.J 1999; 14: 1-15. Armitage P Berry G Matthews J N S. Statistical Methods in Medical Research. Fourth edition., Oxford: Blackwell Science, 2005. Aspinal F Addington-Hall J Hughes R Higginson I J. Using satisfaction to measure the quality of palliative care: a review of the literature. J Adv.Nurs. 2003; 42: 324-339.

176


Baker R. Development of a questionnaire to assess patients' satisfaction with consultations in general practice. Br J Gen Pract 1990; 40: 487-490. Baker R Wu A W Teno J M Kreling B Damiano A M Rubin H R Roach M J Wenger N S Phillips R S Desbiens N A Connors A F, Jr. Knaus W Lynn J. Family satisfaction with end-of-life care in seriously ill hospitalized adults. J Am.Geriatr.Soc. 2000; 48: S61-S69. Barclay S Todd C Grande G Lipscombe J. Controlling cancer pain in primary care: the prescribing habits and knowledge base of general practitioners. J Pain Symptom.Manage. 2002; 23: 383-392. Barros AJ Hirakata V N. Alternatives for logistic regression in cross-sectional studies: an empirical comparison of models that directly estimate the prevalence ratio. BMC.Med Res.Methodol. 2003; 3: 21. Bee PE Barnes P Luker K A. A systematic review of informal caregivers' needs in providing homebased end-of-life care to people with cancer. J Clin.Nurs. 2008. Bliss J Cowley S While A. Interprofessional working in palliative care in the community: a review of the literature. Journal of interprofessional care 2000; 14: 281-290. Borgsteede SD Deliens L van der Wal G Francke A L Stalman W A van Eijk J T. Interdisciplinary cooperation of GPs in palliative care at home: a nationwide survey in The Netherlands. Scand.J Prim.Health Care 2007; 25: 226-231. Borgsteede SD Graafland-Riedstra C Deliens L Francke A L van Eijk J T Willems D L. Good endof-life care according to patients and their GPs. Br J Gen Pract 2006; 56: 20-26. Brazil K Bedard M Krueger P Abernathy T Lohfeld L Willison K. Service preferences among family caregivers of the terminally ill. J Palliat.Med. 2005a; 8: 69-78. Brazil K Bedard M Willison K. Factors associated with home death for individuals who receive home support services: a retrospective cohort study. BMC Palliat Care 2002; 1: 2-7. Brazil K Howell D Bedard M Krueger P Heidebrecht C. Preferences for place of care and place of death among informal caregivers of the terminally ill. Palliat Med 2005b; 19: 492-499. Burge F Lawson B Johnston G Cummings I. Primary care continuity and location of death for those with cancer. J Palliat Med 2003; 6: 911-918. Burge F McIntyre P Twohig P Cummings I Kaufman D Frager G Pollett A. Palliative care by family physicians in the 1990s. Resilience amid reform. Can.Fam Physician 2001; 47: 1989-1995. Burt J Shipman C White P Addington-Hall J. Roles, service knowledge and priorities in the provision of palliative care: a postal survey of London GPs. Palliat Med 2006; 20: 487-492. Cantwell P Turco S Brenneis C Hanson J Neumann C M Bruera E. Predictors of home death in palliative care cancer patients. J Palliat Care 2000; 16: 23-28. Cartwright A. Balance of care for the dying between hospitals and the community: perceptions of general practitioners, hospital consultants, community nurses and relatives. Br.J Gen.Pract 1991a; 41: 271-274.

177


Cartwright A. The relationship between general practitioners, hospital consultants and community nurses when caring for people in the last year of their lives. Fam Pract 1991b; 8: 350-355. Centeno C Clark D Lynch T Racafort J Praill D De Lima L Greenwood A Flores L A Brasch S Giordano A. Facts and indicators on palliative care development in 52 countries of the WHO European region: results of an EAPC Task Force. Palliat Med 2007; 21: 463-471. Clayton D Hills M. Statistical Models in Epidemiology, Oxford: Oxford University Press, 1993. Crabtree BF Miller W L. Doing qualitative research, Newbury Park: Sage Publications, 1992. Curtis JR Patrick D L Engelberg R A Norris K Asp C Byock I. A measure of the quality of dying and death. Initial validation using after-death interviews with family members. J Pain Symptom Manage. 2002; 24: 17-31. de Fine Olivarius N Hollnagel H Krasnik A Pedersen P A Thorsen H. The Danish National Health Service Register. Dan Med Bull 1997; 44: 449-453. Department of Health. The NHS Cancer Plan. A plan for investment, A plan for reform, London: Department of Health, 2000. Department of Health, UK. End of Life Care Strategy. Promoting high quality care for all adults at the end of life. 2008. Donabedian A. The quality of medical care. Science 1978; 200: 856-864. Donabedian A. The quality of care. How can it be assessed? JAMA 1988; 260: 1743-1748. Donner A Klar N. Design and Analysis of Cluster Randomisation Trials in Health Research, London: Hodder Arnold, 2000. DSAM, Palliativ behandling i almen praksis og lokalområdet. Klinisk vejledning (in press), København, 2004. Evans R Stone D Elwyn G. Organizing palliative care for rural populations: a systematic review of the evidence. Fam Pract 2003; 20: 304-310. Fakhoury W McCarthy M Addington-Hall J. Determinants of informal caregivers' satisfaction with services for dying cancer patients. Soc.Sci.Med 1996; 42: 721-731. Fakhoury WK McCarthy M Addington-Hall J. The effects of the clinical characteristics of dying cancer patients on informal caregivers' satisfaction with palliative care. Palliat.Med. 1997; 11: 107115. Ferris FD Gomez-Batiste X Furst C J Connor S. Implementing quality palliative care. J Pain Symptom.Manage. 2007; 33: 533-541. Field D. Special not different: general practitioners' accounts of their care of dying people. Soc.Sci.Med 1998; 46: 1111-1120. Franks PJ Salisbury C Bosanquet N Wilkinson E K Kite S Naysmith A Higginson I J. The level of need for palliative care: a systematic review of the literature. Palliat.Med 2000; 14: 93-104.

178


Fukui S Fukui N Kawagoe H. Predictors of place of death for Japanese patients with advancedstage malignant disease in home care settings: a nationwide survey. Cancer 2004; 101: 421-429. Giorgi A. Sketch of a psychological phenomenological method. In: Phenomenology and psychological research. Pittsburgh: Duquesne University Press, 1985; 23-85. Gomes B Higginson I J. Factors influencing death at home in terminally ill patients with cancer: systematic review. British Medical Journal 2006; 332: 515-521. Grande GE Addington-Hall J M Todd C J. Place of death and access to home care services: are certain patient groups at a disadvantage? Soc.Sci.Med 1998; 47: 565-579. Grande GE Farquhar M C Barclay S I Todd C J. Caregiver bereavement outcome: relationship with hospice at home, satisfaction with care, and home death. J Palliat Care 2004a; 20: 69-77. Grande GE Farquhar M C Barclay S I Todd C J. Valued aspects of primary palliative care: content analysis of bereaved carers' descriptions. Br.J Gen.Pract 2004b; 54: 772-778. Grbich C Parker D Maddocks I. The emotions and coping strategies of caregivers of family members with a terminal cancer. J Palliat Care 2001; 17: 30-36. Groenvold M Petersen M A Aaronson N K Arraras J I Blazeby J M Bottomley A Fayers P M de G A Hammerlid E Kaasa S Sprangers M A Bjorner J B. The development of the EORTC QLQ-C15PAL: a shortened questionnaire for cancer patients in palliative care. Eur.J.Cancer. 2006; 42: 5564. Groot MM Vernooij-Dassen M J Crul B J Grol R P. General practitioners (GPs) and palliative care: perceived tasks and barriers in daily practice. Palliat Med 2005; 19: 111-118. Guizzo BS Krziminski C O de Oliveira D L. [The software QSR Nvivo 2.0 in qualitative data analysis: a tool for health and human sciences researches]. Rev Gaucha.Enferm. 2003; 24: 53-60. Guthrie B Wyke S. Personal continuity and access in UK general practice: a qualitative study of general practitioners' and patients' perceptions of when and how they matter. BMC Fam Pract 2006; 7: 11. Hales S Zimmermann C Rodin G. The quality of dying and death. Arch.Intern.Med. 2008; 168: 912918. Hanratty B. Palliative care provided by GPs: the carer's viewpoint. Br.J.Gen.Pract. 2000; 50: 653654. Hansen RP Olesen F Sorensen H T Sokolowski I Sondergaard J. Socioeconomic patient characteristics predict delay in cancer diagnosis: a Danish cohort study. BMC Health Serv.Res 2008; 8: 49. Harding R Higginson I J. What is the best way to help caregivers in cancer and palliative care? A systematic literature review of interventions and their effectiveness. Palliat Med 2003; 17: 63-74. Heje HN. Patient evaluation in general practice. Methodological aspects, influence of patient and GP characteristics and the GPs' experiences with the evaluations [thesis], Aarhus: Research Unit and Department of General Practice, Faculty of Health Sciences, University of Aarhus, 2006.

179


Hickman M Drummond N Grimshaw J. A taxonomy of shared care for chronic disease. J Public Health Med 1994; 16: 447-454. Hickman SE Tilden V P Tolle S W. Family reports of dying patients' distress: the adaptation of a research tool to assess global symptom distress in the last week of life. J Pain Symptom.Manage. 2001; 22: 565-574. Higginson IJ Astin P Dolan S. Where do cancer patients die? Ten-year trends in the place of death of cancer patients in England. Palliat Med 1998; 12: 353-363. Hinton J. How reliable are relatives' retrospective reports of terminal illness? patients' and relatives' accounts compared. Soc.Sci.Med. 1996; 43: 1229-1236. Hosmer DW Lemeshow S. Applied logistic regression, second edition, New York: John Wiley & Sons, 2000. Howat A Veitch C Cairns W. A retrospective review of place of death of palliative care patients in regional north Queensland. Palliat.Med. 2007; 21: 41-47. Hudson P. The experience of research participation for family caregivers of palliative care cancer patients. Int.J Palliat.Nurs. 2003; 9: 120-123. Jacoby A Lecouturier J Bradshaw C Lovel T Eccles M. Feasibility of using postal questionnaires to examine carer satisfaction with palliative care: a methodological assessment. South Tyneside MAAG Palliative Care Study Group. Palliat.Med 1999; 13: 285-298. Jansma FF Schure L M de Jong B M. Support requirements for caregivers of patients with palliative cancer. Patient Educ.Couns. 2005; 58: 182-186. Johnson B Christenen L. Chapter 14. Mixed Research: Mixed Method and Mixed Model Research. In: Educational Research Quantitative, Qualitative, and Mixed Approaches. Sage, 2007. Johnson G Scholes K. Exploring Corporate Strategy, Prentice Hall, 1997. Jones RV Hansford J Fiske J. Death from cancer at home: the carers' perspective. British Medical Journal 1993; 306: 249-251. Jørgensen CK Karlsmose B. Validation of automated forms processing. A comparison of Teleform™ with manual data entry. Comput.Biol.Med. 1998; 28: 659-667. Kaasa S De C F. Palliative care research. Eur.J Cancer 2001; 37 Suppl 8: S153-S159. Kaasa S Loge J H. Quality of life in palliative care: principles and practice. Palliat.Med. 2003; 17: 11-20. Kirchhof KT Kehl K A. Recruiting Participants in End-of-Life Research. Am J Hosp Palliat Care 2008; 24: 515-521. Kitzinger J. Qualitative research. Introducing focus groups. British Medical Journal 1995; 311: 299302. Kristjanson LJ. Validity and reliability testing of the FAMCARE Scale: measuring family satisfaction with advanced cancer care. Soc Sci.Med 1993; 36: 693-701.

180


Kvale S. Interviews: An Introduction to Qualitative Research Interviewing, Thousand Oaks: SAGE Publications, 1996. Lecouturier J Jacoby A Bradshaw C Lovel T Eccles M. Lay carers' satisfaction with community palliative care: results of a postal survey. South Tyneside MAAG Palliative Care Study Group. Palliat.Med. 1999; 13: 275-283. Leungh WC. How to design a questionnaire. Student BMJ 2001; 9: 143-145. Lidstone V Butters E Seed P T Sinnott C Beynon T Richards M. Symptoms and concerns amongst cancer outpatients: identifying the need for specialist palliative care. Palliat.Med 2003; 17: 588-595. Mainz J Olesen F. Krav til spørgeskemaer I - validitet, reliabitet og diskrimationsevne. Månedsskr Prakt Lægegern 1994; 72: 1367-1373. Malacrida R Bettelini C M Degrate A Martinez M Badia F Piazza J Vizzardi N Wullschleger R Rapin C H. Reasons for dissatisfaction: a survey of relatives of intensive care patients who died. Crit Care Med. 1998; 26: 1187-1193. Malterud K. Shared understanding of the qualitative research process. Guidelines for the medical researcher. Fam Pract 1993; 10: 201-206. Mayland C Williams E Ellershaw J. How well do current instruments using bereaved relatives' views evaluate care for dying patients? Palliat.Med. 2008; 22: 133-144. McIlfatrick S. Assessing palliative care needs: views of patients, informal carers and healthcare professionals. J Adv.Nurs. 2007; 57: 77-86. McPherson CJ Addington-Hall J M. Judging the quality of care at the end of life: can proxies provide reliable information? Soc Sci.Med 2003; 56: 95-109. Medigovich K Porock D Kristjanson L J Smith M. Predictors of family satisfaction with an Australian palliative home care service: a test of discrepancy theory. J Palliat.Care 1999; 15: 48-56. Michiels E Deschepper R Van Der Kelen G Bernheim J L Mortier F Vander S R Deliens L. The role of general practitioners in continuity of care at the end of life: a qualitative study of terminally ill patients and their next of kin. Palliat.Med. 2007; 21: 409-415. Mikkelsen TH Søndergaard J Sokolowski I Jensen A B Olesen F. Cancer survivors' rehabilitation needs in a primary healthcare context. submitted 2008. Milberg A Strang P. What to do when 'there is nothing more to do'? A study within a salutogenic framework of family members' experience of palliative home care staff. Psychooncology 2007; 16: 741-751. Milne J Oberle K. Enhancing rigor in qualitative description: a case study. J Wound.Ostomy.Continence.Nurs. 2005; 32: 413-420. Mitchell G Cherry M Kennedy R Weeden K Burridge L Clavarino A O'Rourke P Del Mar C. General practitioner, specialist providers case conferences in palliative care--lessons learned from 56 case conferences. Aust.Fam Physician 2005; 34: 389-392.

181


Mitchell GK. How well do general practitioners deliver palliative care? A systematic review. Palliat.Med 2002; 16: 457-464. Miyashita M Morita T Sato K Hirai K Shima Y Uchitomi Y. Good death inventory: a measure for evaluating good death from the bereaved family member's perspective. J.Pain Symptom.Manage. 2008; 35: 486-498. Mok E Chan F Chan V Yeung E. Family experience caring for terminally ill patients with cancer in Hong Kong. Cancer Nurs. 2003; 26: 267-275. Morita T Chihara S Kashiwagi T. A scale to measure satisfaction of bereaved family receiving inpatient palliative care. Palliat Med 2002; 16: 141-150. Morita T Hirai K Sakaguchi Y Maeyama E Tsuneto S Shima Y. Measuring the quality of structure and process in end-of-life care from the bereaved family perspective. J Pain Symptom.Manage. 2004; 27: 492-501. Neergaard MA Olesen F Jensen A B Sondergaard J. Palliative care for cancer patients in a primary health care setting: Bereaved relatives' experience, a qualitative group interview study. BMC Palliat Care 2008; 7: 1. Nickelsen TN. [Data validity and coverage in the Danish National Health Registry. A literature review]. Ugeskrift for læger 2001; 164: 33-37. Nielsen JD Palshof T Mainz J Jensen A B Olesen F. Randomised controlled trial of a shared care programme for newly referred cancer patients: bridging the gap between general practice and hospital. Qual Saf Health Care 2003; 12: 263-272. Nielsen JD Palshof T Olesen F. [Cross-sectorial cooperation regarding cancer patients in a recently started care program. Ideas and themes based on focus group interviews with general practitioners and oncologists]. Ugeskr Laeger 1999; 161: 2074-2078. Nørgaard M Skriver M V Gregersen H Pedersen G Schønheyder H C Sørensen H T. The data quality of haematological malignancy ICD-10 diagnoses in a population-based hospital discharge registry. Eur J Cancer Prev 2005; 14: 201-206. Norman A Sisler J Hack T Harlos M. Family physicians and cancer care. Palliative care patients' perspectives. Can.Fam Physician 2001; 47:2009-12,;%2015-6.: 2009-2012. Norris K Merriman M P Curtis J R Asp C Tuholske L Byock I R. Next of kin perspectives on the experience of end-of-life care in a community setting. J Palliat Med. 2007; 10: 1101-1115. O'Brien RM. A Caution Regarding Rules of Thumb for Variance Inflation Factors. Quality and Quantity 2007; 41: 673-690. O'Connor M Lee-Steere R. General practitioners' attitudes to palliative care: a Western Australian rural perspective. J Palliat Med 2006; 9: 1271-1281. Olesen F Mainz J. Krav til spørgeskemaer II. Hvorledes udarbejdes et spørgeskema? Månedsskr Prakt Lægegern 1994; 72: 1461-1468. Patrick DL Engelberg R A Curtis J R. Evaluating the quality of dying and death. J Pain Symptom.Manage. 2001; 22: 717-726.

182


Pedersen CB Gotzsche H Moller J O Mortensen P B. The Danish Civil Registration System. A cohort of eight million persons. Dan.Med Bull. 2006; 53: 441-449. Pope C Ziebland S Mays N. Qualitative research in health care. Analysing qualitative data. British Medical Journal 320: 114-116. Pope C Ziebland S Mays N. Qualitative research in health care. Analysing qualitative data. British Medical Journal 2000; 320: 114-116. Pritchard P Hughes J. Shared Care The Future Imperative?, London: Royal Society of Medicine Press, 1995. Proot IM Abu-Saad H H Crebolder H F Goldsteen M Luker K A Widdershoven G A. Vulnerability of family caregivers in terminal palliative care at home; balancing between burden and capacity. Scand J Caring.Sci. 2003; 17: 113-121. Protti D. Comparison of information technology in general practice in 10 countries. Healthc.Q. 2007; 10: 107-116. Rasmussen NK. Modelspørgeskema til undersøgelse af sundhed og sygelighed i befolkningen [In Danish], København: DIKE, 1994. Ringdal GI Jordhoy M S Kaasa S. Family satisfaction with end-of-life care for cancer patients in a cluster randomized trial. J Pain Symptom.Manage. 2002; 24: 53-63. Rogers A Karlsen S Addington-Hall J. 'All the services were excellent. It is when the human element comes in that things go wrong': dissatisfaction with hospital care in the last year of life. J Adv.Nurs. 2000; 31: 768-774. Rothman KJ Greenland S. Modern Epidemiology. Second Edition, Philadelphia: Lippicott-Raven Publishers, 1998. Rothman KJ. Epidemiology. An introduction, New York: Oxford University Press, 2002. Saaty RW. The analytic hierarchy process and SWOT analysis-what it is and how it is used. Math Model 1987; 9: 161-178. Sandelowski M. Whatever happened to qualitative description? Res Nurs.Health 2000; 23: 334340. Shipman C Addington-Hall J Barclay S Briggs J Cox I Daniels L Millar D. Providing palliative care in primary care: how satisfied are GPs and district nurses with current out-of-hours arrangements? Br.J Gen.Pract 2000; 50: 477-478. Sjogren P. [Palliative medicine--an independent medical specialty?]. Ugeskrift for læger 2007; 169: 3735. Stange KC Gotler R S. Mixed methods and diverse perspectives. Ann.Fam.Med. 2006; 4: 290-291. Stata Statistical Software: Release 9.0, College Station, TX: StataCorp LP, 2005.

183


Steinhauser KE Christakis N A Clipp E C McNeilly M McIntyre L Tulsky J A. Factors considered important at the end of life by patients, family, physicians, and other care providers. JAMA 2000a; 284: 2476-2482. Steinhauser KE Clipp E C McNeilly M Christakis N A McIntyre L M Tulsky J A. In search of a good death: observations of patients, families, and providers. Ann.Intern.Med. 2000b; 132: 825-832. Sullivan-Bolyai S Bova C Harper D. Developing and refining interventions in persons with health disparities: the use of qualitative description. Nurs.Outlook 2005; 53: 127-133. Sulmasy DP McIlvane J M Pasley P M Rahn M. A scale for measuring patient perceptions of the quality of end-of-life care and satisfaction with treatment: the reliability and validity of QUEST. J Pain Symptom Manage. 2002; 23: 458-470. Sykes NP Pearson S E Chell S. Quality of care of the terminally ill: the carer's perspective. Palliative Medicine 1992; 6: 227-236. Tang ST McCorkle R. Determinants of place of death for terminal cancer patients. Cancer Invest 2001; 19: 165-180. Teno JM Clarridge B Casey V Edgman-Levitan S Fowler J. Validation of Toolkit After-Death Bereaved Family Member Interview. J Pain Symptom Manage. 2001; 22: 752-758. Tetsche M Nørgaard M Skriver M V Andersen E S Lash T L Sørensen H T. Accuracy of ovarian cancer ICD-10 diagnosis in a Danish population-based hospital discharge registry. Eur J Gynaecol Oncol 2005; 26: 266-270. The Danish National Board of Health. The Danish Register of Causes of Death [In Danish], 2008. The National Board of Health. National Cancer Plan II • Denmark National Board of Health recommendations for improving cancer healthcare services, Copenhagen: The National Board of Health, 2005. The National council for palliative care, http:, and www.ncpc.org.uk/palliative_care.html. Palliative care explained. 2007. Thomas C. The place of death of cancer patients: can qualitative data add to known factors? Soc.Sci.Med 2005; 60: 2597-2607. Thomas C Morris S M Clark D. Place of death: preferences among cancer patients and their carers. Soc Sci Med 2004; 58: 2431-2444. Thomas K Noble B. Improving the delivery of palliative care in general practice: an evaluation of the first phase of the Gold Standards Framework. Palliat Med 2007; 21: 49-53. Thompson ML Myers J E Kriebel D. Prevalence odds ratio or prevalence ratio in the analysis of cross sectional data: what is to be done? Occup Environ Med 1998; 55: 272-277. van der Heide A de Vogel-Voogt E Visser A P van der Rijt C C van der Maas P J. Dying at home or in an institution: perspectives of Dutch physicians and bereaved relatives. Support.Care Cancer 2007; 15: 1413-1421.

184


Vedsted P Sokolowski I Heje H N. Data quality and confirmatory factor analysis of the Danish EUROPEP questionnaire on patient evaluation of general practice. Scand J Prim Health Care. 2008; 26: 174-180. von B, I Eliasson G Sarvimaki A Mattsson B Hjortdahl P. Patients' views on interpersonal continuity in primary care: a sense of security based on four core foundations. Fam Pract 2006; 23: 210-219. Wakefield MA Beilby J Ashby M A. General practitioners and palliative care. Palliat.Med 1993; 7: 117-126. Weibull A Olesen F Neergaard M A. Caregivers' active role in palliative home care - to encourage or to dissuade? A qualitative descriptive study. BMC Palliat Care 2008; 7: 15. WHO. Cancer Pain Relief and Palliative Care. Report of a WHO Expert Committee, Geneva: World Health Organization, 1990. Wilkinson HJ. Assessment of patient satisfaction and hospice: a review and an investigation. Hosp.J 1986; 2: 69-94. World Health Organization. Pain relief and palliative care. In: National cancer control programmes, policies and managerial guidelines. Geneva: 2002; 86-91. Zou G. A modified poisson regression approach to prospective studies with binary data. Am.J Epidemiol 2004; 159: 702-706.

185

186

APPENDICES

Appendix A: Interview guide, final, relatives (In Danish)

Interview guide, pårørende I er jo her i dag, fordi I alle har mistet en af jeres kære. Jeg synes, vi skal starte med en præsentationsrunde, hvor I siger jeres navn og fortæller kort om, hvem det er, I har mistet… Vi skal i dag snakke mest om den sidste tid, hvor I vidste, at der ikke var mere at gøre. Det er den del, der foregik hjemme eller overgangen fra hospital til hjemmet, vi er mest interesserede i. Vil I fortælle lidt om den tid? … A)Den praktiserende læges rolle 1. Hvordan var indsatsen fra den praktiserende læges side? (Hvad gik godt? Hvad gik skidt? Hvorfor? Hvad var svært? Hvordan kunne det blive bedre?) 2. Hvordan ønsker I den praktiserende læge skulle have støttet jer bedre? 3. Hvilke forventninger havde I til ham/hende? 4. Hvilken rolle synes I, den praktiserende læge skulle have i sådanne forløb? 5. Føler du, at du som pårørende fik støtte nok? Flg. skal belyses:

Viden nok? Ressourcer nok? (tid, psykologiske) Tilgængelighed? Kontinuitet? Barrierer? Sociale foranstaltninger (hjælpemidler, pension, sygemelding, plejeorlov)

B) Overførselsforretning 1. Hvordan oplevede I overgangen fra behandling på hospitalet til lindrende pleje hjemme? (Afslutningen på hospitalet, aflevering til praktiserende læger / hospital, kontinuitet, shared care?) 2. Hvordan var samarbejdet mellem den praktiserende læge og hospitalsafdelingen? (Hvad var svært? Hvordan kunne det blive bedre?) C) Den interpersonelle relation 1. Hvordan var samarbejdet mellem den praktiserende læge og hjemmeplejen? (Hvad var svært? Hvordan kunne det blive bedre?) 2. Hvordan var samarbejdet mellem den praktiserende læge og det palliative team eller sygehus afd.? (Hvad var svært? Hvordan kunne det blive bedre?) D) Dødssted 1. Hvor døde din kære? 2. Var det der, du ønskede, at han/hun skulle dø? 3. Hvor ønskede han/hun at dø? E) Organisation af diagnostik, behandling og pleje 1. Hvordan synes I, at den lindrende behandling bør organiseres? - Hvordan skal opgaverne fordeles? - Hvordan skal samarbejdet være? F) Afsluttende spørgsmål Nu har vi snakket en del om den sidste tid med Jeres kære. - Er der noget I føler, vi har glemt? - Er der særlige budskaber vi skal have med hjem? RESUME Mange tak for hjælpen…

Appendix B: Interview guide, final, GPs (In Danish)

Interview guide, Praktiserende læger Inden vi starter med at snakke om palliation, må I gerne sige, hvem I er, hvilket årstal I er født, hvor I har praksis, og hvilken type praksis. A) Den interpersonelle relation 1. Hvis I tænker på de palliative forløb, I har haft, hvordan fungerer samarbejdet med de andre involverede faggrupper? (Hvad var svært? Hvordan kunne det blive bedre?) 2. Hvilke forventninger har I til samarbejdsparterne? (hjemmesygeplejerskerne, hospitalslægerne) 3. Hvordan fungerer kommunikationsvejene? (tilgængelighed, hemmelige telefonnumre) 4. Hvordan fungerede ”overførselsforretningen”? (Hvad gik godt/skidt? Hvordan kunne det blive bedre? ) (Afslutningen på hospitalet, aflevering fra hospitalet (=Overførselsforretning), kontinuitet, shared care?) 5. Hvordan blev I involveret i forløbet? (Hvordan gives ansvaret til Jer? Er I selv aktive?) 6. Hvordan bruger I det palliative team? (Hvad er godt? Hvad er skidt? Hvordan kan det blive bedre? Hvordan sikrer vi, at patienten får rette behandling?) 7. Hvordan ser I jeres rolle i forhold til de pårørende? (Hvem har ansvaret for dem?) B) Oplevelse Hvad kan du forestille dig, at kolleger kan have imod at tage sig af døende patienter? - Hvilke barrierer kan forekomme? (kulturelle, organisatoriske, personlige, tidsmæssige, faglige, økonomiske, miste patienter, prioritering) C) Den sundhedsfaglige ydelse 1. Føler I, at I har viden nok til at tage sig af de døende og deres pårørende i eget hjem? - Hvor søger man oplysninger? - Hvad manglede I at vide? 2. Hvad karakteriserer en god praktiserende læge, set ud fra patient og pårørendes synsvinkel? 3. Er der ressourcer nok i almen praksis? (Og hvordan opnås det, der mangler?) - Tidsmæssige - Faglige - Personlige - Økonomiske D) Organisation af diagnostik, behandling og pleje 1. Hvem skal gøre hvad? - Hvad synes du din rolle er? - Hvad synes du hjemmeplejens rolle er? - Hvad synes du hospitalets rolle er? 2. Hvad ligger der i ordet ”tovholder”, og hvem skal det være?(behandleren? administratoren? koordinatoren?) 3. Hvordan sikrer man, at hver part har lige præcis den information, som de har brug for? ( Fælles IT-net, fælles journal, hotline, formel konferencetid, fælles undervisning) 4. Hvordan kan man gennemføre forbedringer af indsatsen? - Organisation: - Struktur - Individ: - Viden(Uddannelse) - Holdninger - Ressourcer E) Afsluttende spørgsmål Nu har vi snakket en del om palliation. - Er der noget I føler, vi har glemt? - Er der særlige budskaber vi skal have med hjem? RESUME Mange tak for hjælpen…

Appendix C: Interview guide, final, mixed professionals (In Danish)

Interview guide, blandet sundhedspersonale Inden vi starter med at snakke om palliation, må I gerne sige, hvem I er, hvilket årstal I er født, hvor I arbejder, og hvad I laver. A)Vellykket/ikke vellykket forløb 1. Hvis I tænker på det sidste forløb med et palliativt forløb i hjemmet, I har været involveret i, hvad gik så godt i det forløb? (Hvad gik skidt? Hvorfor? Hvad var svært? Hvordan kunne det blive bedre?) 2. Hvordan fungerede ”overførselsforretningen”? (Hvad gik godt/skidt? Hvordan kunne det blive bedre?) (Afslutningen på hospitalet, aflevering til praktiserende læger/hospital (= overførselsforretning), kontinuitet, shared care?) B) Den interpersonelle relation 1. Hvordan var samarbejdet med de andre involverede faggrupper? (Hvad var svært? Hvordan kunne det blive bedre?) 2. Kender I nok til hinandens kompetencer og muligheder? (Fordomme? / Oplevede?) 3. Hvilke ønsker om relationer og forventninger har I til hinanden? 4. Hvordan ser I jeres rolle i forhold til de pårørende? (Hvem har ansvaret for dem?) 5. Hvordan bruger I det palliative team? (Hvad er godt? Hvad er skidt? Hvordan kan det blive bedre? Hvordan sikrer vi, at patienten får rette behandling?) C) Oplevelse Hvad kan du forestille dig, at kolleger kan have imod at tage sig af døende patienter? - Hvilke barrierer kan forekomme? (kulturelle, organisatoriske, personlige, tidsmæssige, faglige, Økonomiske, miste patienter, prioritering) D) Den sundhedsfaglige ydelse 1. Føler du, at der er viden nok i almen praksis til at tage sig af de døende og deres pårørende i eget hjem? - Hvordan kan man finde ud af, hvilken viden man mangler? - Hvor søger man oplysninger? - Hvad manglede du at vide? Til de praktiserende læger og hjemmesygeplejerskerne. 2. Hvad karakteriserer en god læge, set ud fra patientens og pårørendes synsvinkel? 3. Hvad karakteriserer en god sygeplejerske, set ud fra patientens og pårørendes synsvinkel? 4. Er der ressourcer nok i almen praksis? (Og hvordan opnås det, der mangler?) - Tidsmæssige - Faglige - Personlige - Økonomiske E) Organisation af diagnostik, behandling og pleje 1. Hvem skal gøre hvad? - Hvad synes du din rolle er? Hvad synes du ……..s rolle er? 2. Hvad ligger der i ordet ”tovholder”, og hvem skal det være? 3. Hvordan sikrer man, at hver part har lige præcis den information, som de har brug for? ( Fælles IT-net, fælles journal, hotline, formel konferencetid, fælles undervisning) 4. Hvordan kan man gennemføre forbedringer af indsatsen? - Organisation: - Struktur - Individ: - Viden(Uddannelse) - Holdninger - Ressourcer F) Afsluttende spørgsmål Nu har vi snakket en del om palliation. - Er der noget I føler, vi har glemt? - Er der særlige budskaber vi skal have med hjem? RESUME Mange tak for hjælpen…

Appendix D: Interview guide, final, community nurses (In Danish)

Interview guide, hjemmesygeplejersker Inden vi starter med at snakke om palliation, må I gerne sige, hvem I er, og hvor I arbejder. A)Vellykket/ikke vellykket forløb 1. Hvis I tænker på det sidste forløb med et palliativt forløb i hjemmet, I har været involveret i, hvad gik så godt i det forløb? (Hvad gik skidt? Hvorfor? Hvad var svært? Hvordan kunne det blive bedre?) C) Den interpersonelle relation 1. Hvordan var samarbejdet med den praktiserende læge? (Hvad var svært? Hvordan kunne det blive bedre?) 2. Kender du nok til lægens kompetencer og muligheder? (fordomme? / oplevede?) 3. Hvilke ønsker om relationer har du til lægen / andre samarbejdspartnere? 4. Hvordan fungerede ”overførselsforretningen”? (Hvad gik godt/skidt? Hvordan kunne det blive bedre?) (Afslutningen på hospitalet, aflevering til primærsektoren (=overførselsforretning), kontinuitet, shared care?) 5. Hvordan ser du jeres rolle i forhold til de pårørende? (Hvem har ansvaret for dem?) 6. Hvordan bruger du det palliative team? (Hvad er godt? Hvad er skidt? Hvordan kan det blive bedre? Hvordan sikrer vi rette patient får rette behandling?) 7. Flere praktiserende læger har fortalt, at de kan føle sig ”skubbet til side” af de Palliative Teams. Er det en følelse, du også kan have? D) Den sundhedsfaglige ydelse 1. Føler du, at der er viden nok i almen praksis til at tage sig af de døende og deres pårørende i eget hjem? 2. Er der ressourcer nok i almen praksis? (Og hvordan opnås det, der mangler?) (Tidsmæssige, Faglige, Personlige, Økonomiske) 3. Hvad kan du forestille dig, at praktiserende læger kan have imod at tage sig af døende patienter? - Hvilke barrierer kan forekomme? (kulturelle, organisatoriske, personlige, tidsmæssige, faglige, Økonomiske, miste patienter, prioritering) 4. Hvilke forventninger har du til lægen / andre samarbejdspartnere? 5. Hvad karakteriserer en god læge, set ud fra patientens og pårørendes synsvinkel? 6. Hvad karakteriserer en god sygeplejerske, set ud fra patientens og pårørendes synsvinkel?

E) Organisation af diagnostik, behandling og pleje 1. Hvem skal gøre hvad? - Hvad synes, du din rolle er? Hvad synes, du lægens rolle er? 2. Hvad ligger der i ordet ”tovholder”, og hvem skal det være? 3. Hvordan sikrer man, at hver part har lige præcis den information, som de har brug for? (fælles IT-net, fælles journal, hotline, formel konferencetid, fælles undervisning) 4. Hvordan kan man gennemføre forbedringer af indsatsen? - Organisation: - Struktur - Individ: - Viden(Uddannelse) - Holdninger - Ressourcer F) Afsluttende spørgsmål Nu har vi snakket en del om palliation. - Er der noget, I føler, vi har glemt? - Er der særlige budskaber, vi skal have med hjem? RESUME Mange tak for hjælpen…

Appendix E. Topic guide themes in interviews with professionals Themes The health professionals‟ interpersonal relations

Issues Co-operation with other health professionals

Examples of questions How did the co-operation with other health professionals function in palliative pathways you have been involved with?* Expectations to other health What do you expect from the other professionals health professionals? Availability of other health How did you get in touch with other professionals health professionals in palliative pathways you have been involved with?* Transferring patients between How was the discharge from the hospitals and the patients‟ homes hospital in palliative pathways you have been involved with?* How was the possible admission to the hospital in the palliative pathway in palliative pathways you have been involved with?* How was the co-operation in these situations between the sectors?* Experiences of What barriers / facilitators do you professionals‟ see among groups of other and barriers own health professionals to participating in palliative care? How could it be improved? The effort of health Knowledge Do you have enough knowledge professionals about palliative care? How could it be improved? Role of GP What are the GP‟s roles in palliative care Role of CN What are the CN‟s roles in palliative care Role of discharging hospital staff What are the hospital‟s roles in palliative care Organisation of The team leader role Who was the team leader in palliative treatment palliative pathways you have been and care involved with?* Shared information How do you ensure that all Implementation of improvements involved professionals get the information needed? ¨ How can you implement the suggestions you have had during this interview?(Both in relation to organisation and individual improvements) Ideal organisation How do you think palliative care ideally should be organised? How could it be obtained? *Adding the questions: What was good / bad? How could it be improved?

Year Country

1993 Can

2000 USA

Scale/ Questionnaire

FAMCAREScale 1 (3) 2 (4) 3 (5)

SUPPORT(6)

Two 4-item scales, one measuring satisfaction with patient comfort and the other measuring satisfaction with communication and decision-making

To interview relatives about satisfaction with care after the patient‟s death n=767, 78%

2

Validating, study n= 183/113, 36% /49% 3 n=200, 28%

1

Interviewer administrated or postal study 20-item Likert-type scale (5 possible answers): How content are you with: (Some questions are shortened. red) I. The patient‟s pain relief 2. Information about prognosis 3. Answers from health professionals 4. Information given about side effects 5. Referrals to specialists 6. Availability of a hospital bed 7. Family conferences held 8. Speed of symptom treatment 9. Doctor‟s attention to patient 10. Test / treatment performance 11. Availability of doctors to the family 12. Availability of nurses to the family 13. Coordination of care 14. Time required to make a diagnosis 15. The way the family is included 16. Information given about pain 17. Information given about tests 18. Doctor‟s thoroughness 19. Doctor‟s follow up 20. Availability of the doctor

Items

To measure family satisfaction with advanced cancer care, i.e. the degree to which the family members are content with health care provider behaviour

Aim Number of participants Response rate

Not developed for home care

Not developed for postal questionnaire

Not an overall question of satisfaction with the entire pathway

Too may items

Not differentiation between primary and secondary health care

Not designed to a retrospective setting

Not designed to palliative pathways 2 (In ref , item 14 was excluded)

Problems in relation to our study

Appendix F. Scales and questionnaires developed for relatives in palliative care (1;2)

Year Country

2001 USA


QODD (7;8)

n=252 23%

To measure the family members‟ retrospective ratings of the quality of the dying experience for the decedent‟s last seven days

Aim Number of participants Response rate Interviewer administered 31-items: How would you rate this aspect of dying experience? (Likert 0-10) (Some questions are shortened. red) 1. Having pain under control 2. Having control of event 3. Being able to feed oneself 4. Having control of bladder, bowels 5. Being able to breathe comfortably 6. Having energy to do things 7. Spend time with your children 8. - with your friends and other family 9. Spend time alone 10. Be touched and hugged 11. Say goodbye to your loved ones 12. Have the means to end your life 13. Discuss wishes for end-of-life care 14. Feel at peace with dying. 15. Avoid worry about your loved ones 16. Be unafraid of dying. 17. Find meaning and purpose in life 18. Die with dignity and respect 19. Laugh and smile 20. Avoid dialysis or ventilation 21. Location of death 22. Die with loved ones present 23. State at moment of death 24. Visit from spiritual advisor 25. Spiritual service or ceremony 26. Have health care costs provided 27. Funeral arrangements in order 28. Spend time with spouse, partner 29. Spend time with pets 30. Clear up bad feelings 31. Attend important events

Items

Not a primary health care focus


Only about the last 7 days of the palliative period



Year Country

1995 2000 UK

1999 UK

1997 UK


VOICES 1 (9) 2 (10) 3 (11) 4 (12)

(No name) 1 (13) 2 (14)

(No name) (15;16)

Administered via interview Satisfaction variables: 1. Do you feel the help and care deceased got from district nurses was excellent, good, fair, or poor? 2. Altogether, what do you think about care deceased got from the GP in the last year of life. Would you describe the care as: excellent, good, fair or poor? 3. Altogether, what do you feel about care deceased got from the doctors while he/she was in hospital. Would you describe it as excellent, good, fair or poor?

To measure bereaved relatives‟ satisfaction with service quality n=1858, 69%

2

n=67, 60% n=156, 44%

1

Postal questionnaire (Likert scale) 86 questions Main topics: Information Care received in the community Care at hospitals and hospices Provision of equipment Carers‟ post bereavement health and support

158 questions 1 Telephone interviews 2 Interview versus postal survey 3 Postal survey

Items

To measure satisfaction of bereaved lay caregivers with the care received in the community by those dying of cancer

2

n=3696, 69% n=69/161, 56%/52% 3 n=139, 59% 4 , n=229 53%

1

To measure bereaved relatives‟ satisfaction with palliative care provided by their GP





Too many questions

Too many questions

A high weight on symptom manage

Only satisfaction with GPs effort


Year Country

1999 USA

1998 Swiss

1986 USA

2001 USA


(No name) (17)

(No name) (18)

(No name) (19)

TIME (Toolkit AfterDeath Bereaved Family Member Interview) (20)

n=156, 54%

To measure quality of care at a health care organisation (hospital, hospice, nursing home) at the end of life from the perspective of bereaved relatives

44%, n=266

To assess patients‟ and relatives‟ concern asking bereaved relatives

32%, n=390

To identify information given and perceptions of care quality in the view of bereaved relatives to patients who had died in an intensive care unit (ICU)

80 %, n=55

To measure hospice patients‟ primary care givers‟ ratings of social services From the perspectives of the bereaved relatives


Telephone-administered structured interview (A mix of question types)

Postal questionnaire (Likert scale)

Postal questionnaire (Open ended and Likert scale)

Structured telephone interview (Likert scale)

Items

Not developed for home care

Not developed for postal surveys

Developed for hospice care

Developed to ICU care


Developed for a hospice setting


Japan 2008

Good death inventory (21)

Japan 2004

Japan 2002

CES (23)

Sat-Fam-IPC (24)

Family MSAS-GDI (22)

Year Country


n=850, 64%

To measure satisfaction with an in-patient palliative care services

n=1225, 37%

To evaluate quality of care structure and process for endof-life care; To identify areas of care requiring improvement

n=103, 51%

To assess level of symptom distress experienced by a dying patient in the last week of life

n=189, 57%

To measure the evaluation of a good death from the bereaved relative‟s perspective


Postal survey 34 items Likert scale

Postal survey 28 items Likert scale

Interviewer administrated 11 items Likert scale

Postal survey Likert scale (1-7) 18 domains – 54 questions

Items

Developed to satisfaction with an in-patient palliative care services

Not developed for a home care setting

Focuses on symptom distress

Not developed for a home care setting


References of Appendix F (1) Aspinal F, Addington-Hall J, Hughes R, Higginson IJ. Using satisfaction to measure the quality of palliative care: a review of the literature. J Adv Nurs 2003 May;42(4):324-39. (2) Mayland C, Williams E, Ellershaw J. How well do current instruments using bereaved relatives' views evaluate care for dying patients? Palliat Med 2008 Mar;22(2):133-44. (3) Kristjanson LJ. Validity and reliability testing of the FAMCARE Scale: measuring family satisfaction with advanced cancer care. Soc Sci Med 1993 Mar;36(5):693-701. (4) Ringdal GI, Jordhoy MS, Kaasa S. Family satisfaction with end-of-life care for cancer patients in a cluster randomized trial. J Pain Symptom Manage 2002 Jul;24(1):53-63. (5) Medigovich K, Porock D, Kristjanson LJ, Smith M. Predictors of family satisfaction with an Australian palliative home care service: a test of discrepancy theory. J Palliat Care 1999;15(4):48-56. (6) Baker R, Wu AW, Teno JM, Kreling B, Damiano AM, Rubin HR, et al. Family satisfaction with end-of-life care in seriously ill hospitalized adults. J Am Geriatr Soc 2000 May;48(5 Suppl):S61-S69. (7) Patrick DL, Engelberg RA, Curtis JR. Evaluating the quality of dying and death. J Pain Symptom Manage 2001 Sep;22(3):717-26. (8) Curtis JR, Patrick DL, Engelberg RA, Norris K, Asp C, Byock I. A measure of the quality of dying and death. Initial validation using after-death interviews with family members. J Pain Symptom Manage 2002 Jul;24(1):17-31. (9) Addington-Hall J, McCarthy M. Regional Study of Care for the Dying: methods and sample characteristics. Palliat Med 1995 Jan;9(1):27-35. (10) Addington-Hall J, Walker L, ones C, arlsen S, cCarthy M. A randomised controlled trial of postal versus interviewer administration of a questionnaire measuring satisfaction with, and use of, services received in the year before death. J Epidemiol Community Health 1998;52:802-7. (11) Hanratty B. Palliative care provided by GPs: the carer's viewpoint. Br J Gen Pract 2000 Aug;50(457):653-4. (12) Rogers A, Karlsen S, Addington-Hall J. 'All the services were excellent. It is when the human element comes in that things go wrong': dissatisfaction with hospital care in the last year of life. J Adv Nurs 2000 Apr;31(4):768-74. (13) Jacoby A, Lecouturier J, Bradshaw C, Lovel T, Eccles M. Feasibility of using postal questionnaires to examine carer satisfaction with palliative care: a methodological assessment. South Tyneside MAAG Palliative Care Study Group. Palliat Med 1999 Jul;13(4):285-98. (14) Lecouturier J, Jacoby A, Bradshaw C, Lovel T, Eccles M. Lay carers' satisfaction with community palliative care: results of a postal survey. South Tyneside MAAG Palliative Care Study Group. Palliat Med 1999 Jul;13(4):275-83.

(15) Fakhoury WK, McCarthy M, Addington-Hall J. The effects of the clinical characteristics of dying cancer patients on informal caregivers' satisfaction with palliative care. Palliat Med 1997 Mar;11(2):107-15. (16) Fakhoury W, McCarthy M, Addington-Hall J. Determinants of informal caregivers' satisfaction with services for dying cancer patients. Soc Sci Med 1996 Mar;42(5):721-31. (17) Archer KC, Boyle DP. Toward a measure of caregiver satisfaction with hospice social services. Hosp J 1999;14(2):1-15. (18) Malacrida R, Bettelini CM, Degrate A, Martinez M, Badia F, Piazza J, et al. Reasons for dissatisfaction: a survey of relatives of intensive care patients who died. Crit Care Med 1998 Jul;26(7):1187-93. (19) Wilkinson HJ. Assessment of patient satisfaction and hospice: a review and an investigation. Hosp J 1986;2(4):69-94. (20) Teno JM, Clarridge B, Casey V, Edgman-Levitan S, Fowler J. Validation of Toolkit AfterDeath Bereaved Family Member Interview. J Pain Symptom Manage 2001 Sep;22(3):7528. (21) Miyashita M, Morita T, Sato K, Hirai K, Shima Y, Uchitomi Y. Good death inventory: a measure for evaluating good death from the bereaved family member's perspective. J Pain Symptom Manage 2008 May;35(5):486-98. (22) Hickman SE, Tilden VP, Tolle SW. Family reports of dying patients' distress: the adaptation of a research tool to assess global symptom distress in the last week of life. J Pain Symptom Manage 2001 Jul;22(1):565-74. (23) Morita T, Hirai K, Sakaguchi Y, Maeyama E, Tsuneto S, Shima Y. Measuring the quality of structure and process in end-of-life care from the bereaved family perspective. J Pain Symptom Manage 2004 Jun;27(6):492-501. (24) Morita T, Chihara S, Kashiwagi T. A scale to measure satisfaction of bereaved family receiving inpatient palliative care. Palliat Med 2002 Mar;16(2):141-50.

Appendix G. Themes in questionnaires GP questionnaire Validation questions about the cause of death (72-item questionnaire) Information about closest relative and the community nurse / centre involved. GP knowledge of patient before the palliative pathway Length of the palliative period Type of contact Patient‟s wish for and actual place of death Care for the relative during the palliative period and after bereavement Cooperation with CNs, hospital doctors, palliative specialist team Especially when discharging the patient from the hospital When the patient is at home Evaluation the primary care sector‟s effort Evaluation of GP‟s own effort Overall view on palliative care in primary care Demographic data of the GP‟s practice Open question on comments to the questionnaire CN questionnaire (46-item questionnaire)

Questionnaire for bereaved relatives (65-item questionnaire)

Validation questions about the cause of death Information about closest relative and the CN / centre involved. CN knowledge of patient before the palliative pathway Type of contact Patient‟s wish for and actual place of death Care for the relative during the palliative period and after bereavement Cooperation with GPs, hospital doctors, palliative specialist team When discharging the patient from the hospital When the patient is at home Evaluation the primary care sector‟s effort Evaluation of CN‟s own effort Overall view on palliative care in primary care Demographic data of the CN Open question on comments to the questionnaire Demographic data of deceased Length of the palliative period Type of contact to professionals Patient‟s wish for and actual place of death Care for the relative during the palliative period and after bereavement Cooperation among GPs, CNs, hospital doctors, palliative specialist team When discharging the patient from the hospital When the patient is at home Evaluation the primary care sector‟s effort Demographic data of the relative Open question on comments to the questionnaire

Appendix H. Variables with data retrieved from official registers used in the analysis in Papers 3 – 5. Used in Data used for Variable Data from Original data paper analysis Patient’s age The county 3, 4 and 5 Civil registration Dichotomized into 18at time of hospital discharge numbers 65, 66+ death registry Patient’s The county 3, 4 and 5 Civil registration Male, female gender hospital discharge numbers registry Patient’s The county 3, 4 and 5 ICD10 diagnosis Categorised into lung, primary hospital discharge colo-rectal, breast, cancer registry prostate, other diagnosis Place of death Register of Causes 3, 4 and 5 Home, nursing In Paper 3 of Death home, hospital dichotomized into or hospice and home, institutional. death in other In Papers 4 and 5 places categorised into home, nursing home, institutional GP home Health service 3 and 4 Continuous Dichotomized visits in the registry variable into:No/yes last 3 months

Appendix I. Variables from questionnaires used in the analysis in Paper 3 – 5, with coherent questions. Variable

Data from

Question

Patient‟s marital status

GP questionnaires

Patient having children

GP questionnaires

GP involvement in palliation

GP questionnaires

GP‟s knowledge prior to palliative period

GP questionnaires

Unplanned home visits by GP

GP questionnaires

GP gave private number to patient to use in out-ofoffice hours

GP questionnaires

GP had made a plan with the patient for whom to contact in outof-office hours

GP questionnaires

Hvad var patientens ægteskabelige status? - Ugift, ikke tidligere gift - Gift / Samboende - Enlig (skilt / separeret) - Enlig (enke / enkemand) - Ugift, men samboende med anden pårørende - Ved ikke (GP question 8) Havde patienten børn? - Ja, deriblandt hjemmeboende børn - Ja, men ikke hjemmeboende børn - Nej - Ved ikke (GP question 9) Havde du /din praksis nogen form for kontakt med patienten eller familien i det palliative forløb? Ja Nej (GP Question 13) Hvordan vil du karakterisere dit / din praksis' faglige kendskab til patienten før det palliative forløb? - Meget godt - Godt - Nogenlunde - Ikke særlig godt - Kendte ikke patienten (GP question 6) Tog du initiativ til uopfordrede hjemmebesøg hos denne patient, d.v.s. besøg i hjemmet, hvor du foreslog / tog initiativ til besøget? Ja Nej (GP question 16) Har du opgivet dit private telefonnummer til patient og / eller pårørende i det palliative forløb? Ja, det gør jeg generelt altid Ja, det gør jeg i enkelte udvalgte forløb Nej (GP question 17) Aftalte du med patienten og pårørende, hvem der skulle kontaktes, når du / dine kompagnoner ikke var tilgængelige? Nej Ja (GP question 18)

Used in paper 3 and 4

Dichotomized into: Single Married or cohabiting

3 and 4

No Yes at home Yes not at home

3 and 4

No Yes

3 and 4

Dichotomized into Poor (1,2 on a 1-5-point scale) Good (3, 4 and 5

3 and 4

No Yes

3 and 4

No Yes

3 and 4

No Yes

GP had contact with relatives

GP questionnaires

Community nurse involvement

GP questionnaires

Specialist team involvement

GP and CN questionnaires

Duration of palliative period in total


Duration of palliative period at home


Evaluation of the palliative pathway

Questionnaires for relatives

Relative‟s age at time of filling in questionnaire


Havde du kontakt med pårørende i forløbet? Ja Nej, af hvilken årsag: Der var ingen pårørende De pårørende ønskede ingen kontakt Anden årsag, hvilken (GP question 26) Hvis hjemmeplejen slet ikke var involveret i forløbet, gå til spørgsmål 53 side 23 If the questionnaire were filled in but question number 44 to 52 all were blank, we took this as a no. (GP questionnaire page 20 ) Hvis der ikke har været et palliativt team involveret i forløbet, beder vi dig gå til side 20 øverst. If the questionnaire were filled in but question number 40 to 43 all were blank, we took this as a no. (GP questionnaire page 17 ) Hvor lang tid skønner du, at det palliative forløb varede i alt (fra det tidspunkt, hvor patientens væsentligste behov var palliativt, og al kurativ behandling var stoppet, og frem til dødens indtræden)? Antal uger(< 1 uge skrives som 0) (GP question 11) (Extra question CNs) Hvor mange af de uger, du angav ovenfor, skønner du, patienten sammenlagt var i hjemmet (enten i patientens eget hjem, på fast plejehjemsplads eller i pårørendes hjem, hvis det var der, plejen foregik)? Antal uger(< 1 uge skrives som 0) (GP question 12) (Extra question CNs) Hvordan vil du karakterisere den samlede tid hjemmet, hvor afdøde var døende, i forhold til, hvordan du mener, det burde have været? Rigtig godt Godt Rimeligt Dårligt Rigtig dårligt (Question 45 to relatives) Hvilket årstal er du født? (Question 51 to relatives)

3 and 4

No Yes

3 and 4

No Yes

3, 4 and 5

No Yes

3

Continuous variable

3, 4 and 5

Continuous variable

4 and 5

Dichotomized into „Not successful‟ („Fairly well‟, „Bad‟, „Very bad‟) and „Successful‟ („Very well‟, „Well‟)

4 and 5

Dichotomized into 18-65 66+

Er du kvinde eller mand? (Question 50 to relatives) Hvilken relation har du til den afdøde? (Sæt kun et kryds) Ægtefælle/samlever Kæreste, men ikke samboende Datter / Søn Søster / Bror Forælder Ven Andet Question 52 to relatives) Boede du i sammen med den afdøde i den sidste tid? (Question 53 to relatives) Har du fuldført en erhvervsuddannelse ? (Hvis du har gennemført flere erhvervsuddannelser, skal du kun oplyse den længst varende) Ja, et eller flere kortere kurser (specialarbejderkurser, arbejdsmarkedskurser) Ja, faglært inden for håndværk, handel, kontor mv. Ja, kort videregående uddannelse under 3 år Ja, mellemlang videregående uddannelse 3-4 år Ja, lang videregående uddannelse mere end 4 år Andet: Nej (Question 57 to relatives) Alder: (CN question 43)

4 and 5

CN Questionnaire

Antal år du har arbejdet i hjemmeplejen i alt: (CN question 45)

5

Nurses‟ extra education or pathways in palliative care

CN Questionnaire

5

Nurses‟ knowledge prior to palliative period

CN Questionnaire

Har du eller er du i gang med en professionel palliationsuddannelse / erfaring? Nej Ja, hvis ja, hvilken (CN question 46) Hvordan vil du karakterisere dit kendskab til patienten før det palliative forløb? (Sæt venligst kun et kryds) Meget godt Godt Nogenlunde Ikke særlig godt Kendte slet ikke patienten (CN question 5)

Gender of relative Relation to deceased

Questionnaires for relatives Questionnaires for relatives

Relative living with patient


Relative‟s vocational education


CNs‟ age

CN Questionnaire

Years as community nurse

4 and 5

Male Female Dichotomized into Non-spouse Spouse

4 and 5

No Yes

4 and 5

Dichotomized into 3 years or less > 3 years

5

Dichotomized into Min-39 39+ Dichotomized into 0-5 5+ No Yes

5

Dichotomized into Poor (1,2 on a 1-5-point scale) Good (3, 4 and 5)

Nurses‟ contact with relatives

CN Questionnaire

Nurses‟ home visits pr day

CN Questionnaire

GP involvement

CN questionnaires

Specialist team involvement

CN questionnaires

Har du haft kontakt med pårørende i forløbet? Ja Nej, af hvilken årsag: Der var ingen pårørende De pårørende ønskede ingen kontakt Anden årsag (CN question 8) Hvor ofte udførte hjemmeplejen (dig, dine kolleger eller hjemmehjælperne) følgende praktiske opgaver hos patienten i den sidste uge op til dødens indtræden? Flere gange om dagen En gang om dagen Flere gange om ugen Højst en gang om ugen Slet ikke (CN question 6) We used GP questionnaire information, but if the GP did not reply we used: Har du eller dine kollegaer været i forbindelse med den praktiserende læge i det palliative forløb? (CN question17) Hvis der ikke har været et palliativt team involveret i forløbet, beder vi dig gå til spørgsmål 31 Side 13 If the questionnaire were filled in but question number 27 to 30 all were blank, we took this as a no. (CN questionnaire page 9 )

5

No Yes

5

Dichotomized into Less than once a day Once a day or more

5

No Yes

5

No Yes

Appendix J: Cover letter, GPs (In Danish)

Løbenummer Januar 2007 Praksis-overskrift Praksis-navn Praksis-adresse Praksis-postnummer Praksis-postdistrikt

Kære Kollega

Døende cancer patienter har ofte behov for palliativ behandling og pleje, men der mangler indsigt i behov og ønsker for indsatsen i almen praksis og i samarbejdet med hjemmeplejen og sekundærsektoren. Vi vil analysere en serie nyligt afsluttede palliative forløb set fra dit, den pårørendes og hjemmesygeplejerskens perspektiv. Via sygehusenes dataregistre og CPR-registret har vi identificeret en nylig afdød cancer patient, der var tilmeldt din praksis. Vi beder dig derfor udfylde vedlagte skema om det palliative forløb. I kompagniskabspraksis beder vi den læge, som kendte patienten bedst, besvare skemaet. Hvis flere af de registrerede afdøde cancerpatienter var tilmeldt din praksis, modtager du mere end et skema. Vi ved, at det er et tidskrævende skema, så vi beder dig udfylde alle dem, du har mulighed for. Du bedes dog besvare spørgsmål 1-5 i alle skemaer. Vi håber, at du vil hjælpe os med at skaffe ny viden om emnet. Kun ved at afdække behov, ønsker og mangler kan vi foreslå tiltag til forbedring af indsatsen. Send venligst spørgeskemaet / spørgeskemaerne til Forskningsenheden for Almen praksis senest mandag den 5. februar 2007. Der vedlægges en frankeret svarkuvert. Hvert helt udfyldte skema honoreres med 360,16 kr., svarende til 3 moduler a 10 min. Honoraret udbetales, når dataindsamlingen er afsluttet. Spørgeskemaundersøgelsen er godkendt af Datatilsynet og DSAM og P.L.O´s Udvalg vedrørende multipraksisundersøgelser og har været forelagt den Videnskabsetiske Komité for Århus Amt. Har du spørgsmål eller kommentarer, er du meget velkommen til at kontakte os.

Med venlig hilsen

Mette Asbjørn Neergaard Projektansvarlig Speciallæge i almen medicin Direkte tlf.nr. 89 42 60 28 E-mail: [email protected]

Jens Søndergaard Praktiserende læge Seniorforsker Lektor, ph.d.

Anders Bonde Jensen Overlæge, ph.d. Onkologisk afd. D Århus Universitets hospital

Frede Olesen Forskningsleder Praktiserende læge Adj. professor Dr.med.

2007

Lægeskema

Foto: Morten Storgaard

Behandling og pleje af døende kræftpatienter i hjemmet

Løbenummer xxxx

Appendix K: Questionnaire, GPs (In Danish)

1

Kender ikke pårørende - gå venligst til spørgsmål 5 side 4

Andet:

Ven

Forælder

Søster / Bror

Datter / Søn

Kæreste, men ikke samboende

Ægtefælle / samlever

27987

Forbeholdt kodning

2 Hvilken voksen pårørende (18 år eller derover) var tættest på patienten og samtidig mest aktiv i plejen og behandlingen i det palliative forløb? (Sæt kun et kryds)

Hvis du har svaret ja ovenfor, beder vi dig gå videre med spørgeskemaet.

Hvis du har svaret nej ovenfor, kan den afdøde patient ikke indgå i projektet, og du skal ikke besvare resten af spørgeskemaet. Vi takker dig for hjælpen og beder dig returnere denne side i den frankerede svarkuvert.

Nej

Ja

af sin cancersygdom inklusive dens følgesygdomme?

cpr

1 Døde Fornavn Efternavn

Vi ønsker kun at inkludere afdøde cancerpatienter, hvor dødsårsagen kan tilskrives deres cancer sygdom. Vi har derfor brug for din hjælp til at validere registeroplysningerne og afgøre følgende:

Validering af registeroplysningerne

Dit navn / stempel: (Udfyldes altid)


2

(Vi ringer ikke til den pårørende, men bruger evt. telefonnummer til at lokalisere adresse)

Telefon:

Postnr. og by:

Adresse (gade, nr., sal):

Navn:

Brug venligst blokbogstaver

27987

3

27987

Vi definerer i spørgeskemaet det palliative forløb som den periode, hvor patientens væsentligste behov var palliativt, og hvor al kurativ behandling var stoppet, og frem til dødens indtræden.

Telefon:

4 Vi vil gerne have din hjælp til bedst muligt at identificere den pårørende, du beskrev i spørgsmål 2, og som skal deltage i spørgeskemaundersøgelsen. (Det er en stor hjælp for os, selv hvis du kun kan udfylde en del af oplysningerne)

Adresse (gade, nr.):

Lokalcenter / Hjemmeplejekontor:

Postnr. og by:

Forbeholdt kodning

Navn:

Den hjemmesygeplejerske, der var mest involveret i forløbet, var: Brug venligst blokbogstaver

5 Vi har brug for din hjælp til at finde den hjemmesygeplejerske, der var mest involveret i forløbet i hjemmet. Hvis du ikke har registreret et navn på en sygeplejerske, vil vi stadig bede dig anføre navnet på lokalcentret / hjemmepleje- eller områdekontoret.


Hvis du har markeret, at den pårørende ikke bør have et skema - gå venligst til spørgsmål 5 side 3

Anden årsag, hvilken:

Sproglige årsager (Vil ikke kunne forstå spørgeskemaet)

Psykiske årsager (F.eks. patologisk sorgreaktion)

Mentale årsager (F.eks. senilitet)

Den pårørende bør ikke modtage et spørgeskema af følgende primære årsag:

Pårørende formodes at kunne besvare et spørgeskema

3 Vi sender den nærmeste pårørende et spørgeskema om ca. 1 uge, medmindre du mener, at den pårørende ikke er i stand til at besvare et skema.


4

10 Hvilken type beskæftigelse havde patienten op til sin cancersygdom? (Sæt kun et kryds) Fuldtidsarbejde (Både lønnet og selvstændig) På fleksjobsordning På nedsat arbejdstid Under uddannelse På orlov (f.eks. barselsorlov eller uddannelsesorlov) Hjemmegående Arbejdsløs Sygemeldt Under omskoling eller revalidering Førtidspensionist / pensionist / efterlønsmodtager Ved ikke 27987

9 Havde patienten børn? (Sæt kun et kryds) Ja, deriblandt hjemmeboende børn Ja, men ikke hjemmeboende børn Nej Ved ikke

Ugift, ikke tidligere gift Gift / Samboende Enlig (skilt / separeret) Enlig (enke / enkemand) Ugift, men samboende med anden pårørende Ved ikke

8 Hvad var patientens ægteskabelige status? (Sæt kun et kryds)

7 Hvordan var patientens brug af almen praksis, efter dit skøn, før det palliative forløb? (Sæt kun et kryds) Patienten var hyppig bruger - også for småting Patienten var hyppig bruger - altid med relevante problemstillinger Patienten havde et gennemsnitlig forbrug af lægeydelser Patienten brugte sjældent læge Jeg kender ikke til patientens tidligere brug af almen praksis

6 Hvordan vil du karakterisere dit / din praksis' faglige kendskab til patienten før det palliative forløb? (Sæt kun et kryds) Meget godt Godt Nogenlunde Ikke særlig godt Kendte ikke patienten

Følgende spørgsmål handler om dit kendskab til patienten inden det palliative forløb


Hjemmebesøg

I konsultationen

Telefonkonsultation

En gang om dagen

Flere gange om ugen

5

En gang om ugen

En gang hver 14. dag

Mindre end hver 14. dag

14 Hvilken type og hvor ofte havde du / din praksis kontakt med patienten eller familien i størstedelen af det palliative forløb? (Sæt venligst et kryds i hver række)

Hvis du har svaret Nej til spørgsmål 13 gå til spørgsmål 31 på side 13

Anden grund, anfør hvilken:

Patienten blev udskrevet til pårørendes hjem, der ligger langt væk

Jeg / vi blev ikke kontaktet af udskrivende afdeling

Patienten / familien ønskede ikke kontakt til mig / os

Jeg / vi blev ikke kontaktet af patienten / familien

Patienten kom direkte fra hospitalet til hospice

Jeg / praksis henviste til Det Palliative Team med det samme

Patienten blev henvist til Det Palliative Team fra hospitalsafdelingen

Patienten døde meget hurtigt uden en egentlig palliativ fase

Patienten var indlagt det meste af tiden

Nej - Hvis nej, af hvilken årsag havde din praksis ikke kontakt til patient og familie (Sæt gerne flere krydser)

Ja

27987

Slet ikke

Forbeholdt kodning

13 Havde du /din praksis nogen form for kontakt med patienten eller familien i det palliative forløb?

12 Hvor mange af de uger, du angav ovenfor, skønner du, patienten sammenlagt var i hjemmet (enten i patientens eget hjem, på fast plejehjemsplads eller i pårørendes hjem, hvis det var der, plejen foregik)? uger Antal uger(< 1 uge skrives som 0)

11 Hvor lang tid skønner du, at det palliative forløb varede i alt (fra det tidspunkt, hvor patientens væsentligste behov var palliativt, og al kurativ behandling var stoppet, og frem til dødens indtræden)? Antal uger(< 1 uge skrives som 0) uger

Følgende spørgsmål handler om det palliative forløb og din kontakt med patienten i forløbet


Flere gange om ugen

En gang om ugen

En gang hver 14. dag

Mindre end hver 14. dag Slet ikke

Forbeholdt kodning

6

27987

18 Aftalte du med patienten og pårørende, hvem der skulle kontaktes, når du / dine kompagnoner ikke var tilgængelige? Nej Ja - hvis ja, beder vi dig anføre hvem: (Sæt gerne flere krydser) Praktiserende læge med andet ydernummer, jeg havde lavet aftale med Vagtlægen Forbeholdt Hospitalet kodning Det Palliative Team Anden, anfør hvem:

Jeg havde givet det til hjemmeplejen / hjemmesygeplejersken Det var et meget kort / hurtigt forløbende forløb Det var ikke relevant i dette palliative forløb Det faldt mig ikke ind i det konkrete forløb Det gør jeg generelt aldrig Andet, anfør hvad:

17 Har du opgivet dit private telefonnummer til patient og / eller pårørende i det palliative forløb? Ja, det gør jeg generelt altid Ja, det gør jeg i enkelte udvalgte forløb Nej - hvis nej, beder vi dig anføre hvorfor: (Sæt evt. flere krydser)

16 Tog du initiativ til uopfordrede hjemmebesøg hos denne patient, d.v.s. besøg i hjemmet, hvor du foreslog / tog initiativ til besøget? Ja Nej - hvis nej, beder vi dig anføre hvorfor: (Sæt evt. flere krydser) Jeg havde altid fast aftalte besøg Det var et meget kort / hurtigt forløb Jeg havde ikke tid til at udføre uopfordrede hjemmebesøg Det var ikke relevant i dette palliative forløb Forbeholdt Det faldt mig ikke ind i det konkrete forløb kodning Det gør jeg generelt aldrig Andet, anfør hvad:

Hjemmebesøg

I konsultationen

Telefonkonsultation

En gang om dagen

15 Hvilken type og hvor ofte havde du / din praksis kontakt med patienten eller familien den sidste uge af det palliative forløb? (Sæt venligst et kryds i hver række)


Noget tid

Lidt tid

Ingen tid

Ikke relevant

7

27987

20 Udfyldte du / din praksis nogen form for terminal ansøgning (både være om orlov og medicin) i forløbet? Ja Nej - hvis nej, anfør hvorfor: (Sæt gerne flere krydser) Den blev udfyldt af en anden Der var ikke behov for terminalerklæring Jeg oplevede ikke patienten som terminal (Uventet / pludselig død) Forbeholdt Patient / pårørende ønskede det ikke kodning Jeg var bange for at patienten ville opleve det som en dødsdom Anden grund:

Patientens tanker om den nærmeste pårørendes fremtid Patientens relation til venner og familie i øvrigt

Patientens tanker om plejeorlov til pårørende Patientens relation til de nærmeste pårørende

Patientens tanker om, hvordan den sidste tid skulle komme til at foregå Patientens tanker om begravelse og testamente Patientens tanker om terminalerklæring

Patientens økonomiske og arbejdsrelaterede problemstillinger

Patientens religiøse og / eller spirituelle problemstillinger Patientens etiske og / eller eksistentielle problemstillinger

Patientens kognitive symptomer (f.eks. dårlig hukommelse, koncentrationsbesvær)

Patientens psykiske symptomer (f.eks. uro, angst, krise, depression)

Patientens fysiske symptomer (f.eks. smerte, kvalme, åndenød, træthed, appetitløshed)

Meget tid

19 Hvor meget tid brugte du på følgende aspekter i din kontakt med patienten i det palliative forløb? (Sæt venligst et kryds i hver række)


H j e m (Ikk me ho e s p leje patie hje m) nten F a s t bo lig på p leje Afla hje stn m i n g sp l ads p

Forbeholdt kodning

8

27987

Hvis patienten døde det ønskede sted, eller hvis patienten ikke havde udtrykt noget ønske, gå direkte til side 10 øverst.

Andet sted, anfør venligst hvilket:

Hospice

Hospital

Hjemme hos pårørende

Aflastningsplads på plejehjem

Fast plads på plejehjem / plejebolig

Ældre bolig / beskyttet bolig

22 Hvor døde patienten? (Sæt kun et kryds) Hjemme (Ikke plejehjem / ældrebolig / beskyttet bolig)

I slutningen af det palliative forløb

I midten af det palliative forløb

I begyndelsen af det palliative forløb

ste d , a nfø r ve

Ja - hvis ja, hvor? (Sæt venligst et kryds i hver række)

Ho spi ce And et

21 Udtrykte patienten over for dig et ønske om hvor han/hun helst ville dø? Nej

Følgende spørgsmål handler om dødssted og ønske om dødssted


å p leje hje m ital

Ho sp

t: n l i g st h vilk e

Ve d ik ke

Ved ikke

Nej

Ja

9

25 Hvis patienten ikke døde det ønskede sted: Havde du indtryk af, at det var et problem for de pårørende, at patienten døde et andet sted end ønsket?

Ved ikke

Nej

Ja

24 Hvis patienten ikke døde det ønskede sted: Havde du indtryk af, at det var et problem for patienten at skulle dø et andet sted end ønsket?

Ved ikke

Anden grund, anfør venligst hvilken:

På grund af ønske fra hjemmeplejen

På grund af ønske fra de pårørende

På grund af ændret ønske fra patienten i den allersidste del af det palliative forløb

På grund af opkald til 112

På grund af indlæggelse via vagtlæge

23 Hvis patienten ikke døde det ønskede sted: Hvorfor gjorde han/hun ikke det? ( Sæt evt. flere krydser) På grund af uforudsete klager / symptomer diagnosticeret i normal arbejdstid


27987

Forbeholdt kodning

Forbeholdt kodning

Kontakt ved hjemmebesøg

Telefonisk kontakt Kontakt i konsultationen

Meget ofte

10

Ofte

Mindre ofte

27 Hvilken type kontakt havde du med den nærmeste pårørende?: (Sæt venligst et kryds i hver række) Sjældent

27987

Aldrig

Hvis du slet ikke havde kontakt med patientens pårørende, gå venligst til spørgsmål 31 side 13


De pårørende ønskede ingen kontakt

Nej, af hvilken årsag: Der var ingen pårørende

26 Havde du kontakt med pårørende i forløbet? Ja

Følgende spørgsmål handler om den pårørende, du oplyste navn på tidligere. Denne pårørende vil i det følgende blive omtalt som "den nærmeste pårørende". Selvom du ikke kunne anføre et navn tidligere, så udfyld skemaet, idet du tænker på den pårørende, du havde mest kontakt til i forløbet.


Omsorg for den pårørendes psykiske velbefindende i forløbet (f.eks. angst, krise, depression)

Den pårørendes etiske og / eller eksistentielle problemstillinger Den pårørendes økonomiske og arbejdsrelaterede problemstillinger

Den pårørendes relation til patienten Den pårørendes relation til børn og øvrige familie Den pårørendes religiøse og / eller spirituelle problemstillinger

Den pårørendes tanker om plejeorlov

Den pårørendes tanker om begravelse og testamente Den pårørendes tanker om terminalerklæring

Pårørendes ønske om "dødssted"

Den pårørendes tanker om, hvordan den sidste tid skulle komme til at foregå

Den pårørendes tanker om patientens fysiske symptomer

Meget tid

11

Noget tid

Lidt tid

Ingen tid

27987

Ikke relevant

28 Hvor meget tid brugte du på følgende aspekter i din kontakt med den nærmeste pårørende i det palliative forløb? (Sæt venligst et kryds i hver række)


Nej

ved ikke

Ja, vi har haft personlig kontakt

Ja, vi har haft telefonisk kontakt

Nej, det har jeg generelt aldrig

12

Nej, men vi har en konkret aftale om kontakt

Nej. Jeg / vi tilbød det, men den pårørende ønskede det ikke

Nej, men jeg / vi har opfordret den pårørende til at kontakte mig / os ved behov

Nej, men jeg / vi har tilbudt opfølgende konsultation

27987

30 Har du / din praksis haft en kontakt, der omhandlede forløbet med den nærmeste pårørende efter dødsfaldet? (Sæt kun et kryds)

Drøftede du den pårørendes ønske om at deltage aktivt i plejen med den pårørende? Drøftede du den pårørendes mulighed (både psykisk og fysisk) for at deltage aktivt i plejen med den pårørende? Deltog den nærmeste pårørende aktivt i behandlingen (medicingivning, sårskift, etc.)? Drøftede du den pårørendes ønske om at deltage aktivt i behandlingen med den pårørende? Drøftede du den pårørendes mulighed (både psykisk og fysisk) for at deltage aktivt i behandlingen med den pårørende? Var den nærmeste pårørende til stede i hjemmet det meste af tiden? Tog den nærmeste pårørende plejeorlov i forløbet? Tog den nærmeste pårørende ferie / fridage i forløbet for at passe patienten?

Deltog den nærmeste pårørende aktivt i plejen (madning, badning, etc.)?

Ja

29 Følgende spørgsmål handler om den pårørendes aktive deltagelse i forløbet (Sæt venligst et kryds i hver række)


Forbeholdt kodning

Forbeholdt kodning

Andet, anfør venligst hvad:

Efter hjemmeplejens ønske

Efter de pårørendes ønske

Efter patientens ønske

På grund af ringe kendskab til patienten

På grund af egen tidsmangel

13

På grund af sociale omstændigheder (f.eks. manglende ressourcer hos pårørende)

På grund af psykiske symptomer (f.eks. delir eller konfusion)

På grund af fysiske symptomer (f.eks. svært behandlelige smerter eller kvalme)

27987

Forbeholdt kodning

32 Hvorfor henviste du / din praksis patienten til et hospital, palliativt team eller hospice: (Sæt evt. flere krydser)

Hvis du / din praksis ikke henviste patienten til et hospital, palliativt team eller hospice, gå til spørgsmål 34 side 15

3.

2.

1.

31A Hvis du henviste til en hospitalsafdeling, skriv venligst hvilke type specialer:

Ingen steder

Andet, anfør venligst hvad:

Socialforvaltningen

Psykolog

Fysioterapeut

Hospice

Hospitalsafdeling (Anfør venligst nedenfor, hvilke typer afdelinger)

Palliativt team

31 Hvor henviste du eller andre læger i din praksis patienten til i det palliative forløb? (Sæt gerne flere krydser)

Følgende spørgsmål handler om samarbejdspartnere i det palliative forløb


Anden person, angiv hvem:

Hjemmesygeplejerske

14

Praktiserende læge i din eller anden praksis

Det Palliative Team

Læge på sygehusafdeling

Forbeholdt kodning

27987

Ja - hvis ja, beder vi dig anføre, hvem du har konfereret med: (Sæt evt. flere krydser)

Nej, men set i bakspejlet burde jeg have gjort det

Nej, det havde jeg ikke behov for

33 Konfererede/drøftede du faglige aspekter i dette palliative forløb med andre (Ikke pårørende)?


15

27987

36 Blev du kontaktet af en hospitalsafdeling, inden patienten blev udskrevet til palliativ pleje i hjemmet? Ja Nej

35 Indhentede du supplerende informationer fra hospitalsafdelinger ud over de informationer og epikriser, der kom fra hospitalet? Ja Nej

Om informationen til patient og pårørende

Om særlige behov i hjemmet

Om bivirkninger til behandlingen

Om behandlingsplanen

Om hvilke problemstillinger man kunne forvente i det videre forløb

Om diagnose

34 Fik du tilfredsstillende information fra den hospitalsafdeling, der udskrev patienten til det palliative forløb? (Sæt venligst et kryds i hver række) Ikke TilfredsOvervejende Overvejende Utilfredsrelevant stillende tilfredsutilfredsstillende for dette stillende stillende forløb

Hvis patienten ikke nåede at blive udskrevet til hjemmet i det palliative forløb, gå til spørgsmål 39 side 16

Følgende spørgsmål handler om samarbejdet med hospitalsafdelinger i det palliative forløb (dog ikke palliative teams)


Noget enig

Noget uenig

Helt uenig

Ved ikke

Godt

Rimeligt

Dårligt

Rigtig Ikke relevant dårligt /ved ikke

Rigtig godt

Godt

16

Rimeligt

Dårligt

27987


39 Når du tænker på dette palliative forløb som helhed, hvordan vil du så karakterisere kvaliteten af hospitalsafdelingernes samlede indsats i relation til det palliative forløb (ikke palliative teams)? (Sæt venligst kun et kryds)

Rigtig godt

38 Når du tænker på dette palliative forløb som helhed, hvordan vil du så karakterisere dit samarbejde med hospitalsafdelingerne (ikke palliative teams)? (Sæt venligst kun et kryds)

At jeg savnede personligt kendskab til kolleger på hospitalet At kendskab til læger på hospitalet betød/ ville have betydet noget positivt i forhold til indsatsen

At epikriserne kom i tide At jeg havde en fast kontaktperson blandt de ansatte på hospitalet

At der var tilstrækkelig faglig kontakt mellem mig / min praksis og afdelingerne At der var en "overlevering" af patienten fra hospitalet til mig / min praksis ved udskrivelsen

At jeg var klar over min egen rolle i det palliative forløb

Helt enig

37 I hvilken grad er du enig i følgende udsagn? Under patientens palliative forløb oplevede jeg: (Sæt venligst et kryds i hver række)


M.h.t. information til patient og pårørende

M.h.t. særlige behov i hjemmet

M.h.t. bivirkninger til behandlingen

M.h.t. hvilke problemstillinger man kunne forvente i det videre forløb M.h.t. behandlingsplan

Tilfredsstillende

17

Overvejende Overvejende tilfredsutilfredsstillende stillende

40 Var den information du fik fra et palliativt team tilfredsstillende: (Sæt venligst et kryds i hver række) Utilfredsstillende

27987

Er ikke relevant for dette forløb

Hvis der ikke har været et palliativt team involveret i forløbet, beder vi dig gå til side 20 øverst

Følgende spørgsmål handler om samarbejdet med palliative teams


At der var tilstrækkelig telefonisk kontakt mellem Det Palliative Team og mig / min praksis At Det Palliative Team havde for travlt til at samarbejdet fungerede At Det Palliative Team bestræbte sig på at bevare mig som en del af behandlerteamet At Det Palliative Team bestræbte sig på at bevare hjemmeplejen som en del af behandlerteamet At der var en fast kontaktperson i teamet, så jeg vidste, hvem jeg skulle kontakte At Det Palliative Team var organiseret på en måde, der faciliterede samarbejdet At det var et alvorligt problem, at Det Palliative Team ikke var tilgængeligt efter normal arbejdstid At jeg savnede kendskab til de ansatte i Det Palliative Team At kendskab til Det Palliative Team betød / ville have betydet noget positivt i forhold til indsatsen

At Det Palliative Team var let at få fat i

At jeg fik mere viden om palliation At jeg blev hægtet af behandlingsforløbet At der var en god dialog om behandlingsplanen At der var en god dialog om ansvarsfordelingen

At samarbejdet lettede mit arbejde

18

27987

41 I hvilken grad er du enig i følgende udsagn? Under samarbejdet med et palliativt team oplevede jeg: (Sæt venligst et kryds i hver række) Er ikke relevant for Helt Noget Noget Helt denne enig enig uenig uenig patient


Godt

Rimeligt

Dårligt


Rigtig godt

Godt

19

Rimeligt

Dårligt

27987


43 Når du tænker på dette palliative forløb som helhed, hvordan vil du så karakterisere kvaliteten af Det Palliative Teams samlede indsats? (Sæt venligst kun et kryds))

Rigtig godt

42 Når du tænker på dette palliative forløb som helhed, hvordan vil du så karakterisere samarbejdet med Det Palliative Team? (Sæt venligst kun et kryds)


At hjemmesygeplejerskerne bestræbte sig på at bevare mig som en del af behandlerteamet At der var én kontaktsygeplejerske, så jeg vidste, hvem jeg skulle kontakte At hjemmesygeplejerskerne var organiseret på en måde, der faciliterede samarbejde At jeg savnede personligt kendskab til hjemmesygeplejerskerne At kendskab til hjemmesygeplejerskerne betød / ville have betydet noget positivt i forhold til

At hjemmesygeplejerskerne havde for travlt til, at samarbejdet fungerede

At hjemmesygeplejerskerne var lette at få fat i At der var tilstrækkelig telefonisk kontakt mellem hjemmesygeplejerskerne og mig / min praksis

At der var en god dialog om ansvarsfordelingen

At hjemmesygeplejerskerne var engagerede i forløbet At hjemmesygeplejerskerne var opdateret m.h.t. viden omkring palliativ behandling og pleje At der var en god dialog om behandlingsplanen

Helt enig

20

Noget enig

Noget uenig

Helt uenig

44 I hvilken grad er du enig i følgende udsagn? Under det aktuelle palliative forløb oplevede jeg: (Sæt venligst et kryds i hver række)

27987

Ved ikke/ ikke relevant

Hvis du / din praksis overhovedet ikke har været involveret i forløbet, beder vi dig gå til spørgsmål 54 side 24

Hvis hjemmeplejen slet ikke var involveret i forløbet, gå til spørgsmål 53 side 23

Følgende spørgsmål handler om samarbejdet med hjemmeplejen (hjemmesygeplejerskerne og hjemmehjælperne)


Forbeholdt kodning

Forbeholdt kodning

Forbeholdt kodning

I høj grad

I nogen grad

21

I mindre grad

Slet ikke

27987

49 Savnede du (flere) tværfaglige møder med hjemmeplejen i forbindelse med dette palliative forløb?

Nej, jeg deltog ikke af anden grund. Anfør, venligst hvilken:

Nej, jeg deltog ikke p.g.a. tidsmangel

48 Deltog du / din praksis i et eller flere fælles møder men hjemmeplejen om dette palliative forløb? Ja

Ja

47 Blev der foreslået et eller flere fælles møder med hjemmeplejen om dette palliative forløb med deltagelse af dig? Nej (gå til spørgsmål 49)

Anden, anfør hvem:

Det Palliative Team

Hospitalet

Vagtlægen

Praktiserende læge med andet ydernummer, jeg havde lavet aftale med

Ja - hvis ja, beder vi dig anføre hvem: (Sæt gerne flere krydser)

46 Aftalte du med hjemmeplejen, hvem der skulle kontaktes, når du / din praksis ikke var tilgængelig? Nej

Andet, anfør hvad:

Det gør jeg generelt aldrig

Det faldt mig ikke ind i det konkrete forløb

Det var ikke relevant i dette palliative forløb

Ved behov ville jeg give det til patient og pårørende i stedet

Jeg havde givet det til patient og / eller pårørende

Nej - hvis nej, beder vi dig anføre hvorfor: (Sæt gerne flere krydser)

45 Har du opgivet dit private telefonnummer til hjemmeplejen det palliative forløb? Ja


Godt

Rimeligt

Dårligt

Rigtig Ikke relevant /ved ikke dårligt

Rigtig godt

Godt

22

Rimeligt

Dårligt

Rigtig dårlig

27987

51 Når du tænker på dette palliative forløb som helhed, hvordan vil du så karakterisere kvaliteten af hjemmesygeplejerskernes samlede indsats? (Sæt kun et kryds)

Rigtig godt

50 Når du tænker på dette palliative forløb som helhed, hvordan vil du så karakterisere samarbejdet med hjemmesygeplejerskerne og hjemmeplejen? (Sæt kun et kryds)


Godt

Rimeligt

Dårligt

Rigtig dårligt

Rigtig godt

Godt

23

Rimeligt

Dårligt

Rigtig dårligt

27987

53 Når du tænker på dette palliative forløb som helhed, hvordan vil du så karakterisere kvaliteten af din egen eller din praksis' samlede indsats? (Sæt kun et kryds)

Rigtig godt

52 Når du tænker på dette palliative forløb som helhed, hvordan vil du så karakterisere kvaliteten af primærsektorens (almen praksis, hjemmeplejen og hjemmesygeplejerskerne) samlede indsats? (Sæt kun et kryds)

Hvis du / din praksis overhovedet ikke har været involveret i forløbet, beder vi dig gå til spørgsmål 54 side 24

I det følgende beder vi dig vurdere dette palliative forløb som helhed


Noget enig

Nej

Nej

Noget uenig

Meget ofte

Ofte

Mindre ofte

24

Sjældent

Ja, jeg bruger den: (Sæt kun et kryds)

Aldrig

57 Kender du DSAMs kliniske vejledning: "Palliation i primærsektoren"? Nej

Ja

56 Ønsker du flere kurser i emner inden for palliation?

Ja

55 Har du været på kursus i emner inden for palliation?

Praktiserende læger har ikke tid til at behandle døende patienter i hjemmet

Jeg føler mig alene i arbejdet med døende patienter At arbejde med døende patienter er hårdt og kommer ind imellem "for tæt på" Jeg er generelt meget tilgængelig for mine palliative patienter

Helt enig

54 I hvilken grad er du enig i følgende udsagn? (Sæt venligst et kryds i hver række) Helt uenig

Følgende spørgsmål handler om dit forhold til palliativ behandling generelt


27987

Ved ikke

Noget enig

Noget uenig

Helt uenig

Ved ikke

Anden person, hvem:

25

Den sundhedsperson, der har været mest involveret i forløbet

Den sundhedsperson, som patient og pårørende ønsker

Den behandlende læge på sygehuset

En sygeplejerske fra den behandlende afdeling

En person fra Det Palliative Team

Den praktiserende læge

En hjemmesygeplejerske

27987

Forbeholdt kodning

59 Hvis der var en fast person fra sundhedssektoren, der skulle være "tovholder" i forløbet (have det overordnede ansvar for forløbet og koordinere samarbejdet mellem involverede fagpersoner), hvem mener du så, det skulle være. (Sæt kun et kryds)

Jeg mener, at de palliative patienter aktivt burde henvises tilbage til mig efter udskrivelsen fra sygehuset Der er behov for en tjekliste for palliation af cancerpatienter, der indeholder bl.a. medicinske, psykologiske, økonomiske, funktionelle og sociale aspekter Der er behov for en fælles palliationsjournal for både hospitalspersonale, praktiserende læger og hjemmesygeplejersker Hvis der skal være en fælles palliationsjournal for både hospitalspersonale, praktiserende læger og hjemmesygeplejersker kunne den med fordel være internetbaseret

Jeg oplever, at patienter føler sig overladt til sig selv, når de udskrives til et palliativt forløb

Helt enig

58 I hvilken grad er du enig i følgende udsagn?: (Sæt venligst et kryds i hver række)


År

patienter

Ja

Nej Ja

Nej

26

71 Dato for udfyldelse af spørgeskemaet: (DD-MM-ÅÅÅÅ)

P.L.O. fuldtidsstillinger

70 Antal normerede P.L.O. -fuldtidsstillinger under ydernummeret:

Læger

-

27987

- 2 0 0 7

69 Hvor mange læger arbejder i alt for tiden under ydernummeret, inkl. faste vikarer o.l.?

Kompagniskabspraksis med færre læger end ydernummeret er normeret til

Kompagniskabspraksis med flere læger end ydernummeret er normeret til

Kompagniskabspraksis med det antal læger som ydernummeret er normeret til

Delepraksis i gruppepraksis

Gruppepraksis. To eller flere læger i solopraksis der deler praksisfacilliteter

Delepraksis. Ydernummeret er normeret til én læge, men deles af to eller flere læger

Solopraksis. Ydernummeret er normeret til én læge og i praksis arbejder én læge

68 Hvorledes er praksis organiseret (spørgsmålet vedrører faste læger. Se bort fra vikarer, praksisreservelæger og amanuenser)? (Sæt kun et kryds)

67 Er der jævnligt knyttet studenter, praksisreservelæger eller uddannelsesamenuenser til praksis?

66 Er der sygeplejersker ansat i praksis?

Stillinger

65 Det samlede hjælpepersonale med patientkontakt i praksis, omregnet og afrundet til antal hele fuldtidsstillinger:

64 Praksisstørrelse i alt (gruppe 1-sikrede tilmeldt ydernummer):

En landpraksis

En bypraksis

63 Opfatter du praksis som:

62 Kandidatår:

61 Alder:

Kvinde

60 Køn Mand

I det følgende beder vi dig angive følgende om dig og din praksis.


27

72 Hvis du har kommentarer, beder vi dig skrive dem her:


27987

Forbeholdt kodning

Gironummer:

Mette Asbjørn Neergaard Projektansvarlig Speciallæge i almen medicin Forskningsenheden for Almen Praksis i Århus Tlf: 8942 6028 Email: [email protected]

Med venlig hilsen

Tusind tak for hjælpen

Jens Søndergaard Praktiserende læge Seniorforsker, lektor Ph.d.

28

Anders Bonde Jensen Overlæge, ph.d. Onkologisk afdeling D Århus Universitetshospital

27987

Frede Olesen. Praktiserende læge Forskningsleder Adj. professor, dr.med.

Hvis du er interesseret i at modtage E-mail om resultaterne af denne undersøgelse om ca. 1 år sendes en mail til: [email protected]

Praksis SE-nummer: (Vigtigt af hensyn til udbetalingen) Beløbet vil blive indsat på din konto, når dataindsamlingen er afsluttet.

eller på

Kontonummer:

Honoraret ønskes indsat enten på: Reg.nr:

73 Herunder beder vi dig udfylde, hvortil du ønsker honoraret overført:


Appendix L: Cover letter, Community nurses (In Danish)

April 2007

«Løbenummer» «HStittel» «HSnavn» «Lokalcenter» «HSAdresse» «HSPostnummer» «HSPostdidtrikt» Kære «HStittel» Hjemmeplejen spiller en vigtig rolle i behandling og pleje i hjemmet af døende kræftpatienter i den sidste tid, hvor der er behov for palliativ indsats. Der mangler dog viden om behov og ønsker for indsatsen og om samarbejdet med praktiserende læger og sekundærsektoren. Kun ved at afdække disse behov, ønsker og mangler kan man kvalificere forslag til forbedring af indsatsen. Vi håber, at du / I vil hjælpe os med at skaffe ny viden om emnet. Vi vil analysere en serie nyligt afsluttede palliative forløb set fra hjemmesygeplejerskens, den pårørendes og den praktiserende læges perspektiv. Via sygehusenes dataregistre, CPR-registret og den praktiserende læge har vi identificeret en nylig afdød cancer patient: «Ptnavn» (CPR-nr: «PtCPR») «Ptadresse» «Ptby» «Ptpostdistrikt» VIGTIGT: Vi vil bede om, at spørgeskemaet gives til den hjemmesygeplejerske, der var mest involveret i det palliative forløb hos denne afdøde patient. Hvis - hjemmeplejen ikke har taget del i et palliativt forløb hos denne patient eller - patienten var plejehjemsbeboer allerede inden det palliative forløb, bedes du skrive dette på skemaets forside og sende skemaet tilbage ubesvaret i svarkuverten. I nogle tilfælde har vi fået den praktiserende læge til at besvare spørgsmål 0. Det fremgår i så fald af skemaet. Du bedes udfylde resten af skemaet. Hvis du af tidspres eller af andre årsager ikke kan udfylde hele skemaet, må du meget gerne udfylde spørgsmålene til og med spørgsmål 3 og sende skemaet til os. Spørgeskemaundersøgelsen er godkendt af Datatilsynet og har været forelagt den Videnskabsetiske Komité for Århus Amt. Spørgeskemaet, der tager ca. 30 minutter at udfylde, anonymiseres og ingen ud over projektgruppen får at vide, hvad du har svaret. Send venligst spørgeskemaet inden d. 7.maj 2007 i den frankerede svarkuvert. Har du spørgsmål eller kommentarer, er du meget velkommen til at kontakte os. Med venlig hilsen

Mette Asbjørn Neergaard Projektansvarlig Speciallæge i almen medicin Tlf: 8942 6028 E-mail: [email protected]

Jens Søndergaard Praktiserende læge Seniorforsker Lektor, ph.d.

Anders Bonde Jensen Overlæge, ph.d. Onkologisk afd. D Århus Universitetshospital

Frede Olesen Forskningsleder Praktiserende læge Adj. professor Dr.med.

Appendix M: Questionnaire, Community nurses (In Danish)

2007

Hjemmesygeplejerskeskema



Løbenummer xxxx

Nej

Forbeholdt kodning


1

Sproglige årsager (Vil ikke kunne forstå spørgeskemaet)

Psykiske årsager (f.eks. svær behandlingskrævende depression)

Mentale årsager (f.eks. senilitet)

Den pårørende er ikke er i stand til at udfylde et spørgeskema af følgende årsag:

Pårørende formodes at kunne besvare et spørgeskema

44487

Forbeholdt kodning

2 Vi sender den nærmeste pårørende et spørgeskema om ca. 1 uge, medmindre du mener, at den pårørende ikke er i stand til at besvare et skema.

Kender ikke pårørende - gå venligst til spørgsmål 4 side 2

Andet:

Ven

Forælder

Søster / bror

Datter / søn


1 Hvilken voksen pårørende var mest involveret i plejen og behandlingen i det palliative forløb? (Sæt venligst kun et kryds) Ægtefælle / samlever

Vi har brug for din hjælp til at identificere den af patientens pårørende, der skal have tilsendt et spørgeskema

Vi definerer det palliative forløb som perioden fra patientens problemstillinger blev af lindrende og ikke af helbredende karakter og frem til dødens indtræden

Hvis du har svaret ja ovenfor, beder vi dig gå videre med spørgeskemaet.

Hvis du har svaret nej ovenfor og skal du ikke besvare resten af spørgeskemaet. Vi takker dig for hjælpen og beder dig returnere denne side i den frankerede svarkuvert.

Ja / det formodes

af sin cancersygdom inklusive dens følgesygdomme?

0 Døde Fornavn Efternavn, cpr

Vi ønsker kun at inkludere afdøde cancerpatienter, hvor dødsårsagen kan tilskrives deres cancer sygdom. Vi har derfor brug for din hjælp til at validere registeroplysningerne og afgøre følgende:

Dit navn:


Nej

Ved ikke

Andet, anfør hvad:

Andet, anfør hvad:

Hjælp til ernæring

Sårpleje

Medicingivning

Personlig pleje og hygiejne

2

44487

Kode

Kode

6 Hvor ofte udførte hjemmeplejen (dig, dine kolleger eller hjemmehjælperne) følgende praktiske opgaver hos patienten i den sidste uge op til dødens indtræden? (Sæt venligst et kryds i hver række) Flere gange En gang Flere gange Højst en Slet ikke om dagen om dagen om ugen gang om ugen

Følgende spørgsmål handler om din kontakt med patienten i det palliative forløb

Kendte slet ikke patienten

Ikke særlig godt

Nogenlunde

Godt

Meget godt

5 Hvordan vil du karakterisere dit kendskab til patienten før det palliative forløb? (Sæt venligst kun et kryds)

Ja

4 Kom hjemmeplejen (dig, dine kolleger eller hjemmehjælperne) i patientens hjem før det palliative forløb?

Følgende spørgsmål handler om din kontakt med patienten før det palliative forløb

(Vi ringer ikke til den pårørende, men bruger evt. telefonnummer til at lokalisere adresse)

Telefon:

Postnr. og by:

Adresse: (gade, nr., sal):

Navn:

3 Vi vil gerne have din hjælp til at finde frem til denne pårørende, der skal deltage i spørgeskemaundersøgelsen. (Det er en stor hjælp for os, også hvis du kun kan udfylde en del af oplysningerne)


Patientens tanker om den nærmeste pårørendes fremtid Patientens relation til venner og familie i øvrigt

Patientens tanker om plejeorlov til pårørende Patientens relation til de nærmeste pårørende

Patientens tanker om, hvordan den sidste tid skulle komme til at forløbe Patientens tanker om begravelse og testamente Patientens tanker om terminalerklæring

Patientens økonomiske og arbejdsrelaterede problemstillinger

Patientens religiøse og / eller spirituelle problemstillinger Patientens etiske og / eller eksistentielle problemstillinger

Patientens kognitive symptomer (f.eks. dårlig hukommelse, koncentrationsbesvær)

Patientens psykiske symptomer (f.eks. uro, angst, krise, depression)

Patientens fysiske symptomer (f.eks. smerte, kvalme, åndenød, træthed, appetitløshed)

3

44487

7 Hvor meget tid brugte du på følgende aspekter i din kontakt med patienten i det palliative forløb? (Sæt venligst et kryds i hver række) Meget Noget Lidt Ingen Ikke tid tid tid tid relevant


Forbeholdt kodning

Omsorg for den pårørendes psykiske velbefindende i forløbet (f.eks. angst, krise, depression)

Den pårørendes etiske og / eller eksistentielle problemstillinger Den pårørendes økonomiske og arbejdsrelaterede problemstillinger

Den pårørendes relation til patienten Den pårørendes relation til børn og øvrige familie Den pårørendes religiøse og / eller spirituelle problemstillinger

Den pårørendes tanker om plejeorlov

Den pårørendes tanker om begravelse og testamente Den pårørendes tanker om terminalerklæring

Pårørendes ønske om "dødssted"

Den pårørendes tanker om, hvordan den sidste tid skulle komme til at forløbe

Den pårørendes tanker om patientens fysiske symptomer

Meget tid

4

Noget tid

Lidt tid

Ingen tid

44487

Ikke relevant

9 Hvor meget tid brugte du på følgende aspekter i din kontakt med den nærmeste pårørende i det palliative forløb? (Sæt venligst et kryds i hver række)

Den pårørende, der var mest involveret i plejen og behandlingen i det palliative forløb, kalder vi herefter den nærmeste pårørende

Hvis du ikke havde kontakt med de pårørende overhovedet, gå venligst til spørgsmål 11 side 6


De pårørende ønskede ingen kontakt

Der var ingen pårørende

Nej, af hvilken årsag:

8 Har du haft kontakt med pårørende i forløbet? Ja

Følgende spørgsmål handler om den pårørende


Tog den nærmeste pårørende plejeorlov i forløbet? Tog den nærmeste pårørende ferie / fridage i forløbet for at passe patienten?

10A (Sæt venligst et kryds i hver række)

Hvor meget drøftede du den pårørendes mulighed (både psykisk og fysisk) for at deltage aktivt i behandlingen med ham / hende (medicingivning, sårskift, etc.)?

Hvor meget drøftede du den pårørendes ønske om at deltage aktivt i behandlingen med ham / hende (medicingivning, sårskift, etc.)?

Hvor meget deltog den nærmeste pårørende aktivt i behandlingen (medicingivning, sårskift, etc.)?

Hvor meget drøftede du den pårørendes mulighed (både psykisk og fysisk) for at deltage aktivt i plejen med ham / hende (madning, badning, etc.)?

Hvor meget drøftede du den pårørendes ønske om at deltage aktivt i plejen med ham / hende (madning, badning, etc.)?

Hvor meget deltog den nærmeste pårørende aktivt i plejen (madning, badning, etc.)?

Meget

5

Noget

Ja

Lidt

Nej

Slet ikke

10 Følgende spørgsmål handler om den pårørendes aktive deltagelse i forløbet (Sæt venligst et kryds i hver række)


44487

Ved ikke

Ved ikke

Forbeholdt kodning

M.h.t. behandlingsplanen M.h.t. bivirkninger til tidligere onkologisk behandling M.h.t. særlige behov i hjemmet M.h.t. information givet til patient og pårørende

M.h.t. symptomer og problemstillinger, der var sandsynlige i det videre forløb

M.h.t. diagnose og prognose

6

44487

13 Fik du tilfredsstillende information fra den hospitalsafdeling / det ambulatorium, der sendte patienten hjem til det palliative forløb? (Sæt venligst et kryds i hver række) Er ikke TilfredsOvervejende Overvejende Utilfredsrelevant stillende tilfredsutilfredsstillende for dette stillende stillende forløb

At kendskab til ansatte på hospitalsafdelingen betød/ ville have betydet noget positivt i forhold til indsatsen

At jeg savnede kendskab til ansatte på hospitalsafdelingen

At der var god faglig kontakt mellem mig og afdelingerne At der var en god "overlevering" af ansvaret for patienten ved udskrivelsen / afslutningen fra hospitalet At jeg skulle bruge uhensigtsmæssigt meget tid på at indhente information fra hospitalet

12 I hvilken grad er du enig i følgende udsagn? Under patientens palliative forløb oplevede jeg: (Sæt venligst et kryds i hver række) Helt Noget Noget Helt Ved ikke enig enig uenig uenig

Andet:

Ved ikke

Efter ambulant forløb på hospitalet

11 Hvordan startede patientens palliative forløb i hjemmet? Efter indlæggelse på hospitalsafdeling

Følgende spørgsmål handler om samarbejdet med hospitalsafdelinger i det palliative forløb (dog ikke palliative teams).


Nej

Godt

Rimeligt

Dårligt


Rigtig godt

Godt

7

Rimeligt

Dårligt

44487


16 Når du tænker på dette palliative forløb som helhed, hvordan vil du så karakterisere kvaliteten af hospitalets samlede indsats i relation til det palliative forløb (ikke palliative teams)? (Sæt venligst kun et kryds)

Rigtig godt

15 Når du tænker på dette palliative forløb som helhed, hvordan vil du så karakterisere dit samarbejde med hospitalet (ikke palliative teams)? (Sæt venligst kun et kryds)

Ja

14 Indhentede du supplerende informationer fra hospitalet ud over den information, der kom fra hospitalet?


Forbeholdt kodning

8

44487

Hvis du slet ikke har haft kontakt til den praktiserende læge, beder vi dig gå til spørgsmål 27 side 11

Andet, anfør hvad:

Den praktiserende læge var ikke tilgængelig

Den praktiserende læge var ikke interesseret i kontakt

Forløbet gik så hurtigt, at der ikke var tid til at kontakte lægen

Der var kun behov for pleje

Patienten eller pårørende ønskede det ikke

Patienten var allerede tilknyttet Det Palliative Team ved udskrivelsen

Nej - hvis nej beder vi dig anføre hvorfor: (Sæt gerne flere krydser)

17 Har du eller dine kollegaer været i forbindelse med den praktiserende læge i det palliative forløb? Ja

Følgende spørgsmål handler om samarbejdet med patientens praktiserende læge. Ved kompagniskab af læger handler det om den læge, der tog sig mest af forløbet.


At jeg savnede kendskab til den praktiserende læge At kendskab til den praktiserende læge betød / ville have betydet noget positivt i forhold til indsatsen

At den praktiserende læge var let at få fat i At der var tilstrækkelig telefonisk kontakt mellem den praktiserende læge og mig At den praktiserende læge var let at få til at komme på hjemmebesøg At den praktiserende læge havde for travlt til, at samarbejdet fungerede

At den praktiserende læge ikke ville involvere sig i det palliative forløb At den praktiserende læge var opdateret omkring palliativ behandling og pleje At der var en god dialog om behandlingsplanen At der var en god dialog om ansvarsfordelingen

At den praktiserende læge var tilstrækkelig engageret i forløbet

9

44487

18 I hvilken grad er du enig i følgende udsagn om den praktiserende læge, der var mest involveret i forløbet? Under det aktuelle palliative forløb oplevede jeg: (Sæt venligst et kryds i hver række) Ved ikke / ikke Helt Noget Noget Helt relevant enig enig uenig uenig


Nej

Ved ikke

Nej

Ved ikke

Nej

Ved ikke

Nej

Ved ikke

I nogen grad

I mindre grad

Slet ikke

Godt

Rimeligt

Dårligt


Rigtig godt

Godt

10

Rimeligt

Dårligt

44487


26 Når du tænker på dette palliative forløb som helhed, hvordan vil du så karakterisere kvaliteten af den praktiserende læges samlede indsats? (Sæt venligst kun et kryds)

Rigtig godt

25 Når du tænker på dette palliative forløb som helhed, hvordan vil du så karakterisere samarbejdet med den praktiserende læge? (Sæt venligst kun et kryds)

I høj grad

24 Har du savnet deltagelse af den praktiserende læge i tværfaglige møder i forbindelse med dette palliative forløb? (Sæt venligst kun et kryds)

23 Har du savnet (flere) tværfaglige møder i forbindelse med dette palliative forløb? (Sæt venligst kun et kryds) I høj I nogen I mindre Slet ikke grad grad grad

Ja

22 Deltog den praktiserende læge i tværfaglige møder i forbindelse med dette palliative forløb?

Ja

21 Blev der holdt tværfaglige møder i forbindelse med dette palliative forløb?

Ja

20 Aftalte den praktiserende læge med dig eller dine kollegaer, hvem der skulle kontaktes, når han / hun ikke var tilgængelig?

Ja

19 Gav den praktiserende læge sit private telefonnummer til patient, pårørende, dig eller andre i hjemmeplejen i det palliative forløb?


At jeg savnede kendskab til de ansatte i Det Palliative Team At kendskab til Det Palliative Team betød / ville have betydet noget positivt i forhold til indsatsen

At Det Palliative Team havde for travlt til, at samarbejdet fungerede At Det Palliative Team bestræbte sig på at bevare mig som en del af behandlerteamet At der var én kontaktperson i teamet, så jeg vidste, hvem jeg skulle kontakte At Det Palliative Team var organiseret på en god måde i forhold til vores samarbejde At det var et alvorligt problem, at Det Palliative Team ikke var tilgængeligt efter normal arbejdstid

At der var en god dialog om ansvarsfordelingen At Det Palliative Team var let at få fat i At der var tilstrækkelig telefonisk kontakt mellem Det Palliative Team og mig

At der var en god dialog om behandlingsplanen

At jeg blev hægtet af behandlingsforløbet

At jeg fik mere viden om palliation

At samarbejdet lettede mit arbejde

Helt enig

11

Noget enig

Noget uenig

Helt uenig

44487

Er ikke relevant for denne patient / ved ikke

27 I hvilken grad er du enig i følgende udsagn? Under samarbejdet med et palliativt team oplevede jeg: (Sæt venligst et kryds i hver række)

Hvis der ikke har været et palliativt team involveret i forløbet, beder vi dig gå til spørgsmål 31 Side 13

Følgende spørgsmål handler om samarbejdet med palliative teams


Overvejende Overvejende tilfredsutilfredsstillende stillende Utilfredsstillende

Godt

Rimeligt

Dårligt

Rigtig godt

12

Godt

Rimeligt

Dårligt

Er ikke relevant for dette forløb


44487


30 Når du tænker på dette palliative forløb som helhed, hvordan vil du så karakterisere kvaliteten af Det Palliative Teams samlede indsats? (Sæt venligst kun et kryds)

Rigtig godt

29 Når du tænker på dette palliative forløb som helhed, hvordan vil du så karakterisere samarbejdet med Det Palliative Team? (Sæt venligst kun et kryds)

M.h.t. informationen til patient og pårørende

M.h.t. bivirkninger til behandlingen

M.h.t. behandlingsplanen

M.h.t. patientens symptomer M.h.t. hvilke symptomer og problemstillinger man kunne forvente i det videre forløb

Tilfredsstillende

28 Har den information, du har fået fra et palliativt team, været tilfredsstillende? (Sæt venligst et kryds i hver række)


13

44487

32 Når du tænker på dette palliative forløb som helhed, hvordan vil du så karakterisere kvaliteten af din egen og dine kollegers (andre hjemmesygeplejerskers) samlede indsats? Ikke relevant (Sæt venligst kun et kryds) /ved ikke Rigtig Godt Rimeligt Dårligt Rigtig godt dårligt

31 Når du tænker på dette palliative forløb, hvordan vil du så karakterisere kvaliteten af primærsektorens (almen praksis og hjemmeplejen) samlede indsats? (Sæt venligst kun et kryds) Ikke relevant /ved ikke Rigtig Godt Rimeligt Dårligt Rigtig godt dårligt

Følgende spørgsmål handler om dette palliative forløb som helhed


H j e m (Ikk me ho ep s leje patie h j e m / nten Fas bes t bo ky t t l i e g t bo på lig) p leje hje m

I begyndelsen af det palliative forløb

Afla stn

ing sp l ad

Ved ikke

Nej

Ved ikke

Ja, vi har planlagt kommende kontakt

Aflastningsplads på plejehjem

14

44487

Hvis patienten ikke på noget tidspunkt ønskede at dø hjemme, jr. spørgsmål 33, gå direkte til spørgsmål 38 side 15


Hospice

Hospital

Hjemme hos pårørende

Ja, vi har haft telefonisk kontakt Ja, vi har haft personlig kontakt

15

Nej, men jeg/vi har opfordret den pårørende til at kontakte os ved behov

Nej, den pårørende ønskede det ikke

Nej

38 Har du eller din kollega haft kontakt med den nærmeste pårørende efter dødsfaldet? (Sæt kun et kryds)

Ja

Fast plads på plejehjem / pleje bolig - gå til spørgsmål 38 side 15

Forbeholdt kodning

Nej

Forbeholdt kodning

44487

37 Hvis patienten ikke døde hjemme, trods ønske herom: Havde du indtryk af, at det var et problem for de pårørende, at patienten døde et andet sted end ønsket?

Ja

36 Hvis patienten ikke døde hjemme trods ønske herom: Havde du indtryk af, at det var et problem for patienten at skulle dø et andet sted end ønsket?

Ved ikke


På grund af ønske fra hjemmeplejen

På grund af ønske fra de pårørende

På grund af ændret ønske fra patienten i den allersidste del af det palliative forløb

På grund af at patient eller pårørende kaldte 112

På grund af indlæggelse via vagtlæge

På grund af uforudsete klager / symptomer uden for normal arbejdstid

35 Hvis patienten ikke døde hjemme, trods ønske herom: Hvorfor døde han/hun ikke hjemme? (Sæt evt. flere krydser) På grund af uforudsete klager / symptomer i normal arbejdstid


Ældre bolig / beskyttet bolig - gå til spørgsmål 38 side 15

34 Hvor døde patienten? (Sæt kun et kryds) Hjemme (ikke plejehjem / ældrebolig / beskyttet bolig) - gå til spørgsmål 38 side 15

I slutningen af det palliative forløb

I midten af det palliative forløb

/ be sky tte t

bol ig

Ja - hvis ja, hvor? (Sæt venligst et kryds i hver række)

33 Udtrykte patienten over for dig et ønske om, hvor han/hun helst ville dø? Nej - hvis nej, gå til spørgsmål 34

Følgende spørgsmål handler om dødssted og ønske om dødssted


Ho spi ce And et

ste d , a nfø r ve

jeh jem

sp åp le ital

Ho sp

t: n l i g st h vilk e

Ve d ik ke

Noget enig

Noget uenig

Helt uenig

Ved ikke

Anden person, hvem:

16

Den sundhedsperson, der har været mest involveret i forløbet

Den sundhedsperson, som patient og pårørende ønsker

Den behandlende læge på sygehuset

En sygeplejerske fra den behandlende afdeling

En person fra Det Palliative Team


En hjemmesygeplejerske

44487

Forbeholdt kodning

40 Hvis der var en fast person fra sundhedssektoren, der skulle være "tovholder" i forløbet (have det overordnede ansvar for forløbet og koordinere samarbejdet mellem involverede fagpersoner), hvem mener du så primært, det skulle være? (Sæt kun et kryds)

Palliation af cancerpatienter er en så kompleks og vigtig opgave, at der bør oprettes et decideret team i primærsektoren, der kan stå for opgaven

Jeg føler ofte, at jeg i arbejdet med døende patienter mangler opbakning fra mine kolleger i hjemmeplejen Jeg har ikke tilstrækkelig tid til at behandle døende patienter i hjemmet Jeg oplever, at patienter føler sig overladt til sig selv, når de udskrives til et palliativt forløb Der er behov for en fælles palliationsjournal for både hospitalspersonale, praktiserende læger og hjemmesygeplejersker Hvis der skal være en fælles palliationsjournal for både hospitalspersonale, praktiserende læger og hjemmesygeplejersker, kan den med fordel være internet-baseret

Helt enig

39 I hvilken grad er du enig i følgende udsagn? (Sæt venligst et kryds i hver række)

Følgende spørgsmål handler om dit forhold til palliativ behandling generelt


17



44487

Forbeholdt kodning

År

År

År

-

- 2 0 0 7

Forbeholdt kodning


Med venlig hilsen

Jens Søndergaard Praktiserende læge Seniorforsker, lektor Ph.d.

18

Anders Bonde Jensen Overlæge, ph.d. Onkologisk afdeling D Århus Universitetshospital

44487

Frede Olesen. Praktiserende læge Forskningsleder Adj. professor, dr.med.

Hvis du er interesseret i at modtage e-mail om resultaterne af denne undersøgelse om ca. 1 år sendes en mail til: [email protected]

Mange tak for hjælpen!

Dato for udfyldelse af spørgeskemaet: (DD-MM-ÅÅÅÅ)

Anden, hvilken:

Har arbejdet i et palliativt team

Ja, hvis ja, hvilken: Netværkssygeplejerske i palliation udannet på Hospice Søholm

Nej

46 Har du eller er du i gang med en professionel palliationsuddannelse / erfaring?

45 Antal år du har arbejdet i hjemmeplejen i alt:

44 Antal år du har været færdig som sygeplejerske:

43 Alder:

Kvinde

42 Køn Mand

De sidste spørgsmål handler om dig selv:


Appendix N: Cover letter, Relatives (In Danish)

2007

«Løbenummer» «Pånavn» «Påadresse» «Påpostkode» «Påpostdistrikt»

Kære «Pånavn» Som aftalt sender jeg dig et nyt skema. Vi har fundet frem til dig som nær pårørende til «Ptnavn» Vi ved, at du går igennem en svær tid. Alligevel håber vi, at du har overskud til at hjælpe os med at udfylde dette spørgeskema. Vi vil gerne være med til at forbedre indsatsen overfor kommende uhelbredeligt syge kræftpatienter og deres pårørende. Vi vil derfor afdække behov, ønsker og mangler i den sidste tid før døden. Såfremt det ikke var dig, der var den mest involverede pårørende, bedes du videregive skemaet til den pårørende, der var mest involveret i forløbet. Vi har bl.a. spurgt hjemmeplejen og praktiserende læger, om de mente, at vi kunne sende disse spørgeskemaer. Vi håber ikke, at det kan virke stødende på nogen måde. Skemaerne bliver anonymiseret således, at læger, sygeplejersker og andre personalegrupper ikke får at vide, hvad du har svaret. Kun nedenstående forskergruppe får indblik i de enkelte besvarelser. Spørgeskemaundersøgelsen er godkendt af Datatilsynet og har været forelagt den Videnskabsetiske Komité for Århus Amt. Når skemaet er udfyldt ligges det i den vedlagte frankerede svarkuvert og sendes. Hvis du har problemer med at udfylde skemaet selv, må du gerne få hjælp af en anden pårørende. Har du spørgsmål eller kommentarer, er du meget velkommen til at kontakte os. Vi sender et brev for at minde om spørgeskemaet ca. en uge efter fristen, så hvis du ikke ønsker at deltage, kan skemaet sendes uudfyldt til os, så undgår du at modtage påmindelsen. Hvis du ikke er pårørende til ovennævnte person, undskylder vi, og beder dig om at kontakte os. Der er plads til kommentarer på den sidste side i skemaet. Med venlig hilsen

Mette Asbjørn Neergaard Projektansvarlig, læge Forskningsenheden for Almen Praksis i Århus Tlf: 8942 6028 Email: [email protected]

Jens Søndergaard Praktiserende læge Seniorforsker Lektor, ph.d. Forskningsenheden for Almen Praksis

Anders Bonde Jensen Overlæge, ph.d. Onkologisk afd. D Århus Universitetshospital

Frede Olesen Forskningsleder Praktiserende læge Professor, Dr.med. Forskningsenheden for Almen Praksis

Appendix O: Questionnaire, Relatives (In Danish)

2007

Pårørendeskema



Løbenummer xxxx

Ja (0 = ingen børn)

Forbeholdt kode

(0 = ingen børn)

Nej

Andet:

1

Ja, lang videregående uddannelse mere end 4 år

Ja, mellemlang videregående uddannelse 3-4 år

Ja, kort videregående uddannelse under 3 år

Ja, faglært inden for håndværk, handel, kontor mv.

Ja, et eller flere kortere kurser (specialarbejderkurser, arbejdsmarkedskurser)

13327

Forbeholdt kode

4 Havde den afdøde fuldført en erhvervsuddannelse? (Sæt kun et kryds) (Hvis den afdøde har gennemført flere erhvervsuddannelser, skal du kun oplyse den længst varende)

Samlevende med familiemedlem - angiv hvem:

Samlevende med kæreste

Gift

Enlig (enke / enkemand)

Enlig (skilt, separeret, afbrudt fast samlivsforhold)

Enlig (ikke tidligere gift eller samlevende)

3 Hvad var den afdødes ægteskabelige status? (Sæt kun et kryds)

2 Hvor mange af den afdødes børn under 25 år boede i den afdødes husstand?

1 Hvor mange børn havde den afdøde? (skriv antal)

Nej

0 Boede den afdøde på plejehjem inden han / hun blev erklæret uhelbredelig syg?

I dette spørgeskema har vi af praktiske grunde kaldt Pt-navn for "den afdøde." Vi håber ikke, at dette kan virke stødende på nogen måde.


Ja, hvis ja, - hvilken profession:

Forbeholdt kode

Antal uger(< 1 uge skrives som 0)

2

uger

13327

6A Hvor mange uger skønner du, den afdøde sammenlagt var i hjemmet i den sidste tid (enten i eget hjem, eller i pårørendes hjem, hvis det var der, plejen foregik)?

Udtrykket "Den sidste tid" bruger vi herefter, om den periode, hvor den afdøde var tydeligt døende af sin cancer sygdom.

Nej

6 Arbejdede eller havde den afdøde arbejdet i sundhedsvæsenet, før den afdøde blev uhelbredelig syg?

Pensionist / efterlønsmodtager

Førtidspensionist

På omskoling eller revalidering

Sygemeldt af anden årsag end kræftsygdommen

Sygemeldt pga. kræftsygdommen

Arbejdsløs

Hjemmegående

På orlov (fx barselsorlov eller uddannelsesorlov)

I fleksjob eller lignende

Under uddannelse

På nedsat arbejdstid (< 37 timer)

I fuldtidsarbejde (37 timer eller derover)

5 Hvordan var den afdødes erhvervsmæssige situation, før den afdøde blev uhelbredelig syg? (Sæt evt. flere krydser)


Den afdøde gav udtryk for, at hospitalet burde havde givet mere information om psykiske aspekter (Fx om bekymringer, psykiske reaktioner, krise og sorg) ved afslutningen fra hospitalet Den afdøde gav udtryk for, at hospitalet burde havde givet mere information om sociale aspekter(Fx om forholdet til den afdøde, familie og venner) ved afslutningen fra hospitalet Den afdøde gav udtryk for at føle sig ladt i stikken af fagpersonerne lige efter afslutning fra hospitalet Det var klart, hvor der skulle søges hjælp og vejledning lige efter afslutning fra hospitalet

Den afdøde gav udtryk for, at hospitalet burde havde givet mere information om lægelige aspekter ( Fx om behandling, medicin og sygdommen) ved afslutningen fra hospitalet

Hospitalet kunne have gjort mere for at sikre, at de praktiske ting var i orden i hjemmet inden hjemsendelsen

Meget enig

3

Noget enig

Noget uenig

Meget uenig

7 Er du enig i følgende udsagn? (Sæt venligst et kryds i hver vandrette række)

13327

Kan ikke svare

De følgende 2 sider handler om tiden lige før og efter udskrivelsen / afslutningen fra hospitalet, hvor den afdøde var blevet erklæret uhelbredeligt syg


Jeg følte mig ladt i stikken lige efter afslutningen fra hospitalet Jeg ville ønske, at der havde været lidt mere hjælp til at håndtere den sygdomsmæssige side (behandling og pleje) lige efter afslutning fra hospitalet Jeg ville ønske, at der havde været lidt mere hjælp til at håndtere alle mine bekymringer lige efter afslutning fra hospitalet Jeg ville ønske, at der havde været lidt mere hjælp til at håndtere alle de praktiske ting(f.eks. rengøring og madlavning) lige efter afslutning fra hospitalet

Jeg synes, at hospitalet burde have givet mere information om lægelige aspekter (f.eks. om behandling, medicin og sygdommen) inden afslutningen fra hospitalet Jeg synes, at hospitalet burde have givet mere information om psykiske aspekter (f.eks. om bekymringer, psykiske reaktioner, krise og sorg) inden afslutningen fra hospitalet Jeg synes, at hospitalet burde have givet mere information om sociale aspekter (f.eks. om forholdet til den afdøde, familie og venner) inden afslutningen fra hospitalet

Meget enig

4

Noget enig

Noget uenig

8 Er du enig i følgende udsagn? (Sæt venligst et kryds i hver vandrette række)


Meget uenig

13327

Kan ikke svare

Noget enig

Noget uenig

Meget uenig

Kan ikke svare

Aldrig

godt

5

dårligt

13327

/ved ikke

12 Hvordan vil du karakterisere kvaliteten af hjemmehjælpernes indsats i den sidste tid? (Sæt kun et kryds) Rigtig Godt Rimeligt Dårligt Rigtig Ikke relevant

11 Hvordan vil du karakterisere kvaliteten af hjemmesygeplejerskernes indsats i den sidste tid? (Sæt kun et kryds) Rigtig Godt Rimeligt Dårligt Rigtig Ikke relevant godt dårligt /ved ikke

Hospitalet havde informeret hjemmeplejen tilstrækkeligt om sygdom og behandling Hjemmesygeplejerskerne var lette at få fat i Hjemmesygeplejerskerne var engagerede i forløbet Hjemmesygeplejerskerne var opdaterede i deres viden om det aktuelle sygdomsforløb Hjemmesygeplejerskerne var opdaterede i deres viden om behandling af døende patienter generelt

Hospitalet havde informeret hjemmeplejen tilstrækkeligt om praktiske behov i hjemmet (f.eks. plejeseng eller andre hjælpemidler)

Meget enig

10 Er du enig i følgende udsagn (Sæt venligst et kryds i hver vandrette række)

Hjemmehjælpere

Hjemmesygeplejersker

9 Hvor ofte kom hjemmeplejen i hjemmet, i den sidste tid? (Sæt venligst et kryds i hver vandrette række) Flere Mindre end Hver dag Flere 1 gang gange om 1 gang om gange om om ugen dagen ugen ugen

Hvis der slet ikke var hjemmesygeplejersker eller hjemmehjælpere involveret i den sidste tid - gå venligst direkte til spørgsmål 13 side 6

Denne side handler om hjemmeplejen, d.v.s. hjemmesygeplejerskerne og hjemmehjælperne


Nej

Forbeholdt kode

Hjemmebesøg

Konsultation hos lægen

Telefon konsultation

6

13327

17 Hvor meget kontakt havde den afdøde, du eller andre pårørende med den praktiserende læge i den sidste tid? (Sæt venligst et kryds i hver vandrette række) Flere Mindre end Hver dag Flere 1 gang Aldrig gange om 1 gang om gange om om ugen dagen ugen ugen

Hvis den praktiserende læge slet ikke var involveret i den sidste tid - gå direkte til spørgsmål 22 side 8

Andet

Det var ikke nødvendigt, at den praktiserende læge skulle deltage

Vi ønskede ikke, at den praktiserende læge skulle deltage

Nej, hvis nej, hvorfor? Den praktiserende læge var ikke interesseret i at deltage

Kendte ikke lægen 16 Var den praktiserende læge involveret i den sidste tid? Ja

Ikke særlig godt

Nogenlunde

Godt

15 Hvordan vil du karakterisere dit kendskab til den afdødes praktiserende læge før den sidste tid? (Sæt kun et kryds) Meget godt

Kendte ikke lægen

Ikke særlig godt

Nogenlunde

Godt

14 Hvordan vil du karakterisere den afdødes kendskab til sin praktiserende læge før den sidste tid? (Sæt kun et kryds) Meget godt

De følgende spørgsmål handler om den afdødes praktiserende læge.

Ja

13 Havde du og den afdøde den samme praktiserende læge?

De følgende 3 sider handler om den praktiserende læge


Noget enig

Noget uenig

Meget uenig

Kan ikke svare

Forbeholdt kode

7

13327

20 Hvordan vil du karakterisere kvaliteten af den praktiserende læges indsats i den sidste tid? (Sæt kun et kryds) Rigtig Godt Rimeligt Dårligt Rigtig godt dårlig

Den praktiserende læge vidste nok om behandling af døende patienter generelt Jeg var tryg ved samarbejdet mellem den praktiserende læge og hjemmesygeplejerskerne Den praktiserende læge gjorde en indsats for at få samarbejdet med hjemmesygeplejerskerne til at fungere Før den sidste tid, havde jeg fuldstændig tillid til den praktiserende læges faglige kunnen og dømmekraft Nu har jeg fuldstændig tillid til den praktiserende læges faglige kunnen og dømmekraft

Den praktiserende læge var let at få fat i Den praktiserende læge var engageret i forløbet Den praktiserende læge var let at få til at komme på hjemmebesøg Den praktiserende læge kom på hjemmebesøg af sig selv, uden vi skulle bede om det Den praktiserende læge vidste nok om den afdødes sygdom

Meget enig

19 Er du enig i følgende udsagn om den praktiserende læge i den sidste tid? (Sæt venligst et kryds i hver vandrette række)

Andet:

Hjemmebesøg

Tid i konsultation

Telefonkonsultation

Ja, hvis ja beder vi dig anføre, hvilken aftale I bad om: (Sæt evt. flere krydser)

18 Var der tilfælde hvor den afdøde, du eller andre pårørende ikke kunne få en aftale med den praktiserende læge, selvom I kom igennem til lægens praksis i telefonen? Nej


Noget enig

Noget uenig

Jeg var tryg ved samarbejdet mellem forskellige sygehuse og sygehusafdelinger

Jeg oplevede, at der var et godt samarbejde mellem forskellige sygehuse og sygehusafdelinger

Meget enig

8

Noget enig

Noget uenig

22 Hvordan oplevede du samarbejdet inden for hospitalsvæsenet.? Er du enig i følgende udsagn? (Sæt venligst et kryds i hver vandret række)

Jeg var tryg ved samarbejdet mellem den praktiserende læge og hospitalslægerne Det virkede som om, hospitalslægerne havde hægtet den praktiserende læge af Hospitalslægerne holdt den praktiserende læge orienteret om, hvordan det gik med behandlingen. Alt i alt var jeg tilfreds med samarbejdet mellem den praktiserende læge og hospitalslægerne

Jeg ville ønske, der havde været et bedre samarbejde mellem hospitalslægerne og den praktiserende læge

Meget enig

Meget uenig

Meget uenig

13327

Kan ikke svare

Kan ikke svare

21 Hvordan oplevede du samarbejdet mellem hospitalet og den praktiserende læge.? Er du enig i følgende udsagn? (Sæt venligst et kryds i hver vandrette række)


Nej

Hver dag

Flere gange om ugen

Noget enig

Noget uenig

1 gang om ugen

Meget uenig

Mindre end 1 gang om ugen

Kan ikke svare

Aldrig

9

Rigtig god

God

Rimelig

Dårlig

13327

Rigtig dårlig

26 Hvordan vil du karakterisere kvaliteten af det palliative teams indsats? (Sæt kun et kryds)

Det palliative team var let at få fat i Det var et problem, at vi ikke kunne kontakte det palliative team uden for normal arbejdstid Det palliative team ville gerne samarbejde med hjemmesygeplejersken Det palliative team vejledte hjemmesygeplejersken Alt i alt var jeg tilfreds med samarbejdet mellem det palliative team og hjemmesygeplejersken Det palliative team ville gerne samarbejde med den praktiserende læge Alt i alt var jeg tilfreds med samarbejdet mellem det palliative team og den praktiserende læge

Meget enig

25 Er du enig i følgende udsagn om det palliative team? (Sæt venligst et kryds i hver vandrette række)

Hjemmebesøg

Telefonisk kontakt

Flere gange om dagen

24 Hvor ofte havde du, den afdøde eller andre pårørende kontakt med det palliative team i den sidste tid? (Sæt venligst et kryds i hver vandrette række)

Hvis der ikke var et palliativt team involveret i det terminale forløb - gå til spørgsmål 27 side 10

Ja

23 Var der et palliativt team involveret i det terminale forløb?

Denne side handler om palliative team (specialiserede team, der er specialister i behandling af døende patienter)


Ja, hvis ja, hvor? (Sæt venligst et kryds i hver vandrette række)

Nej, hvis nej, gå til spørgsmål 29 side 11

28 Udtrykte den afdøde overfor dig, et ønske om, hvor han/hun helst ville dø?


Hospice

Hospital

Hjemme hos mig (hvis du ikke delte bolig med den afdøde)

Aflastningsplads / korttidsplads på plejehjem

Fast bolig på plejehjem / beskyttet bolig

27 Hvor døde den afdøde? (Sæt kun et kryds) Hjemme hos den afdøde (ikke plejehjem / beskyttet bolig)

I slutningen af det terminale forløb

I midten af det terminale forløb

I begyndelsen af det terminale forløb

10

Ho spi ce And et

13327

Forbeholdt kode

Ja, hvis ja, hvor? (Sæt venligst et kryds i hver vandrette række)

Nej, hvis nej, gå til spørgsmål 30

29 Havde du et foretrukket sted, hvor døden skulle finde sted inden den afdødes død?


11

Hvis den afdøde døde det ønskede sted, gå direkte til spørgsmål 34 side 13

Ved ikke


Hospice

Hospital

Hjemme hos dig (hvis du ikke delte bolig med den afdøde)

Aflastningsplads / korttidsplads på plejehjem

Fast bolig på plejehjem / beskyttet bolig

Hjemme hos den afdøde (ikke plejehjem / beskyttet bolig)

30 Hvis du kunne ønske i dag, hvor ville du så ønske, at døden havde fundet sted? (Sæt kun et kryds)

I slutningen af det terminale forløb

I midten af det terminale forløb

I begyndelsen af det terminale forløb

Ho spi ce And et

De følgende 2 sider handler om det sted, døden fandt sted, eller det ønskede dødssted

ve s t e d, a nfø r

ve ste d, a nfø r


H j e m (Ikk me ho ep s leje patie h j e m / nten Fa s bes tb ky o t t l ig p e t bo åp lig) leje A f h l a j em stn ing / be sp s l k a y d tte t bol leje ig hje m

sp åp

ital

Ho sp

: nlig s t hvi lke t

ke

Ve d ik

H j e m (Ikk me ho ep s leje pati h j em enten / be Fas sk tb y o t t l ig p et b åp olig lej ) A e f h l a jem stn ing / be sp s l k a y d tte t bol leje ig hje m sp åp

ital

Ho sp

t: nlig st h vilk e

ke Ve d ik 13327

Forbeholdt kode

En del

Lidt

Slet ikke

Meget

12

En del

Lidt

Slet ikke

33 Hvis den afdøde ikke døde det ønskede sted: Hvor ked var du af, at afdøde døde et andet sted end ønsket? (Sæt kun et kryds)

Meget

32 Hvis den afdøde ikke døde det ønskede sted: Hvor ked var den afdøde af at skulle dø et andet sted end ønsket? (Sæt kun et kryds)

Ved ikke


Pres fra fagpersoner, anfør hvem:

Det palliative team kunne ikke behandle tilstrækkeligt

Hjemmeplejen kunne ikke behandle tilstrækkeligt

Vagtlæge kunne ikke behandle tilstrækkeligt

Den praktiserende læge kunne ikke behandle tilstrækkeligt

Ønske fra dig og / eller de andre pårørende

31 Hvis den afdøde ikke døde det ønskede sted: Hvorfor gjorde den afdøde ikke det? ( Sæt evt. flere krydser) Ændret ønske fra patienten umiddelbart inden døden


13327

Ved ikke

Ved ikke

Forbeholdt kode

Jeg tog aktivt del i behandlingen (f.eks. medicingivning, sårskift) Jeg tog mere del i pleje og / eller behandling, end jeg havde overskud til Det, at jeg tog del i pleje og / eller behandling gjorde, at jeg havde svært ved at få sagt farvel Der var ingen fagpersoner, der snakkede med mig om min deltagelse i pleje og behandling Hvis jeg kunne skrue tiden tilbage, ville jeg tage mindst lige så meget del i behandling og pleje, som jeg gjorde

Jeg tog aktivt del i plejen og pasningen (f.eks. sengebadning, daglig hygiejne, madning)

13

34 Er du enig i følgende udsagn om den sidste tid? (Sæt venligst et kryds i hver vandrette række) Meget enig Noget enig

Noget uenig

Meget uenig

De følgende spørgsmål handler om din deltagelse og oplevelse af den sidste tid


13327

Kan ikke svare

Forbeholdt kode

Forbeholdt kode

Den praktiserende læge Ansatte i Det Palliative Team


Ansatte i Det Palliative Team

15

Anden person, anfør venligst hvilken:


14

Sygeplejerske eller anden ansat på hospital (ikke Det Palliative Team)


13327

Hjemmehjælperne

Hjemmehjælperne

Læge på hospitalsafdeling (ikke Det Palliative Team)

Hjemmesygeplejerskerne



Andre pårørende / familiemedlemmer Venner / bekendte / naboer

Andre pårørende / familiemedlemmer

Venner / bekendte / naboer

Den afdøde

38 Hvem savnede du til at tage sig af dit behov og ønske om psykisk støtte i forløbet? (Sæt gerne flere krydser) Ingen





Den afdøde

36 Hvem savnede du til at tage sig af dit behov og ønske om fysisk hjælp i forløbet? (Sæt gerne flere krydser) Ingen






Hjemmehjælperne



Hjemmehjælperne



Den afdøde

13327

Forbeholdt kode

Forbeholdt kode

37 Hvem snakkede med dig om dit behov og ønske om psykisk støtte i forløbet? (Samtaler, støtte til samtale med den afdøde, m.v.) (Sæt gerne flere krydser) Ingen





Den afdøde

35 Hvem snakkede med dig om dit behov og ønske om fysisk hjælp i forløbet (rengøring, aflastning i hjemmet, m.v.)? (Sæt gerne flere krydser) Ingen


Hjemmesygeplejerskerne Hjemmehjælperne Den praktiserende læge Ansatte i Det Palliative Team


Hjemmehjælperne




16





Hjemmehjælperne




Den afdøde

13327

Forbeholdt kode

Forbeholdt kode


17





Hjemmehjælperne



Dig selv

Den afdøde

13327

Forbeholdt kode

42 Hvem, synes du, burde have været tovholder / central person (koordinerede pleje, behandling og samarbejde mellem alle involverede parter) i forløbet? (Sæt kun et kryds) Ingen



40 Hvem savnede du til at tage sig af behov for juridisk og økonomisk støtte? (Sæt gerne flere krydser) Ingen





Forbeholdt kode

Dig selv




Den afdøde

41 Hvem, følte du, var tovholder / central person (koordinerede pleje, behandling og samarbejde mellem alle involverede parter) i forløbet? (Sæt kun et kryds) Ingen


Den afdøde

39 Hvem snakkede med dig om behov for juridisk og økonomisk støtte? (Pension, terminalydelse, forsikring, m.v.) (Sæt gerne flere krydser) Ingen


Kan ikke svare

God

Rimelig

Dårlig

Rigtig dårlig

18

Rigtig godt

Godt

Rimeligt

Dårligt

13327

Rigtig dårligt

45 Hvordan vil du karakterisere den samlede tid hjemmet, hvor afdøde var døende, i forhold til, hvordan du mener, det burde have været? (Sæt kun et kryds)

Rigtig god

44 Hvordan vil du karakterisere kvaliteten af det samlede sundhedsvæsens indsats (hospital, hjemmeplejen, den praktiserende læge, mv.)? (Sæt kun et kryds)

Nu har jeg fuldstændig tillid til det samlede sundhedsvæsen (hjemmeplejen, egen læge, ansatte på sygehuse, vagtlæger, mv.)

Før den sidste tid, havde jeg fuldstændig tillid til det samlede sundhedsvæsen (hjemmeplejen, egen læge, ansatte på sygehuse, vagtlæger, mv.)

43 Er du enig i følgende udsagn? (Sæt venligst et kryds i hver vandrette række) Meget Noget Noget Meget enig enig uenig uenig


Forbeholdt kode

Forbeholdt kode

Nej

Ja

19

13327

48 Føler du, at du har fået tilstrækkelig hjælp og støtte af fagpersoner i tiden efter forløbet?

Anden person, hvem?

Ansatte på hospice

Præst

Ansatte på hospitalet

Det palliative team

Din egen praktiserende læge

Afdødes praktiserende læge

Hjemmehjælperne


47 Hvilke af de fagpersoner, der var involveret i forløbet, har tilbudt dig kontakt i tiden efter forløbet? (Sæt gerne flere krydser) Ingen

Anden person, hvem?

Ansatte på hospice

Præst


Det palliative team



Hjemmehjælperne


46 Hvilke af de fagpersoner, der var involveret i forløbet, har du haft kontakt til i tiden efter forløbet? (Sæt gerne flere krydser) Ingen

De næste to sider handler om tiden efter den afdødes død


Anden person, hvem?

Psykolog uden relation til forløbet

Ansatte på hospice

Præst


Det palliative team



Hjemmehjælperne


20

13327

Forbeholdt kode

49 Hvilke fagpersoner mener du, der burde være en støtte for dig som pårørende i tiden efter forløbet?(Sæt evt. flere krydser) Ingen


1 9

Kvinde

Mand

(0 = ingen børn) (0 = ingen børn)

Forbeholdt kode

Nej

Andet:

21

Ja, lang videregående uddannelse mere end 4 år

Ja, mellemlang videregående uddannelse 3-4 år

Ja, kort videregående uddannelse under 3 år

Ja, faglært inden for håndværk, handel, kontor mv.

Ja, et eller flere kortere kurser (specialarbejderkurser, arbejdsmarkedskurser)

13327

Forbeholdt kode

57 Har du fuldført en erhvervsuddannelse ? (Sæt kun et kryds) (Hvis du har gennemført flere erhvervsuddannelser, skal du kun oplyse den længst varende)

Samlevende

Gift

Enlig (enke, enkemand)

Enlig (skilt, separeret, afbrudt fast samlivsforhold)

Enlig (ikke tidligere gift eller samlevende)

56 Hvad er din ægteskabelige status nu? (Sæt kun et kryds)

55 Hvor mange af dine børn under 25 år bor i din husstand? (skriv antal)

54 Hvor mange børn har du? (skriv antal)

Nej

53 Boede du i sammen med den afdøde i den sidste tid? Ja

Andet:

Ven

Forælder

Søster / Bror

Datter / Søn


Ægtefælle/samlever

52 Hvilken relation har du til den afdøde? (Sæt kun et kryds)

51 Hvilket årstal er du født?

50 Er du kvinde eller mand?

De følgende 2 sider handler om dig selv


Under uddannelse

Under uddannelse

På omskoling eller revalidering Førtidspensionist Pensionist / efterlønsmodtager


På omskoling eller revalidering

Førtidspensionist

Pensionist / efterlønsmodtager

Ja, hvis ja, - hvilken profession:

Andet:

Jeg meldte mig syg

Jeg tog ferie

Ja, hvis ja - hvordan? Jeg tog plejeorlov

22

60 Tog du fri fra arbejde, fordi du skulle passe den afdøde: Nej

Nej

13327

Forbeholdt kode



64 Dato for udfyldelse af skemaet: (F.eks. 5. februar skrives 05 / 02 - 2007)

Dårligt

Mindre godt

Godt

Vældigt godt

23

/

Dårligt 63 Hvordan synes du, dit eget helbred er i dag alt i alt? (Sæt kun et kryds) Fremragende

Mindre godt

Godt

Vældigt godt

- 2 0 0 7

62 Hvordan synes du, dit eget helbred var, før den afdøde blev uhelbredeligt syg? (Sæt kun et kryds) Fremragende


Arbejdsløs

59 Arbejdede du i sundhedsvæsenet, før den afdøde blev uhelbredelig syg?

Hjemmegående Arbejdsløs

På orlov (fx barselsorlov eller uddannelsesorlov)

Hjemmegående

I fleksjob eller lignende På orlov (fx barselsorlov eller uddannelsesorlov)

I fleksjob eller lignende

Forbeholdt kode


61 Hvordan er din erhvervsmæssige situation nu? (Sæt evt. flere krydser) I fuldtidsarbejde (37 timer eller derover)



58 Hvordan var din erhvervsmæssige situation før afdødes sidste tid? (Sæt evt. flere krydser) I fuldtidsarbejde (37 timer eller derover)


13327

Forbeholdt kodning


Med venlig hilsen

Tusind tak for hjælpen

Jens Søndergaard Praktiserende læge Seniorforsker Lektor, ph.d. Forskningsenheden for Almen Praksis Århus

24

Anders Bonde Jensen Overlæge, ph.d. Onkologisk afdeling D Århus Sygehus

13327

Frede Olesen Forskningsleder Praktiserende læge Adj. professor Dr.med. Forskningsenheden for Almen Praksis Århus

Hvis du er interesseret i at modtage en e-mail om resultaterne af denne undersøgelse om ca. 1 år sendes mail til: [email protected]

