ARTICLE IN PRESS Respiratory Medicine (2006) 100, 1254–1261
Defining and investigating difficult asthma: Developing quality indicators C.O. Prys-Picarda,, S.M. Campbellb, J.G. Ayresc, J.F. Milesd, R.M. Nivena, for the Consensus on Difficult Asthma Consortium UK (CODAC-UK)1 a
North West Lung Research Centre, Wythenshawe Hospital, Southmoor Road, Wythenshawe, Manchester M23 9LT, UK b National Primary Care Research and Development Centre, Williamson Building, University of Manchester M13 9PL, UK c Department of Environmental and Occupational Medicine, University of Aberdeen, Liberty Safe Work Research Centre, Foresterhill Road, Aberdeen AB25 2ZP, UK d North Manchester General Hospital, Delaunays Road, Manchester M8 5RB, UK Received 5 April 2005; accepted 15 October 2005
KEYWORDS Difficult; Asthma; Consensus techniques; RAND appropriateness method
Summary Background: There is no agreed definition of ‘difficult asthma’ or what investigations should be available to investigate these patients. Patients with difficult asthma remain symptomatic on high levels of treatment and are high users of medical resources. Aim: To develop a set of quality indicators for the definition and investigation of difficult asthma. Method: Modified RAND Appropriateness Method was used. An expert panel composed of nine hospital asthma specialists who run ‘difficult’ asthma clinics and were identified from a shortlist of key workers in the field. Indicators were rated as necessary to define and investigate difficult asthma. Results: Difficult asthma was defined as ‘symptoms persisting beyond therapy consistent with step 4 of the British Thoracic Society (BTS) guidelines’ (high dose inhaled corticosteroids and long acting b2-agonists). Eighty-three indicators were identified (40 relating to definition and 43 relating to investigations). Of these 32 (39%) were rated as necessary: 7 out of 40 (18%) for defining difficult asthma and 23 out of 43 (53%) for investigations. Indicators of high medical resource usage were
Corresponding author. Tel.: +0161 291 5050; fax: +0161 291 2550.
E-mail address:
[email protected] (C.O. Prys-Picard). The CODAC panel consisted of Prof. Jon Ayres-Aberdeen, Prof. Neil Barnes-London, Dr. Christine Bucknall-Glasgow, Dr. Brian Harrison-Norwich, Dr. Bernard Higgins-Newcastle, Dr. Adel Mansur-Birmingham, Dr. Jon Miles-Manchester, Dr. Rob Niven-Manchester, and Dr. Ian Pavord-Leicester. 1
0954-6111/$ - see front matter & 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.rmed.2005.10.013
ARTICLE IN PRESS Defining and investigating difficult asthma: Developing quality indicators
1255
characteristic of the ‘difficult’ nature of the management of patient with difficult asthma. A framework for the investigation of these patients was created. Conclusion: The listed performance indicators identify a range of requirements that are necessary to define difficult asthma. Targeting of real needs in this group of patients will lead to better patient care and reduction of ‘waste’ in provision of healthcare. & 2005 Elsevier Ltd. All rights reserved.
Introduction Asthma literature is filled with an array of terminology to describe patients who remain symptomatic despite high levels of therapy. These include severe asthma,1 refractory asthma,2 steroid resistant asthma,3 steroid insensitive asthma,4 therapy resistant asthma,5 persistent asthma,6 difficult-to-treat asthma1 and difficult asthma7; terms which have often been used interchangeably. The TENOR study1 recruited patients who were considered to have severe or difficult-to-treat asthma. Of this cohort only 48% of patients were felt to have severe asthma by their physicians leaving half labelled ‘difficult asthma’ with lesser degrees of asthma severity. The variety of terms that exist need to be clarified so that all authors can use an agreed terminology. Severe asthma refers to those patients with physiologically severe disease that remains unresponsive to high dose therapy, often implying regular or frequent oral corticosteroid therapy. A number of empirical grading scales exist for severe asthma8–10 on the presumption that it is the only disease process being measured and is the only condition causing symptoms. These are based upon symptoms, fixed physiological criteria and therapeutic thresholds. Difficult asthma refers to asthma of mild, moderate or severe physiological severity that may co-exist with another condition, unresponsive to conventional asthma therapy. This secondary condition may be structural as in bronchiectasis, functional as in dysfunctional breathing or vocal cord dysfunction, or psycho-social in origin. This secondary condition is responsible for the prolongation of respiratory symptoms which are then wrongly attributed as being due to asthma. Without appropriate management of both the asthma and any other co-morbid condition, these patients remain symptomatic and become high users of medical resources with frequent GP attendances and hospital admissions.11,12 There is often an escalation of inappropriate therapy when the co-
morbid condition goes unrecognised, and the severity of asthma is over-estimated. There are a number of specialist asthma clinics in the UK that accept referrals for difficult or severe asthma. The results of a structured approach to assessment of such patients have recently been published.5,13 These studies are both observational in nature. Robinson et al. identified that 32% of patients had alternative or additional diagnoses, psychiatric illness or non-compliance with therapy. Heaney et al. made similar observations but went further by identifying predictors of true therapy resistance (inhaled corticosteroid dose 42000 mg BDP equivalent, previous assessment by a respiratory specialist and FEV1 at referral o70% predicted). It has not been widely agreed which facilities should be available for the investigation of these patients and practice is often anecdotal and dependant on local availability. Good-quality evidence (randomised double blind placebo controlled trials) does not exist in this field of diagnostic testing due to the heterogeneity of the patients and subsequent difficulty in performing sufficiently large trials. In situations where evidence is sparse or uncertain14–16 consensus techniques, which explore consensus among a group of experts by synthesising opinions in combination with available evidence, are an increasingly important mechanism for developing quality indicators.17 This study used a consensus technique to define ‘difficult asthma’ and to determine which facilities should be available for the investigation of patients with difficult asthma.
Methods We used a modified RAND Appropriateness Method,18,19 which has been described as the only systematic method of combining expert opinion and evidence.15 It combines characteristics of both the Delphi Technique (e.g. postal questionnaire) and the Nominal Group Technique (e.g. face-toface meeting).17
ARTICLE IN PRESS 1256
C.O. Prys-Picard et al.
Questionnaire compilation A list of 83 indicators was generated based on previous recommendations of the ERS Task Force on Difficult/Therapy resistant asthma20 and the Proceedings of the American Thoracic Society Workshop on refractory asthma.2 Indicators were grouped according to two key issues: (1) How important are certain indices in defining difficult asthma? (40 indicators) and (2) The importance of specific investigations in the assessment of difficult asthma (43 indicators).
Expert panel formation The expert panel was drawn from those clinicians running severe or difficult asthma clinics in England and Scotland, authors of key trials and recognised experts, and was selected to provide geographical balance throughout the United Kingdom. A shortlist of 11 panellists was invited to participate. All agreed but only 9 (82%) were available for the panel meeting (see Acknowledgements). Panel sizes between 9 and 12 members provide results that have been shown to be reproducible by second panels19 whilst facilitating group discussion and preventing the group from becoming too unwieldy.21
Round 1–postal round The list of indicators was sent to each panellist by post. Panellists were asked, without reference to each other, to rate the indicators. Standard Likert scales were used consisting of continuous integer scales of 1–9, where 1 signified the indicator was
Box 1
absolutely unnecessary and 9 meant it was absolutely necessary.
Round 2–expert panel meeting At the meeting panellists were given an anonymised summary of the panel’s first round scores and each panellist was fed back their own response from round one for each indicator. Panellists were given the opportunity to discuss any indicator they wished to. After discussing the relevant issues, each indicator was re-rated. At the panel meeting the standard Likert scale of 1 (absolutely unnecessary) to 9 (absolutely necessary) was considered insufficiently sensitive for rating indicators relating to the issues of difficult asthma. The panel meeting is a dynamic process, which allows stem questions and indicators to be clarified or modified. This process ensures that any ambiguities of meaning and important omissions are minimised as recommended by Baker.22 The Likert scales were discussed and subsequently refined (Box 1). The indicators were rated according to the following terms of reference. ‘Difficult’ is from the physicians’ perspective, and the indicators represent an ‘ideal’ rather than what is necessarily currently available to all physicians. Consensus was defined as existing where the median score from the overall panel was 7 or greater without disagreement.19 Disagreement was defined as where at least 33% of the panel members rated in both the upper(7–9) and lower(1–3) tertile. Indicators rated 8 or 9 in a RAND method have been found to be more reproducible if the same indicators were rated by a panel composed of
Revised rating scales used in the second round
Rating scale used for diagnosing difficult asthma 9 Pathognomic 7,8 Necessary criteria 4,5,6 Equivocal, unproven or undecided necessity 2,3 Not a necessary criteria 1 No relevance to difficult asthma Rating scales used to assess the necessity of investigations in the assessment of difficult asthma 9 Absolutely necessary for all patients 7,8 Necessary and must be available for all patients 4,5,6 May be useful but not a necessity 2,3 Not routinely necessary for assessment 1 No relevance to difficult asthma assessment
ARTICLE IN PRESS Defining and investigating difficult asthma: Developing quality indicators similar experts.23 All analyses are based on second round ratings.
Results Seven out of 40 indicators were rated as necessary for defining difficult asthma. Difficult asthma was defined as ‘symptoms [suggestive of asthma] despite BTS Step 4 therapy’ (Table 1). Other necessary criteria included unscheduled attendances at general practices and at Accident and Emergency departments, three indicators relating to hospital admissions and use of frequent/continual oral corticosteroids. Thirty indicators were identified as being equivocal or unnecessary for defining difficult asthma (Tables 2 and 3). There was disagreement over two indicators in the definition of difficult asthma. These were ‘Prebronchodilator FEV1 as percentage of predicted’ Table 1 Indicators necessary for the definition of difficult asthma. Median panel score Pathognomic 9 Useful criteria 8 7
Symptoms despite Step 4 of BTS Guidelines None Unscheduled GP attendances Accident and Emergency department attendances Hospital admissions greater than 24 h Number of ICU admissions Number of ICU admissions with raised inflation pressures Use of frequent/continual oral corticosteroids
Table 2 Indicators of uncertain relevance in the definition of difficult asthma. Median panel score Equivocal/ unproven 4,5,6
Peak flow patterns or variability Disability/QoL score Presence of co-morbid psychiatric history
1257
and ‘Fixed airflow obstruction (post-bronchodilator/ post-steroid trial, FEV1/FVCo70%)’. Twenty-three investigations were rated as necessary to assess all patients with difficult asthma (overall panel rating of 47) (Table 4). Of these, non-physiological indicators were the only two with an overall panel rating of 9 (assessment of environment and psychological well-being) with an assessment of compliance with medication rated with an overall panel rating of 8. Other investigations rated with an overall panel rating of 8 included peak flow patterns/variability, flow/volume loop, plain chest radiograph, and assessment of vocal cord dysfunction and dysfunctional breathing/hyperventilation. Fifteen other investigations were rated as being necessarily available for all patients diagnosed as having difficult asthma. These included spirometry with reversibility, total lung capacity, smoking status and bone densiometry (Table 4). In addition, 14 investigations were rated as being equivocal, which should be available, but are not necessary for all patients, including bronchoscopy and exercise induced bronchospasm (Table 5). Six investigations were rated as being not routinely required for the assessment of difficult asthma including inhaled allergen challenge, thyroid function and progressive exercise test (for cardiac disease) (Table 6). There was disagreement over one indicator relating to ‘induced sputum differential white cell count’ in the investigation of difficult asthma. A full list of ratings for all 83 indicators is available.
Discussion Defining difficult asthma This paper has characterised difficult asthma as symptoms despite an arbitrary level of therapy (BTS Step 49) combined with continued unpredictable or emergency healthcare resource usage. The level of therapy differs from the ERS definition,20 which had a treatment threshold of high dose inhaled corticosteroids (did not include long acting b2 agonists or other add-on therapies). This paper has, for the first time, described the development of a set of quality indicators for the investigation of patients with difficult asthma which can now be assessed in clinical practice. Whilst this consensus definition of difficult asthma incorporates most true severe asthma patients, there are some patients with physiologically severe asthma, who do not report symptoms,
ARTICLE IN PRESS 1258 Table 3
C.O. Prys-Picard et al. Indicators that are not necessary in the definition of difficult asthma.
Median panel score Not necessary 2,3
No relevance to difficult asthma 1
Pre-bronchdilator FEV1 as percentage of predicted Fixed airflow obstruction (post-bronchodilator/post-steroid trial, FEV1/ FVCo70%) Bronchial hyper-reactivity (PC20) Presence of significant co-morbid chest disease Blood eosinophilia Sputum differential white cell count Response to inhaled allergen challenge Skin-prick testing Total IgE estimation Allergen-specific RAST Arterial PO2 estimation Arterial PCO2 estimation (for hypercapnia) Arterial pH/PCO2 estimation (for hyperventilation) Bacterial culture/isolation Bronchoscopic appearances Airway smooth muscle hypertrophy/hyperplasia on bronchial wall biopsy Fibroblast accumulation on bronchial wall biopsy Plain chest radiograph HRCT evidence of bronchiectasis HRCT bronchial wall thickening HRCT ground glass shadowing 3 He contrast MRI ventilation defects Assessment of corticosteroid absorption/efficacy Use of leukotriene antagonists Use of regular nebulisers Continuous parenteral terbutaline infusions Continuous intravenous aminophylline infusions Anti-IgE immunoglobulin therapy Steroid sparing agents, e.g. methotrexate, azathioprine, gold, cyclophosphamide, mycophenylate Fungal growth/identification on sputum culture
inadequately self-manage or are under perceivers of asthma severity.24,25 It is believed that these patients have ongoing airway inflammation and reactivity and are susceptible to high risk asthma attacks while not reporting or experiencing symptoms. In reality these patients do have severe asthma, and are at risk of emergency healthcare resource usage, but do not complain of an ‘appropriate’ level of symptomatology, and as such, lie outside the consensus definition derived here. They do however remain within the broader concept of difficult asthma. The identification and management of this group of patients remains of major importance with specific reference to the prevention of high-risk asthma events. In patients with difficult asthma, identifying any co-morbid condition that exists with asthma is central to symptom reduction. Unfortunately, in this group of patients, history, physical signs and
physiological parameters are often unhelpful and sometimes misleading. By maintaining a high degree of clinical suspicion and by following a standardised and methodical investigational process, it is hoped that therapy will match true asthma severity, co-morbid conditions will be treated appropriately, iatrogenic morbidity reduced and healthcare usage rationalised.
Severe versus difficult asthma All patients with severe persistent asthma as defined by GINA will also fall under the category of difficult asthma. In that respect, the definitions are similar. The important differentiator is that in difficult asthma, the symptom level rating may be artificially high due to the presence of another unrecognised condition, and the treatment level inflated due to lack of efficacy.
ARTICLE IN PRESS Defining and investigating difficult asthma: Developing quality indicators Table 4
1259
Investigations that are necessary for patients with difficult asthma.
Median panel score Absolute necessity in all patients 9 Must be available for assessment tool 8
7
Table 5
Assessment of environment/occupation Assessment of psychological well-being Peak flow patterns/variability Flow/volume loop Assessment of compliance with medication Plain chest radiograph Assessment for vocal cord dysfunction Assessment of dysfunctional breathing/hyperventilation Spirometry with reversibility Bronchial hyper-reactivity (PC20) Skin prick testing Blood eosinophilia Total IgE estimation Allergen specific RAST HRCT chest Arterial Blood Gas estimation in acute presentation for documentation of type II respiratory failure Assessment of smoking status Assessment of medication side-effects Assessment of Body Mass Index Diffusion capacity (KCO) Total lung capacity (TLC) Assessment of corticosteroids efficacy Bone densiometry
Investigations that should be available, but are not necessary, for all patients with difficult asthma.
Median panel score May be useful but not a necessity 4,5,6
Oesophageal pH monitoring for GORD Arterial blood gas estimation for PO2, pH or PCO2 Inductance plethysmography Food exclusion diet Induced sputum differential white cell count Exhaled nitric oxide (eNO) CT sinuses Bronchoscopy Endo-bronchial biopsy Lavage for differential white cell count ENT assessment Trial IM Triamcinolone in non-compliant/non-absorbant patients Assessment of OSA (Epworth+overnight oximetry) Exercise induced bronchospasm
Investigation of difficult asthma This study has developed a set of investigations deemed necessary for the initial investigation of difficult asthma. Investigation is aimed at confirming or excluding asthma and identifying other comorbid conditions. The incidence of the latter is high in difficult asthma.
The findings from an integrated approach to investigating difficult asthma patients5,13 has confirmed the high incidence of additional diagnoses. This study has provided a framework for the detection of co-morbid pathology in these patients. Measures developed using consensus techniques have high face validity. Subsequent developmental
ARTICLE IN PRESS 1260
Table 6 Investigations that are not routinely necessary for the investigation of patients with difficult asthma. Median panel score Not routinely required for assessment 2,3 Inhaled allergen challenge Oral/nasal aspirin challenge Thyroid function Assessment of corticosteroid absorption Exhaled breath condensate analysis Progressive exercise test No relevance to difficult asthma assessment 1 None
work is required to provide empirical evidence, as far as is possible, to test for acceptability, feasibility, reliability, sensitivity to change and validity of the measures.17 The challenge when applying quality indicators is that they tend to focus on care for populations or average patients but they need also to reflect the context and circumstances of individual patients. Therefore, while the investigations rated as 47 should be conducted routinely for all patients, in interpreting these findings, we propose that, as a minimum requirement, a specialist difficult asthma clinic would ideally have access to all the investigations rated 4 or higher as listed in Tables 4 and 5. These investigations provide a template for good practice and for responding to individual patient contexts. However, factors such as local availability will cause disparity between these ‘ideals’ and reality. The definition of difficult asthma presented here encompasses a larger proportion of asthma patients than is currently seen in difficult or severe asthma clinics. There may be a discrepancy between clinical need and service provision at present. Enabling greater patient access to specialist services may require either expansion of the number of centres or increasing the capacity of the present ones. Since this group of patients use a disproportionately high proportion of the asthma healthcare budget11,12 it is reasonable to expect that this will be a cost-effective measure.
C.O. Prys-Picard et al. Limitations ‘Difficult’ is a subjective term and its definition with respect to asthma may vary between different user groups. The panel was composed exclusively of asthma specialists in the UK with responsibility for running specialist asthma clinics. There was also no patient involvement. The combined view on the definition of difficult asthma is likely to be valid elsewhere, but differences in health care systems and daily practice between countries, partly caused by national regulations, mean that investigation protocols may need to be customised.26 The definition of difficult asthma should apply as equally to a paediatric population, but the relative appropriateness of investigations may vary and has not been addressed in this study. There are no currently adequate freely available methods for assessing compliance/concordance with medication or corticosteroid efficacy. However, this study rated the necessity of investigations required to provide quality care, not their current feasibility.
Conclusion This paper formalises the conceptual definitions of ‘difficult’ as opposed to ‘severe’ asthma and provides a list of clinical attributes required to define difficult asthma. This is useful as a guide to clinicians but also sets a standard to provide uniformity in clinical trials. The central tenet is that symptoms persist despite an arbitrarily defined ‘reasonable’ level of medication and they do so for a number of disparate reasons. These persisting symptoms may represent uncontrolled or unresponsive severe asthma, or may be due to the presence of other conditions. Appropriate therapy of the latter may result in improvement or resolution of the patients’ symptoms. The function of a difficult asthma clinic is to bring together the appropriate skills and facilities in a multi-disciplinary team to evaluate a patient holistically, to confirm a diagnosis of asthma and its severity and identify other conditions and contributory factors which otherwise make it ‘difficult’ to treat.
Reference 1. Dolan CM, Fraher KE, Bleecker ER, et al. Design and baseline characteristics of the epidemiology and natural history of asthma: outcomes and treatment regimens (TENOR) study: a large cohort of patients with severe or difficult-to-treat asthma. Ann Allergy Asthma Immunol 2004;92(1):32–9.
ARTICLE IN PRESS Defining and investigating difficult asthma: Developing quality indicators 2. American Thoracic Society. Proceedings of the ATS workshop on refractory asthma: current understanding, recommendations, and unanswered questions. Am J Respir Crit Care Med 2000;162(6):2341–51. 3. Leung DY, Szefler SJ. New insights into steroid resistant asthma. Pediatr Allergy Immunol 1998;9(1):3–12. 4. Wamboldt FS, Spahn JD, Klinnert MD, et al. Clinical outcomes of steroid-insensitive asthma. Ann Allergy Asthma Immunol 1999;83(1):55–60. 5. Heaney LG, Conway E, Kelly C, et al. Predictors of therapy resistant asthma: outcome of a systematic evaluation protocol. Thorax 2003;58(7):561–6. 6. Van Ganse E, Laforest L, Pietri G, et al. Persistent asthma: disease control, resource utilisation and direct costs. Eur Respir J 2002;20(2):260–7. 7. Barnes PJ, Woolcock AJ. Difficult asthma. Eur Respir J 1998;12(5):1209–18. 8. GINA. Gina workshop reports, global strategy for asthma management and prevention. Publication No. 02-3659, 2002. 9. British guideline on the management of asthma. Thorax 2003;58(Suppl 1):1–94. 10. National Asthma Education and Prevention Program. Expert panel report: guidelines for the diagnosis and management of asthma update on selected topics-2002. J Allergy Clin Immunol 2002;110(Suppl 5):S141–219. 11. Smith DH, Malone DC, Lawson KA, Okamoto LJ, Battista C, Saunders WB. A national estimate of the economic costs of asthma. Am J Respir Crit Care Med 1997;156(3 Pt 1): 787–93. 12. Malone DC, Lawson KA, Smith DH. Asthma: an analysis of high-cost patients. Pharm Pract Manage Q 2000;20(1): 12–20. 13. Robinson DS, Campbell DA, Durham SR, Pfeffer J, Barnes PJ, Chung KF. Systematic assessment of difficult-to-treat asthma. Eur Respir J 2003;22(3):478–83. 14. Jones J, Hunter D. Consensus methods for medical and health services research. BMJ 1995;311(7001):376–80. 15. Naylor CD. What is appropriate care? N Engl J Med 1998; 338(26):1918–20.
1261
16. Black N, Murphy M, Lamping D, et al. Consensus development methods: a review of best practice in creating clinical guidelines. J Health Serv Res Policy 1999;4(4):236–48. 17. Campbell SM, Braspenning J, Hutchinson A, Marshall MN. Research methods used in developing and applying quality indicators in primary care. BMJ 2003;326(7393):816–9. 18. Brook RH, Chassin MR, Fink A, Solomon DH, Kosecoff J, Park RE. A method for the detailed assessment of the appropriateness of medical technologies. Int J Technol Assess Health Care 1986;2(1):53–63. 19. Shekelle PG, Kahan JP, Bernstein SJ, Leape LL, Kamberg CJ, Park RE. The reproducibility of a method to identify the overuse and underuse of medical procedures. N Engl J Med 1998;338(26):1888–95. 20. Chung KF, Godard P, Adelroth E, et al. Difficult/therapyresistant asthma: the need for an integrated approach to define clinical phenotypes, evaluate risk factors, understand pathophysiology and find novel therapies. ERS Task Force on Difficult/Therapy-Resistant Asthma. European Respiratory Society. Eur Respir J 1999;13(5):1198–208. 21. Agency for Health Services Research. Using clinical practice guidelines to evaluate quality of care. Rockville: AHCPR; 1995. 22. Baker R. Principles of quality improvement. Part Two: Measuring quality. J Clin Governance 2001;9:153–6. 23. Shekelle PG, Kahan JP, Park RE, Bernstein SJ, Leape LL, Kamberg BA, et al. Assessing appropriateness by expert panels: how reliable? J Gen Intern Med 1996;10:81. 24. Chetta A, Gerra G, Foresi A, Zaimovic A, Del Donno M, Chittolini B, et al. Personality profiles and breathlessness perception in outpatients with different gradings of asthma. Am J Respir Crit Care Med 1998;157(1):116–22. 25. Veen JC, Smits HH, Ravensberg AJ, Hiemstra PS, Sterk PJ, Bel EH. Impaired perception of dyspnea in patients with severe asthma. Relation to sputum eosinophils. Am J Respir Crit Care Med 1998;158(4):1134–41. 26. Marshall MN, Shekelle PG, McGlynn EA, Campbell S, Brook RH, Roland MO. Can health care quality indicators be transferred between countries? Qual Saf Health Care 2003; 12(1):8–12.
