Burnout intervention studies for inpatient elderly care nursing staff ...

International Journal of Nursing Studies 51 (2014) 63–71

Contents lists available at SciVerse ScienceDirect

International Journal of Nursing Studies journal homepage: www.elsevier.com/ijns

Review

Burnout intervention studies for inpatient elderly care nursing staff: Systematic literature review Claudia Westermann a,*, Agnessa Kozak a, Melanie Harling a, Albert Nienhaus a,b a b

University Clinic Hamburg-Eppendorf, Institute for Health Service Research in Dermatology and Nursing, Germany Institution for Statutory Accident Insurance and Prevention in Healthcare and Welfare Services, Germany

A R T I C L E I N F O

A B S T R A C T

Article history: Received 11 July 2012 Received in revised form 23 October 2012 Accepted 1 December 2012

Background: Staff providing inpatient elderly and geriatric long-term care are exposed to a large number of factors that can lead to the development of burnout syndrome. Burnout is associated with an increased risk of absence from work, low work satisfaction, and an increased intention to leave. Due to the fact that the number of geriatric nursing staff is already insufficient, research on interventions aimed at reducing work-related stress in inpatient elderly care is needed. Objective: The aim of this systematic review was to identify and analyse burnout intervention studies among nursing staff in the inpatient elderly and geriatric long-term care sector. Methods: A systematic search of burnout intervention studies was conducted in the databases Embase, Medline and PsycNet published from 2000 to January 2012. Results: We identified 16 intervention studies. Interventions were grouped into workdirected (n = 2), person-directed (n = 9) and combined approaches (work- and persondirected, n = 5). Seven out of 16 studies observed a reduction in staff burnout. Among them are two studies with a work-directed, two with a person-directed and three with a combined approach. Person-directed interventions reduced burnout in the short term (up to 1 month), while work-directed interventions and those with a combined approach were able to reduce burnout over a longer term (from 1 month to more than 1 year). In addition to staff burnout, three studies observed positive effects relating to the client outcomes. Only three out of ten Randomised Control Trials (RCT) found that interventions had a positive effect on staff burnout. Conclusion: Work-directed and combined interventions are able to achieve beneficial longer-term effects on staff burnout. Person-directed interventions achieve short-term results in reducing staff burnout. However, the evidence is limited. ß 2012 Elsevier Ltd. All rights reserved.

Keywords: Burnout Elderly care staff Intervention studies Systematic literature review Prevention Turnover Workload

What is already known about the topic?

* Corresponding author at: University Clinic Hamburg-Eppendorf, Institute for Health Service Research in Dermatology and Nursing, Martinistrasse 52, 20246 Hamburg, Germany. Tel.: +49 40 7410 59516; fax: +49 40 7410 59708. E-mail addresses: [email protected] (C. Westermann), [email protected] (A. Kozak), [email protected] (M. Harling), [email protected] (A. Nienhaus). 0020-7489/$ – see front matter ß 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijnurstu.2012.12.001

Workplace stress is a significant problem for employees working in healthcare, particularly among professions such as nursing. Burnout in the nursing profession is associated with the risk of absenteeism and high staff turnover. A substantial economic impact can be inferred from this syndrome.

64

C. Westermann et al. / International Journal of Nursing Studies 51 (2014) 63–71

What this paper adds Few reviews have been published on the effectiveness of interventions in preventing burnout in healthcare workers. This paper identifies intervention studies in the workplace aimed at preventing burnout for staff in the inpatient elderly care segment. Limited evidence is available for the effectiveness of work-, person-directed and combined interventions in reducing staff burnout. Work-directed and combined intervention programmes potentially have longer-lasting beneficial effects on staff burnout than person-directed intervention programmes. 1. Introduction Within an occupational context, the term burnout is used to describe an individual reaction to chronic stress at work. These reactions are characterised by emotional exhaustion, a distanced, indifferent, cynical attitude towards work and other people (depersonalisation) and a decline in personal accomplishment (Maslach et al., 2001). Burnout is a complex syndrome featuring a large number of unspecific symptoms that can have far-reaching impacts on an individual’s health (Von Ka¨nel, 2008). Burnout is also associated with an increased risk of absence from work, more work days lost due to illness, inability to work and low work satisfaction, and the individuals affected are very likely to give up work completely (Borritz et al., 2006; Camerino et al., 2006; Edwards et al., 2003). Staff who provide inpatient elderly and geriatric long-term care are exposed to a large number of factors that can lead to the development of burnout syndrome (Gandoy-Crego et al., 2009; Zimber, 1998). They are faced with time pressure, heavy workloads, few opportunities to influence their work, confrontation with suffering, dealing with people in need of long-term care, physical stress, shift work and staff shortages (Gandoy-Crego et al., 2009; Glaser et al., 2008; Jenull et al., 2008; McHugh et al., 2011; Zimber, 1998). Other enabling factors include a lack of support and recognition among colleagues and from management (Lewis et al., 2010). Factors not relating to work that may lead to the development of burnout include an individual’s personality and social relationships (GandoyCrego et al., 2009; Narumoto et al., 2008). One particular challenge faced by professional nursing staff that may cause burnout is providing care to people with dementia (Haberstroh et al., 2010; Mackenzie and Peragine, 2003; Rodney, 2000). Due to cognitive impairment and psychological changes among clients, as well as the related loss of ability to cope with day-to-day life, this sort of work is intensive and emotionally demanding (Fuchs-Lacelle et al., 2008; Opie et al., 2002; Van Weert et al., 2005; Weyerer et al., 2004). As a consequence of demographic changes in western societies, the proportion of people with complex diseases (e.g., dementia) who are in need of intensive longterm care is continually on the rise (Prezewowsky, 2007). In the face of this trend the number of geriatric nursing staff is already insufficient (Camerino et al., 2006; Mackenzie et al., 2006). Measures must be taken to boost

motivation and improve the health of the staff members (Borritz et al., 2006; Evers et al., 2002). Therefore, research on interventions aimed at reducing work-related stress among professional nursing staff is needed. The aim of this paper is to investigate the current research on burnout intervention studies in the field of inpatient elderly and geriatric long-term care. The review focuses on the following questions: Which intervention approaches for burnout in the field of inpatient elderly and geriatric long-term care have been published since 2000? Are there studies proving that burnout can be influenced by intervention? Which recommendations for future interventions can be derived from these studies? 2. Methods 2.1. Search strategy and selection criteria Systematic literature research was performed using the Medline, Embase and PsycNet databases. This literature search covered the period from January 2000 to January 2012 and the languages were limited to German, English, French, Spanish, Portuguese and Italian. The search terms used were ‘‘burnout’’ combined with ‘‘occupation* AND intervention’’, ‘‘employ* AND intervention’’, ‘‘intervention AND nursing AND geriatric’’, ‘‘intervention AND nursing AND homes’’. Relevant publications were included regardless of the study design. Included were primary studies, which analysed intervention programmes aimed at preventing burnout for nursing staff in the setting of inpatient elderly and geriatric long-term care. We only included studies which methodically assess the issue of burnout among nursing staff. Studies which examined the effectiveness of burnout intervention in diverse healthcare occupations, but did not provide detailed information on the impatient elderly care staff, where excluded. Further literature sources were taken from the reference list of articles included and other reviews. Studies that did not meet the inclusion criteria, as well as duplicates were excluded and the reasons for their exclusion noted. Wherever it was not possible to make a decision on a study’s inclusion or exclusion based on the abstract, the full text of the study was examined. Relevant publications were assessed by two reviewers working independently. A consensus method was used to resolve disagreements. 2.2. Presentation of intervention approaches and their effects Marine et al. (2006) make a distinction between two types of intervention: work-directed (circumstanceoriented) and person-directed (behavioural). This paper also refers to a third study approach: combined interventions (work- and person-directed). According to Marine et al. (2006) work-directed interventions are carried out with the aim of changing the working environment, work tasks or working methods. Such interventions involve, for example, decreasing workload or increasing job control. Person-directed interventions are aimed at teaching staff skills or techniques to lessen the effects of stressors for the


individual. Such interventions involve, for example, reducing the effects of stress through relaxation or cognitivebehavioural techniques that increase coping skills. Combined interventions are aimed at teaching personal skills and also include measures to change the work environment or improve working methods. Intervention effects that are relevant to burnout were identified. An effect is deemed to exist if the intervention used has a statistically significant positive impact on one of the burnout sub-scales (p < .05). Furthermore, effects on burnout were classified by reference period as being short-term (up to 1 month after the intervention), medium-term (between 1 month and 1 year after the intervention) and long-term (more than 1 year after the intervention) in line with Marine et al. (2006). If reported in primary studies, mean values (M) and standard deviation (SD) as well as p-values were stated in the result part. 2.3. Quality and criteria for the evidence classification of included studies The main component used in this paper to assess the level of evidence is the study design. The study design plays a decisive role in determining the validity of the effects observed. The best possible level of evidence can be achieved with RCTs. These studies reduce the likelihood of systematic design errors (Des Jarlais et al., 2004). Often, however, the conditions in the healthcare sector and a variety of associated issues call for different study designs (Kunz et al., 2009). Randomisation cannot always be justified from an ethical perspective or is often impossible to implement due to organisational factors. However, non-randomised designs can also contribute key findings to the body of evidence (Gordis, 2000; Des Jarlais et al., 2004). The evidence found in the studies included is assigned to quality categories in line with the system developed by Kunz et al. (2009). According to this system, RCTs are assigned to the highest evidence level I. Experimental studies without randomisation and observational studies with control groups (cohort studies, case control studies) are assigned to level II, while observational studies without control groups (crosssectional, before and after studies, case series) are assigned to level III. Case reports, pathophysiological studies or laboratory studies, expert opinions or the expert consensus are assigned to the lowest evidence level – level IV. Limitations affecting the design and the manner in which the intervention measures are performed can potentially influence the quality of the results (Kunz et al., 2008). 3. Results 3.1. Search results The search query resulted in a total of 478 studies, including six reviews of intervention programmes aimed at preventing stress and burnout at work (Awa et al., 2010; Edwards et al., 2003; Marine et al., 2006; Mimura and

65

Griffiths, 2003; Van Wyk and Pillay-Van Wyk, 2010; Van der Klink et al., 2001). Three further relevant studies were found manually. A total of 16 original studies fulfilled the inclusion criteria and made up the basis of this review (Fig. 1). Out of the 16 studies included, three studies originated from Australia, Germany and Canada in each case, two each from the US and the Netherlands, and one study each from Denmark, Italy and the UK. Ten studies have a randomised controlled design, four have a quasi-experimental design, one has a pre-post design and one a longitudinal comparative design. 3.2. Study population and sample size In most of the studies (n = 15) the interventions took place in psychogeriatric long-term care, with the exception of the study by Jensen et al. (2006). Fourteen studies (with exception to Mackenzie et al., 2006) looked specifically at care for clients with dementia. The sample examined consisted of professional nurses working in inpatient elderly and geriatric long-term care. Three studies also examined a small percentage of activity staff members, consenting managers and social workers (Fritsch et al., 2009; Jeon et al., 2012; Richardson et al., 2002). The number of participants varied between 21 and 300. 3.3. Survey instruments used to measure burnout The most common instrument used to measure burnout was the MBI (n = 14). Three of the studies used other/additional survey instruments: the stress screening system for human service providers (BHD system) developed by Hacker et al. (1995) and the General Health Questionnaire developed by Goldberg and Hillier (GHQ, 1979). Because of the different MBI and scales versions used in the studies, only a limited comparison of the effects was possible (Table 1). 3.4. Approaches and study period of interventions Due to the different intervention concepts used in the studies, we grouped them according to their intervention approach. As shown in Table 1, the person-directed approach was the most common type of intervention (n = 9) in this review. Five studies used a combined while two used a work-directed approach. In seven out of the 16 studies, a significant healthpromoting effect was identified. Out of these seven studies, which achieved a statistically significant effect on staff burnout, two applied a person-directed (Haberstroh et al., 2010; Mackenzie et al., 2006), two a work-directed (Baldelli et al., 2004; Fuchs-Lacelle et al., 2008) and three a combined (Jeon et al., 2012; Schrijnemaekers et al., 2003; Van Weert et al., 2005) intervention approach. The duration of the interventions varied from 4 weeks to 18 months (Table 1). Two studies were conducted over a long period of time (up to 18 months of follow-up). Ten studies stretched the study period between 2 months and 1 year. The remaining four studies only observed short time periods (up to 1 month after intervention).

66


Fig. 1. Flow chart.

As shown in Table 2, the work-directed and combined interventions achieved medium to long-term effects on staff burnout (five out of seven), while person-directed interventions tended to achieve short-term effects (two out of nine). 3.4.1. Summary of study results with a work-directed intervention approach Both studies (medium-term) with work-directed interventions had health-promoting effects on staff burnout. In the intervention study performed by Fuchs-Lacelle et al. (2008) a standardised systematic pain assessment was implemented. In comparison to the control group, the authors observed a significant reduction in the average values in the MBI sub-scale emotional exhaustion (p < .03). In addition, the intervention participants reported a general decline in work-related stress (nursing stress scale). The 1-year ongoing exercise and activity programme for clients suffering from dementia that was assessed by Baldelli et al. (2004) showed statistically significant improvements on staff burnout in all MBI sub-scales between the follow-up periods: emotional exhaustion (lack of gratification) t0 M = 13.3, SD = 7.2, t1 M = 9.2 SD = 5.8, p < .05; depersonalisation t0 M = 5.5 SD = 4.2, t1 M = 2.2 SD = 3.3, p < .05; personal accomplishment (personal satisfaction) t0 M = 36.1 SD = 8.5, t1 M = 40 SD = 8.4, p < .05.

3.4.2. Summary of study results with a person-directed intervention approach Nine studies adopted a person-directed intervention approach. Effects on staff burnout were only observed in two studies with a brief follow-up period. Compared with the control groups, statistically significant effects were observed in combined communication training in dealing with individuals suffering from dementia and with colleagues (Haberstroh et al., 2009). These became apparent in the stress sub-scales (BHD system) emotional exhaustion (p < .05), intrinsic motivation (p < .05) and work (dis)satisfaction (p < .05). Positive effects could also be observed in respect of the employees’ social skills. Social skills in particular showed a significant negative correlation with psychological stress (r = .33; p < .01). In the brief mindfulness-based stress reduction (MBSR) intervention a significant Group Time interaction was observed by Mackenzie et al. (2006). At baseline MBSR participants were significantly more exhausted than control participants. Following the intervention, a significant reduction in the mean values in the MBI subscale emotional exhaustion were observed among MBSR participants, whereas the scores of control participants increased (MBSR participants: t0 M = 26.4, SD = 10.4; t1 M = 20.7, SD = 10.4; controls: t0 M = 16.2, SD = 8.8; t1 M = 17.2, SD = 10.6; Group Time interaction p < .05). With respect to depersonalisation, MBSR participants showed relative constant mean values over the testing


67

Table 1 Details of studies included. Reference/Country

Design/LE

Work-directed interventions Fuchs-Lacelle QE/II et al. (2008), CA PP/III Baldelli et al. (2004), IT

Person-directed interventions Kuske et al. (2009), DE RCT/I

Sample size

Instrument burnout

Intervention

Duration in months

Outcome burnout

61

MBI 1981, 1997, 22 items MBI 1986, 1993, 22 Items

Systematic Pain Assessment (PACSLAC) Therapeutic programme of occupational therapy and cognitive rehabilitation for demented patients

3

EE#*

MBI-D 2003, 21 items BHD 1999

Training programme in dementia care Communicating with dementia patients (Training 1) and with colleagues (Training 2) Training programme in managing behavioural symptoms of dementia and peer support Training programme in managing dementiarelated challenging behaviours with a peer support group Mindfulness-based stressreduction programme Ergonomic and psychosocial training Training in knowledge and self-efficacy in dementia care

21

134

Haberstroh et al. (2009), DE

QE/II

79

Visser et al. (2008), AU

RCT/I

52

MBI 1996, 22 items

Davison et al. (2007), AU

RCT/I

132

MBI 1996, 22 items

Mackenzie et al. (2006), CA Jensen et al. (2006), DK

RCT/I

30

RCT/I

210

MBI 1996, 22 items MBI 1986, 22 items MBI 1996, 22 items

Mackenzie and Peragine (2003), CA

QE/II

47

Richardson et al. (2002), UK

RCT/I

86

MBI 1981, 22 items

Zimber et al. (2001), DE

QEII

202

BHD 1995, GHQ 1979

Combined interventions Jeon et al. (2012), AU

RCT/I

194

MBI-HSS 1996, 22 items GHQ-12 1997

Fritsch et al. (2009), US

RCT/I

192

MBI 1996, 14 items

Robison et al. (2007), US

RCT/I

384

MBI 1982 (only DP), 11 items

Van Weert et al. (2005), NL

QE/II

129

Schrijnemaekers et al. (2003), NL

RCT/I

300

MBI-NL 1993, 1994,1995,2000, 20 items MBI-NL 1994, 1993, 20 items

Post-test

a

1 year

* b

* b

* b

12

EE#

3

$

1 year

1

EE#* a, intrinsic motivation"* a, job(dis)satisfaction#* a

1 month

2

$

1 year

n/a

$

1 year

1

EE#* a,b, DP#*

6

$

1

, DP#

, PA"

a

1 year

1 month >1 year

* a

1 year

n/a

PA" directly after intervention, no effect after three months $

3

$

1 year

Training in person-centred care (Training 1) and in dementia care mapping (Training 2) Time slips (TS), a creative expression programme in dementia care Cooperative communication programme for staff and families on dementia units Snoezelen

4

EE#*

2.5

$

1 month

n/a

$

1 year

4.5

EE#*

a

>1 year

Emotion-oriented care for cognitively impaired elderly persons and supervision meetings

8

PA"*

a

1 year

Educational course to increase staff skills in dealing with abuse of the elderly Communication training in dealing with dementia patients, Training in coping with job stress and cooperating with colleagues

a,b

1 month

1 year

Note. LE: level of evidence; EE: emotional exhaustion; DP: depersonalisation; PA: personal accomplishment; RCT: randomised controlled trial; QE: quasiexperimental design; PP: pre post design; n/a: not available. #: decrease; ": improvement; $: no improvement/no decrease. a Comparison between groups. b Pre-post comparison. * p < .05 significant change.

68


Table 2 Number of post-tests and number of observed effects on staff burnout by type of intervention and time interval. Intervention approach

Short term (1 month) n*

n Person-directed Work-directed Combined All

Medium term (1 year)

8 1 2 11

n

Total

n*

n

n*

5 2 3

0 2 2

1 0 1

0 0 1

9 2 5

10

4

2

1

16

2(1a) 1 1 5

Long term (>1 year)

n

n* 2 (3) 2 3 7(8)

Note. n: number of studies; n*: number of studies with effect. a One study with short-term effects only directly after intervention.

periods, whereas control participants mean scores increased significantly (MBSR participants: t0 M = 4.75, SD = 4.5; t1 M = 4.8, SD = 4.4; controls: t0 M = 3.4, SD = 5.1; t1 M = 5.0, SD = 5.9; Group Time interaction p < .05). A significant improvement in the staff-relevant outcomes of life satisfaction (p < .05) and relaxation (p < .05) were observed as well. The self-efficacy training for dealing with patients with dementia (Mackenzie and Peragine, 2003) had a shortterm, positive effect on health only in the MBI sub-scale personal accomplishment directly after intervention (p < .05). However, this result could not be confirmed at follow up after 3 months (p = .61). In a further six person-directed interventions (Table 1), no effects on staff burnout could be observed. Staff training on how to deal with individuals with dementia (Davison et al., 2007; Kuske et al., 2009; Visser et al., 2008; Zimber et al., 2001), with stress (Zimber et al., 2001), with colleagues (Visser et al., 2008; Zimber et al., 2001) and with attacks on clients (Richardson et al., 2002) as well as ergonomic and psycho-social training (Jensen et al., 2006) evidenced no impact on staff burnout (p > .05). The accompanying peer support system analysed by Davison et al. (2007) and Visser et al. (2008) did not show any positive impact either (p > .05). 3.4.3. Summary of study results with a combined intervention approach Five studies used a combined intervention approach. Three out of five interventions showed a significant positive effect on staff burnout. Of these, two studies were designed for the medium-term and one for a long term period. Jeon et al. (2012) observed significant positive effects of Dementia Care Mapping (DCM) on the staff MBI sub-scale emotional exhaustion (p < .05). In addition, the authors also observed a link between low management support and staff burnout symptoms that are perceived to be more severe. The Snoezelen intervention analysed by Van Weert et al. (2005) had a long-term effect on the staff MBI subscale emotional exhaustion in comparison with the control group (intervention group: t0 M = 10.5, SD = 0.8; t1 M = 8.3, SD = 0.9; control group: t0 M = 10.5, SD = 0.8; t1 M = 10.8, SD = 0.9). The authors also observed significant positive effects on the training group in the scale of stress reaction and job satisfaction (Maastricht Work Satisfaction Scale for Healthcare – MAS-GZ). Compared with the staff in the

control group, the certified nursing assistants (CNAs) in Snoezelen facilities were more satisfied with the quality of care (p < .05) and with the interaction with patients with dementia (p < .05). Schrijnemaekers et al. (2003) observed a significant health-promoting effect on the MBI sub-scale personal accomplishment following the implementation of emotion-oriented 24-h care. Further positive effects could be observed on the levels of job satisfaction (p < .05). Both, the creative expression programme ‘‘Time Slips’’ (Fritsch et al., 2009) and the communication and conflict resolution programme for employees and relatives of patients in special care (Robison et al., 2007) showed no positive effects on staff burnout. However, Robison et al. (2007) found a treatment effect over time only for registered nurses (p < .05) but not for CNAs (p = .09). 3.5. Other results: significant impact on client related outcomes In addition to staff burnout, eight out of 16 studies reported about different client related outcomes. Of them, three studies observed significant improvements of the living and psycho-physical conditions (Haberstroh et al., 2009; Robison et al., 2007; Van Weert et al., 2005), while other three did not (Davison et al., 2010; Fritsch et al., 2009; Visser et al., 2008). Haberstroh et al. (2009) observed an improvement in the quality of life of patients with dementia – specifically in their communication skills and mobility (p < .05). Robison et al. (2007) analysed a communication and conflict resolution training programme for staff and patients’ relatives in the special care of clients with dementia. An improvement in client outcomes (aggression, self-abuse or sexual advances, inappropriate dress or disrobing, constant requests for attention or help, and wandering) could be observed 2 months after the intervention, although the clients themselves did not receive any intervention (p < .05). Van Weert et al. (2005) showed a significant effect on resident outcomes regarding depressive behaviour, loss of decorum, and restless behaviour (p < .05). The authors referred to a possible strong interaction between clientrelated outcomes and psychological outcomes (such as stress reactions, job satisfaction and burnout) among CNAs. In further two studies the authors mentioned improvements on client related general condition; however it was not the focal point of their analysis. The observed effect


was only mentioned in the discussion part (Baldelli et al., 2004; Jeon et al., 2012). 3.6. Quality of the included studies Ten out of 16 studies were classified as RCTs (62.5%) and assigned to evidence level I. Of these, three showed positive correlations between the interventions analysed and staff burnout (Jeon et al., 2012; Mackenzie et al., 2006; Schrijnemaekers et al., 2003). The results of Mackenzie et al. (2006), however, have to be viewed critically given the small sample size (n < 50). In three out of five studies assigned to evidence level II, positive effects could be seen on staff burnout (Haberstroh et al., 2009; Fuchs-Lacelle et al., 2008; Van Weert et al., 2005). One study was assigned to evidence level III (Baldelli et al., 2004). This study showed a positive effect on staff burnout. However, this result had to be viewed critically due to the limitations of the study design (no control group, small sample size). An overview is shown in Table 1. 4. Discussion This paper provides an overview of the intervention studies on burnout in inpatient elderly and geriatric longterm care from the period between 2000 and January 2012. One positive aspect is the relatively large number of highquality studies. The publications available, however, only allow for a limited comparison of the effects achieved. The studies differed in terms of the conceptual nature of the approach (design, intervention, number of participants, participant group) as well as with respect to the survey instruments employed. Although 14 out of the 16 studies used the MBI system, a comparison between the interventions observed was restricted by the heterogeneous use of the tool (version, scale items). The need to address increasing psychogeriatric requirements in elderly care was reflected in the settings chosen for the interventions. The setting chosen was primarily specialised dementia care. 4.1. Summary – effects of interventions on staff burnout All studies (n = 2) with a work-directed approach led to a reduction in staff burnout, lasting up to 1 year after intervention (Table 2). Two out of nine interventions with a person-directed approach had an effect on staff burnout for up to 1 month after the intervention. Three out of five combined interventions led to positive effects on staff burnout lasting from up to 4 months to more than 1 year. The majority of RCTs (seven out of ten) found no association between the interventions examined and staff burnout. Evidence for the effectiveness of work-, persondirected and combined interventions in terms of reducing staff burnout is thus limited. Possible causes are listed in Section 4.3. The two work-directed interventions had mediumterm effects on staff burnout. The interventions were, however, very different – pain assessment in the case of Fuchs-Lacelle et al. (2008) and exercise and activity programmes for clients with dementia in the case of

69

Baldelli et al. (2004). Direct comparisons were therefore not feasible. Out of the nine studies using person-directed intervention, only two had a positive impact on staff burnout, with any effects only recorded over the short term. The brief mindfulness-based stress reduction intervention for staff revealed a significant health-promoting effect on the burnout sub-scale emotional exhaustion (Mackenzie et al., 2006). This study was also mentioned in the review performed by Van Wyk and Pillay-Van Wyk (2010), but was assessed differently. The authors only compared mean values in the intervention group with those in the control group at the post-training. Consequently, the different starting values of the two groups were disregarded, meaning that the effect was underestimated. 4.1.1. Unconfirmed short-term effects on staff burnout In a study by Mackenzie and Peragine (2003) shortterm effects on staff burnout could be observed only directly after intervention, but not at follow up (after 3 months). In order to maintain the effects achieved, the authors recommended a continuous approach. As argued by Zimber et al. (2001), a lack of continuity in the training programmes will lead to limited effects. 4.2. Other effects on client-related outcomes In addition to staff burnout, three studies observed positive effects relating to the client outcomes (Haberstroh et al., 2009; Robison et al., 2007; Van Weert et al., 2005). Within this context, Van Weert et al. (2005) referred to a possible marked interaction between factors such as workload, client-related outcomes and psychological outcomes for employees. In line with this hypothesis, Haberstroh et al. (2009) confirmed, on the basis of the pre-post data, significant simultaneous correlations between the social skills of nursing staff and client quality of life. 4.3. Reasons for lack of effects on staff burnout In nine studies, the interventions showed no impact on staff burnout. The following possible reasons were referred to: the fact that the training units do not focus enough, from an organisational perspective, on the participants (Robison et al., 2007; Zimber et al., 2001), lack of objective measuring instruments aimed at ensuring the optimum control of intervention implementation (Jensen et al., 2006), different working conditions and work requirements in the settings analysed (Jensen et al., 2006), too few participants in the nursing facility in question (Davison et al., 2007; Mackenzie and Peragine, 2003; Visser et al., 2008), lack of evaluation at administrative level in the interests of sustainability (Robison et al., 2007; Zimber et al., 2001), lack of continuity (Davison et al., 2007; Kuske et al., 2009; Mackenzie and Peragine, 2003; Zimber et al., 2001), lack of time to devote to care and rigid working conditions (Kuske et al., 2009), inadequate staffing levels and high levels of work-related demands (Davison et al., 2007), insufficient resources for the facilities, e.g. lack of space, lack of aids and means of transportation (Jensen

70


et al., 2006) and ethical problems relating to client observation (Jensen et al., 2006). The accompanying peer support meetings were well received by employees but did not have any impact on staff burnout. Amongst other things, this was due to the fact that the programmes did not last long enough. Moreover, employees were not released from their duties in order to attend the meetings and there were not enough participants. Lack of time and administrative support were further reasons for neglecting peer meetings after the intervention (Davison et al., 2007; Visser et al., 2008). Several studies showed that possible barriers to change have to be identified in the facility itself. The most frequently mentioned factor was a lack of support at management level (Davison et al., 2007; Schrijnemaekers et al., 2003; Visser et al., 2008; Zimber et al., 2001).

accordance with our results, they found that persondirected interventions tended to have short-term effects, whereas interventions based on work-directed and combined approaches tended to have longer-term effects. 4.5. Limitations of this systematic literature review The results may be influenced by publication bias. Studies published in journals that were not listed in the databases used in our review could not be identified in this database search. Due to the heterogeneous methods used and the missing data in individual studies, this paper does not include a comparison of the effect estimates. This systematic literature review is purely descriptive in nature. 5. Conclusion

4.4. Comparison of results with other reviews Four reviews identified in this literature search examined the effectiveness of stress management intervention in healthcare occupations (Edwards et al., 2003; Mimura and Griffiths, 2003; Marine et al., 2006; Van Wyk and Pillay-Van Wyk, 2010). Two further reviews analysed the effectiveness of interventions to prevent stress and burnout across a number of different sectors (Van der Klink et al., 2001; Awa et al., 2010). In contrast to our results, Van der Klink et al. (2001) reported that work-directed interventions did not lead to success in staff coping with stress. The authors identified only one study with a combined approach that yielded a significant effect. Edwards et al. (2003) found that lack of time for client contact was associated with high stress levels among healthcare staff, particularly in long-term nursing care occupations. They also looked at how effective interventions (solely person-directed) are when it comes to dealing with stress. Stress management training combined with relaxation exercises and training measures, including psycho-social training units, helped to reduce burnout among long-term care staff. Mimura and Griffiths (2003) found more evidence in favour of person-directed than work-directed interventions. They were not able to recommend any particular approach for practical implementation, however, as the number of studies examined was too small. Limited evidence was available regarding the effectiveness of person- and work-directed interventions in reducing stress levels in healthcare settings by Marine et al. (2006). The authors proposed that the best period for proving stress-reducing effects is between 6 months and 2 years after the intervention. Van Wyk and Pillay-Van Wyk (2010) stated that short-term measures did not yield any results. The most convincing evidence was delivered by long-term measures. The authors were in favour of regular refresher courses (within 18 months) after the intervention in order to maintain the positive effects of the interventions on staff stress. Awa et al. (2010) analysed cross-sector intervention studies with different approaches (work- and persondirected as well as combined). In line with Van Wyk and Pillay-Van Wyk (2010), they confirmed that interventions with regular refresher courses had longer-term effects. In

Although only a few interventions were able to achieve a positive effect on staff burnout, we found that all three intervention approaches are potentially useful in order to achieve positive effects on staff burnout. Work-directed and combined intervention programmes seem to achieve potentially longer-lasting effects than person-directed intervention programmes. More evidence is needed to make a clear recommendation for an intervention approach to prevent staff burnout. A more comprehensive examination of the causal links between burnout and other factors, e.g. client outcomes and employees’ professional skills, is required. Similarly, there is a need for further longitudinal studies with representative samples and sufficiently long investigation periods. Long-term aspects should be taken into account in future projects. Contributions CW, AN, and MH were responsible for project management. CW and AK were responsible for data collection and analysis. CW was responsible for drafting the manuscript. AK, AN, and MH reviewed the paper for important intellectual content. Competing interests The authors declare that they have no competing interests. References Awa, W.L., Plaumann, M., Walter, U., 2010. Burnout prevention: a review of intervention programs. Patient Education and Counseling 78 (2), 184–190. Baldelli, M.V., Fabbo, A., Costopulos, C., Carbone, G., Gatti, R., Zucchl, P., 2004. Is it possible to reduce job burnout of the health care staff working with demented patients? Archives of Gerontology and Geriatrics (Suppl. 9), 51–56. Borritz, M., Rugulies, R., Bjorner, J.B., Villadsen, E., Mikkelsen, O.A., Kristensen, T.S., 2006. Burnout among employees in human service work: design and baseline findings of the PUMA study. Scandinavian Journal of Public Health 34 (1), 49–58. Camerino, D., Conway, P.M., Van der Heijden, B.I., Estryn-Behar, M., Consonni, D., Gould, D., Hasselhorn, H.M., 2006. Low-perceived work ability, ageing and intention to leave nursing: a comparison among 10 European countries. Journal of Advanced Nursing 56 (5), 542–552.

C. Westermann et al. / International Journal of Nursing Studies 51 (2014) 63–71 Davison, T.E., McCabe, M.P., Visser, S., Hudgson, C., Buchanan, G., George, K., 2007. Controlled trial of dementia training with a peer support group for aged care staff. International Journal of Geriatric Psychiatry 22 (9), 868–873. Des Jarlais, D.C., Lyles, C., Crepaz, N., 2004. Improving the reporting quality of nonrandomized evaluations of behavioral and public health interventions: the TREND statement. American Journal of Public Health 94 (3), 361–366. Edwards, D., Burnard, P., Owen, M., Hannigan, B., Fothergill, A., Coyle, D., 2003. A systematic review of the effectiveness of stress-management interventions for mental health professionals. Journal of Psychiatric and Mental Health Nursing 10 (3), 370–371. Evers, W., Tomic, W., Brouwers, A., 2002. Aggressive behaviour and burnout among staff of homes for the elderly. International Journal of Mental Health Nursing 11 (1), 2–9. Fritsch, T., Kwak, J., Grant, S., Lang, J., Montgomery, R.R., Basting, A.D., 2009. Impact of TimeSlips, a creative expression intervention program, on nursing home residents with dementia and their caregivers. The Gerontologist 49 (1), 117–127. Fuchs-Lacelle, S., Hadjistavropoulos, T., Lix, L., 2008. Pain assessment as intervention: a study of older adults with severe dementia. Clinical Journal of Pain 24 (8), 697–707. Gandoy-Crego, M., Clemente, M., Mayan-Santos, J.M., Espinosa, P., 2009. Personal determinants of burnout in nursing staff at geriatric centers. Archives of Gerontology and Geriatrics 48 (2), 246–249. Glaser, J., Lampert, B., Weigl, M., 2008. Arbeit in der stationa¨ren Altenpflege - Analyse und Fo¨rderungern von Arbeitsbedingungen, Interaktion Gesundheit und Qualita¨t. Wirtschaftsverlag NW - Verlag fu¨r neue Wissenschaft, Dortmund, Berlin, Dresden. Goldberg, D.P., Hillier, V.F., 1979. A scaled version of the General Health Questionnaire. Psychological medicine 9 (1), 139–145. Gordis, L., 2000. Epidemiologie. W.B. Saunders, Marburg. Haberstroh, J., Hampel, H., Pantel, J., 2010. Optimal management of Alzheimer’s disease patients: clinical guidelines and family advice. Neuropsychiatric Disease and Treatment 6, 243–253. Haberstroh, J., Neumeyer, K., Schmitz, B., Pantel, J., 2009. Evaluation eines Kommunikationstrainings fu¨r Altenpfleger in der stationa¨ren Betreuung demenzkranker Menschen (Tandem im Pflegeheim).(Development and evaluation of a training program for nursing home professionals to improve communication in dementia care). Zeitschrift fu¨r Gerontologie und Geriatrie 42 (2), 108–116. Hacker, W., Reinhold, S., Darm, A., Hu¨bner, I., Wollenberger, E., 1995. Beanspruchungsscreening bei Humandienstleistungen (BHD-System). TU Dresden. Institut fu¨r Allgemeine Psychologie und Methoden der Psychologie, Dresden. Jensen, L.D., Gonge, H., Jo¨rs, E., Ryom, P., Foldspang, A., Christensen, M., Vesterdorf, A., Bonde, J.P., 2006. Prevention of low back pain in female eldercare workers: randomized controlled work site trial. Spine 31 (16), 1761–1769. Jenull, B., Brunner, E., Ofner, M., Mayr, M., 2008. Burnout und Coping in der stationa¨ren Altenpflege. Ein regionaler Vergleich an examinierten Pflegekra¨ften.(Burnout and coping in long term care units. A regional comparison in registered nurses). Pflege 21 (1), 16–24. Jeon, Y.H., Luscombe, G., Chenoweth, L., Stein-Parbury, J., Brodaty, H., King, M., Haas, M., 2012. Staff outcomes from the Caring for Aged Dementia Care REsident Study (CADRES): a cluster randomised trial. International Journal of Nursing Studies 49 (5), 508–518. Kunz, R., Burnand, B., Schu¨nemann, H.J., 2008. Das GRADE-System. Ein internationaler Ansatz zur Vereinheitlichung der Graduierung von Evidenz und Empfehlungen in Leitlinien.(The GRADE System. An international approach to standardize the graduation of evidence and recommendations in guidelines). Der Internist 49 (6), 673–680. ¨ bersichtKunz, R., Khan, K.S., Kleijnen, J., Antes, G., 2009. Systematische U sarbeiten und Meta-Analysen - Einfu¨hrung in Instrumente der evi¨ denzbasierten Medizin fu¨r Arzte klinische Forscher und Experten im Gesundheitswesen. Verlag Hans Huber, Hogrefe AG, Bern. Kuske, B., Luck, T., Hanns, S., Matschinger, H., Angermeyer, M.C., Behrens, J., Riedel-Heller, S.G., 2009. Training in dementia care: a clusterrandomized controlled trial of a training program for nursing home staff in Germany. International Psychogeriatric 21 (2), 295–308. Lewis, R., Yarker, J., Donaldson-Feilder, E., Flaxman, P., Munir, F., 2010. Using a competency-based approach to identify the management behaviours required to manage workplace stress in nursing: a critical incident study. International Journal of Nursing Studies 47 (3), 307– 313. Mackenzie, C.S., Peragine, G., 2003. Measuring and enhancing self-efficacy among professional caregivers of individuals with

71

dementia. American Journal of Alzheimer’s Disease & Other Dementias 18 (5), 291–299. Mackenzie, C.S., Poulin, P.A., Seidman-Carlson, R., 2006. A brief mindfulness-based stress reduction intervention for nurses and nurse aides. Applied Nursing Research 19 (2), 105–109. Marine, A., Ruotsalainen, J., Serra, C., Verbeek, J., 2006. Preventing occupational stress in healthcare workers. Cochrane Database of Systematic Reviews (4), CD002892. Maslach, C., Schaufeli, W.B., Leiter, M.P., 2001. Job burnout. The Annual Review of Psychology 52, 397–422. McHugh, M.D., Kutney-Lee, A., Cimiotti, J.P., Sloane, D.M., Aiken, L.H., 2011. Nurses’ widespread job dissatisfaction, burnout, and frustration with health benefits signal problems for patient care. Health Affairs 30 (2), 202–210. Mimura, C., Griffiths, P., 2003. The effectiveness of current approaches to workplace stress management in the nursing profession: an evidence based literature review. Occupational and Environmental Medicine 60 (1), 10–15. Narumoto, J., Nakamura, K., Kitabayashi, Y., Shibata, K., Nakamae, T., Fukui, K., 2008. Relationships among burnout, coping style and personality: study of Japanese professional caregivers for elderly. Psychiatry and Clinical Neurosciences 62 (2), 174–176. Opie, J., Doyle, C., O’Connor, D.W., 2002. Challenging behaviours in nursing home residents with dementia: a randomized controlled trial of multidisciplinary interventions. International Journal of Geriatric Psychiatry 17 (1), 6–13. Prezewowsky, M., 2007. Demografischer Wandel und Personalmanagement: Herausforderungen und Handlungsalternativen vor dem Hintergrund der Bevo¨lkerungsentwicklung. Deutscher Universita¨tsVerlag. Richardson, B., Kitchen, G., Livingston, G., 2002. The effect of education on knowledge and management of elder abuse: a randomized controlled trial. Age and ageing 31 (5), 335–341. Robison, J., Curry, L., Gruman, C., Porter, M., Henderson Jr., C.R., Pillemer, K., 2007. Partners in caregiving in a special care environment: cooperative communication between staff and families on dementia units. The Gerontologist 47 (4), 504–515. Rodney, V., 2000. Nurse stress associated with aggression in people with dementia: its relationship to hardiness, cognitive appraisal and coping. Journal of Advanced Nursing 31 (1), 172–180. Schrijnemaekers, V.J., van Rossum, E., Candel, M.J., Frederiks, C.M., Derix, M.M., Sielhorst, H., van den Brandt, P.A., 2003. Effects of emotionoriented care on work-related outcomes of professional caregivers in homes for elderly persons. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences 58 (1), S50–S57. Van der Klink, J.J., Blonk, R.W., Schene, A.H., van Dijk, F.J., 2001. The benefits of interventions for work-related stress. American Journal of Public Health 91 (2), 270–276. Van Weert, J.C., van Dulmen, A.M., Spreeuwenberg, P.M., Bensing, J.M., Ribbe, M.W., 2005. The effects of the implementation of snoezelen on the quality of working life in psychogeriatric care. International Psychogeriatric 17 (3), 407–427. Van Wyk, B.E., Pillay-Van Wyk, V., 2010. Preventive staff-support interventions for health workers. Cochrane Database of Systematic Reviews (3), CD003541. Visser, S.M., McCabe, M.P., Hudgson, C., Buchanan, G., Davison, T.E., George, K., 2008. Managing behavioural symptoms of dementia: effectiveness of staff education and peer support. Aging and Mental Health 12 (1), 47–55. Von Ka¨nel, R., 2008. Das Burnout Syndrom: Eine medizinische Perspekktive. Praxis (Bern. 1994) 97 (9), 477–487. Weyerer, S., Schaufele, M., Schrag, A., Zimber, A., 2004. Demenzielle Sto¨rungen, Verhaltensauffa¨lligkeiten und Versorgung von Klienten in Einrichtungen der Altentagespflege im Vergleich mit Heimbewohnern: Eine Querschnittsstudie in acht badischen Sta¨dten.(Dementia disorders, behavior problems and the care of clients in geriatric daycare compared to residents in homes for the elderly: a cross-sectional study in eight communities in Baden). Psychiatrische Praxis 31 (7), 339–345. Zimber, A., 1998. Beanspruchung und Stress in der Altenpflege: Forschungsstand und Forschungsperspektiven.(Job strain and stress among geriatric caregivers: A review and a research agenda). Zeitschrift fu¨r Gerontologie und Geriatrie 31 (6), 417–425. Zimber, A., Rudolf, A., Teufel, S., 2001. Arbeitsbelastungen in der Altenpflege reduzieren: Ein Trainingsprogramm fu¨r Mitarbeiter und Leitungskra¨fte.(Reducing work stress in geriatric care: a training program for nursing team and administrators). Zeitschrift fu¨r Gerontologie und Geriatrie 34 (5), 401–407.