Eligibility criteria for the review
Types of studies
This review will include empirical, quantitative studies that are randomised controlled trials including parallel, cross-over and cluster-randomised designs.
Participants
The review will consider work and person-directed interventions for direct-care staff employed in RACFs. Interventions involving aged care staff not employed in direct-care roles in RACFs, aged care staff not employed in RACFs or other family members caring for the older adults will be excluded. For studies involving some RAC staff and some family members as carers, data from RAC staff are included if datasets for these staff were reported separately.
Settings
Residential aged care facilities (RACFs) are defined as aged care homes that provide respite or long term care for older adults who are not able to live at home and whose demands require greater care and support (Productivity Commission, 2019). RACFs, are known by different names globally, such as nursing homes, long-term care facilities and care homes (Rayner et al., 2019). For studies conducted across settings such as community and residential care, data from residential care settings are included if datasets for these staff are reported separately. Studies from all geographical locations will also be considered.
Interventions
This review will consider work and person-directed interventions for preventing occupational burnout in RAC staff employed in RACFs. Person-directed interventions are defined as interventions that are focussed on the individual and can include educational training, engaging in mental and physical relaxation, utilising cognitive behavioural therapy (CBT) and combined approaches including CBT and relaxation (Ruotsalainen et al., 2015). Work-directed interventions are defined as interventions that focused on changing the work environments of staff members such as having organisational support and improving their work schedules (Ruotsalainen et al., 2015). Prevention can be defined as interventions focussed on healthy individuals without a diagnosable condition (Ruotsalainen et al., 2015).
Time frame
In this review, studies from 1 January 2000 to 31 October 2021 will be searched.
Outcome measures
This review will consider studies that focus on occupational stress and burnout prevention as outcome measures. Burnout can be measured by the various burnout scales such as Maslach Burnout Inventory, Burnout Measure, Oldenburg Burnout Inventory and Copenhagen Burnout Inventory. The effects of such interventions on related phenomena such as stress and anxiety will also be measured. Secondary outcome measures can include psychological symptoms such as depression, anxiety and stress.
Search methods for identification of studies
This review will consider published peer-reviewed studies, written in English. Electronic databases for published data used for this search will include Medical Literature Analysis and Retrieval System Online (MEDLINE), EMBASE, PsycINFO and CINAHL. Filters to identify randomised controlled trials will be applied where applicable. The search terms used will be as follows:
(burnout OR “occupational stress” OR “psychological stress”) AND
(intervention OR effect* OR evaluation) AND
(healthcare OR staff or employ* OR job OR work* OR nurs*) AND
(“long-term care” OR “residential aged care” OR “aged care” OR “nursing home*” OR “assisted living” OR “care facilit*” OR “residential home*” OR “care home*” OR “residential care”) AND
(“random* controlled trial*”)
Handsearching of reference lists of systematic reviews will also be conducted. For studies with missing data or ambiguity in research design and methods, the researchers will attempt to contact authors of the study twice for further clarification.
Data collection and analysis
Selection of studies
Studies will be selected by two reviewers independently using Cochrane’s software, Covidence. Titles and abstracts of articles will be retrieved and screened. Next, the reviewers will screen the full text of relevant studies to assess their eligibility for inclusion. Should there be differences in opinion amongst the two reviewers, the reviewers will attempt to resolve these differences by discussion. If disagreement persists between the reviewers, a third reviewer will be consulted to reach a majority agreement.
Data extraction and management
Two reviewers will extract and manage the data independently. Covidence software will be utilised to record data of relevant studies. The extracted data will include some items from the template for intervention description and replication (TIDieR) checklist (Hoffmann et al., 2014) such as the following: brief name of study (item 1), theory related to the intervention (item 2), materials used in the intervention (item 3), procedures (item 4), who provided the intervention (item 5), how was the intervention delivered (item 6), where the interventions occurred (item 7), when and how much- number of sessions, schedule, duration, intensity and dose (item 8).
Assessment of methodological quality of included studies
Assessment of the quality of included studies will be conducted by two reviewers independently. This review will utilise the Cochrane RoB 2 tool (Higgins et al., 2019) which can help assess the methodological quality of randomised controlled trials. The tool consists of 5 types of bias domain including bias arising from the randomization process, deviations from intended interventions, missing outcome data, measurement of the outcome and selection of the reported result (Higgins et al., 2019).
Measures of treatment effect
The intervention effect will be estimated using the SMD with 95% CIs.
Assessment of heterogeneity
We will assess heterogeneity by visually inspecting the forest plots and by using a standard chi-square test with a significance level of α = 0.1 (Deeks et al., 2019). In view of the low power of the test, the I2 statistics, which quantifies inconsistency across trials to assess the impact of heterogeneity on the meta-analysis, will also be considered (Higgins & Thompson, 2002; Higgins et al., 2003). In the event of substantial clinical or methodological heterogeneity, we will not report trial results as the pooled effect estimate in a meta-analysis.
Assessment of reporting biases
If 10 or more trials are included to examine a particular outcome, we will use funnel plots to assess small-trial effects. Potential justifications to account for funnel plot asymmetry include true heterogeneity of effect with respect to trial size, poor methodological design and publication bias (Boutron et al, 2019).
Data synthesis
The quality of included studies will be assessed using the GRADE system and similar studies will be combined in meta-analyses. Standardised Mean Differences (SMDs) will also be utilised in order to combine studies with different burnout scales or measures. Results will be presented in a table format. Any person-directed and work-directed interventions for preventing burnout will be grouped in a single comparison in order to address whether person-directed or work-directed interventions for burnout is effective. A standard meta-analysis will be conducted where person-centred interventions will be compared to inactive control interventions (usual practice or care). In the event where there is sufficient data, separate meta-analyses will be considered to examine the effects of specific types of person-centred interventions (for example, educational training, mental and physical relaxation, teaching CBT and combined CBT and relaxation). When a specific person-centred intervention comprise multiple components, the main feature of the intervention must be clearly defined and categorised accordingly.