Return to work in employees with depression
Cochrane Corner is contributed to Suva Medical by Cochrane Insurance Medicine. An updated Cochrane Review (45 studies involving 12,109 people) on interventions to improve return to work in employees with depression confirms the importance of workplace interventions that work in conjunction with medical treatment.
Content
The case
Mr Dürr, a 43-year-old operations manager at a mid-sized business, was in a motorcycle accident 12 months ago during which he sustained a severe spinal fracture accompanied by temporary paraparesis. After undergoing surgery, he suffered from continuing bladder and sexual dysfunction in spite of extensive rehabilitation measures. Under these circumstances and fearing for his future, he manifested with depression, marked by social anxiety and disturbed sleep. In the course of psychiatric and psychotherapeutic treatment involving antidepressant medication and a positive rehabilitation outcome, the depressive symptoms improved.
Physically, Mr Dürr had then become able to return to work. Nonetheless, he became anxious about whether or not he was still up to the task. Before the accident, he was responsible for managing 20 employees, supervising production using machinery and managing purchasing. The role is highly demanding in terms of the required flexibility, responsibility for decision-making and assertiveness – qualities which the insuree could still only offer to a limited extent. In light of the anticipated challenges, his depressive symptoms returned. In his quarterly meeting, the insuree asked for support from his case manager, who referred him to the consultant psychiatrist. This case recalls a recent training session which discussed a Cochrane meta-analysis about interventions to improve return to work in employees with depression (Nieuwenhuijsen K et al. CDSR 2020). Nine studies involving 1,292 employees demonstrated that employees with depression returned to work sooner if their workplace made specific work-directed interventions in addition to psychiatric treatment (table). Extrapolated over one year, employees who received combined interventions returned to work 25 days earlier. Depressive symptoms also subsided quicker (eight studies involving 1,025 employees) and the employees felt more resilient at work (five studies, 926 employees). The consultant psychiatrist is of the view that the results of the meta-analysis for an additional work-directed intervention are convincing enough to justify covering the costs. Disability insurance is arranging this intervention.
Table
Comparison of work‐directed interventions combined with clinical interventions versus usual care (period of observation in the studies varies from two months to one year)
|
Target value |
Number of studies |
Number of participants |
Statistical method |
Effect size [95% CI] |
Evaluation: |
|---|---|---|---|---|---|
|
1. Numberof days of sickness absence |
9 |
n = 1 292 |
SMD+ |
-0,25 |
moderately robust |
|
2. Employ-ment statusunfit for work |
2 |
n = 1 025 |
Relative risk |
1,08 |
highly robust |
|
3. Depressivesymptoms |
8 |
n = 1 091 |
SMD+ |
-0,25 |
Not veryrobust |
|
4. Resili-ence in theworkplace |
5 |
n = 926 |
SMD+ |
-0,19 |
moderately robust |
Key
*For the evaluation of the «robustness of evidence» see annex belowSMD+ = standardised mean difference. We report the difference between the combined intervention «Adjustments in the workplace combined with clinical intervention» and «the usual care» as a standardised mean difference (SMD). The SMD enables results from multiple studies to be brought together into a single meta-analysis. An SMD of –0.2 or 0.2 indicates a slight effect; an SMD of –0.5 or 0.5 indicates a moderate effect; and an SMD of –0.8 or 0.8 indicates a large effect.
The evidence
The Cochrane meta-analysis «Interventions to improve return to work in depressed people»
Objective of the meta-analysis
This review analyses interventions designed to reduce periods of work disability in employees with depression. Employees with depression often experience short-term, medium-term or even long-term work disability. Returning to work is an important task for the employee themselves, for the employer and for society. Various interventions can be made to achieve this. The review investigates work-directed interventions to improve return to work with or without clinical interventions.
The main objective of work-directed interventions is to reduce the impact of depression on the ability to work. The majority of the studies combined multiple individual interventions, such as contact with supervisors, gradual return to work programmes, adjusting work requirements or working hours, or learning coping strategies for dealing with depression at work. Clinical interventions such as pharmacotherapy, psychotherapy and physiotherapy in line with guidelines are used to reduce depressive symptoms and thus improve ability to work. These interventions were compared with «the usual care» to answer the following questions:
- How many days were employees with depression on sick leave?
- How many employees with depression were off work (at the end of the study)?
- How serious were people’s symptoms of depression?
- How well could employees with depression cope with their work?
What are the results of the meta-analysis?
We found 45 studies involving 12,109 employees with depression. The studies come from Europe (n=34), the USA (n=8), Australia (n=2) and Canada (n=1). We present the key findings for other comparisons and a post-assessment period up to one year:
Workplace changes combined with a clinical programme
- probably reduce the number of days on sick leave, by 25 days on average for each person over one year (9 studies; 1,292 participants) 🙂
- do not reduce the number of people off work at the end of the study (two studies; 1,025 participants) 🙁
- reduce symptoms of depression (eight studies; 1,091 participants) 🙂
- may improve ability to cope with work (five studies; 926 participants) 🙂
Workplace changes alone
- may increase the number of days on sick leave (two studies, 130 participants) 😢
- do not reduce the number of people off work at the end of the study (one study; 226 participants) 😐
- probably do not affect symptoms of depression (four studies; 390 participants) 😐
- may not improve ability to cope with work (one study; 48 participants) 😐
Improved healthcare alone without work-directed interventions (e.g. inclusion in a quality assurance programme, care in line with guidelines, regular phone contact with therapists)
- probably reduces the number of days on sick leave by 20 days (in two well‐conducted studies involving 692 participants, although not in all seven studies involving 1,912 participants) 😐
- probably reduces symptoms of depression (seven studies; 1,808 participants) 🙂
- may reduce ability to cope with work (one study; 604 participants) 🙁
Psychotherapy alone without work-direct interventions
- may reduce the number of days on sick leave by 15 days on average (over the course of one year) (nine studies; 1,649 participants) 🙂
- may reduce symptoms of depression (eight studies; 1,255 participants) 😐
- It remains unclear if psychological therapies alone affect people’s ability to cope with work (one study; 58 participants) 😐
| Key | |
| 🙂 | improvement due to the intervention investigated compared to the control intervention |
| 😐 | no definitive difference between the intervention investigated and the control intervention |
| 🙁 | no difference between the intervention investigated and the control intervention |
| 😢 | exacerbation due to the intervention investigated compared to the control intervention |
Robustness of evidence terminology (Lietz et al. 2020)
The «robustness of evidence» was evaluated by the group of authors.
|
Evaluating the robustness of evidence |
Definition |
|---|---|
|
⊗⊗⊗⊗ Highly robust |
The group of authors is very sure that the real effect is close to the observed effect. |
|
⊗⊗⊗O Moderately robust |
The group of authors is reasonably confident in the observed effect: the real effect is probably close to the observed effect, but it might be substantially different. |
|
⊗⊗OO Not very robust |
The group of authors has limited confidence in the observed effect: there is a high chance that the real effect could differ substantially from the observed effect. |
|
⊗OOO Not robust at all |
The group of authors has very little confidence in the observed effect: the real effect is probably substantially different from the observed effect. |
| Evaluating the robustness of evidence | Definition |
| ⊗⊗⊗⊗ Highly robust | The group of authors is very sure that the real effect is close to the observed effect. |
| ⊗⊗⊗O Moderately robust | The group of authors is reasonably confident in the observed effect: the real effect is probably close to the observed effect, but it might be substantially different. |
| ⊗⊗OO Not very robust | The group of authors has limited confidence in the observed effect: there is a high chance that the real effect could differ substantially from the observed effect. |
| ⊗OOO Not robust at all | The group of authors has very little confidence in the observed effect: the real effect is probably substantially different from the observed effect. |
Correspondence address
Heike Raatz
Cochrane Coordinatorin
Cochrane Insurance Medicine
Departement Klinische Forschung
Universitätsspital Basel
Bibliography
- Nieuwenhuijsen K, Verbeek JH, Neumeyer-Gromen A, Verhoeven AC, Bültmann U, Faber B. Interventions to improve return to work in depressed people. Cochrane Database of Systematic Reviews 2020, Issue 10. Art. No.: CD006237. DOI: 10.1002/14651858.CD006237.pub4.
- Lietz M, Angelescu K, Markes M, Molnar S, Runkel B, Schell L, et al. GRADE: Von der Evidenz zur Empfehlung oder Entscheidung – Entscheidungen zur Kostenerstattung. Z Evid Fortbild Qual Gesundhwes. 2020;150-152:134-41.