Skip to main content

Table 3 Certainty of evidence of work-related interventions

From: Effectiveness of work-related interventions for return to work in people on sick leave: a systematic review and meta-analysis of randomized controlled trials

Population: Adults on full or partly sick leave

Countries: Denmark, Norway, Sweden, The Netherlands

Intervention: Work-related interventions

Comparison: Usual care or other active intervention

Outcome, follow-up time

Relative effect

(95% KI)

Anticipated absolute effects (95% CI)

Number of participants

(studies)

Quality of evidence

(GRADE)

Assumed risk with control

Assumed risk with intervention

Absolute difference (intervention minus control)

Multidisciplinary rehabilitation vs. UC

 Return to work (12 mo)

RR = 1.01 (0.70–1.48)

72 per 100

73 per 100 (50 to 107)

1 more person

321 participants

(3 RCTs)

Very lowa,b,c

 Return to work (12 mo)

-

-

-

One study found no difference between the groups. One study found that the intervention group had shorter time on sick leave than the control group.

1891 participants

(2 RCTs)

Very lowa,b,c

Multidisciplinary rehabilitation vs. active treatment

 Return to work (12 mo)

RR = 1.04 (0.86–1.25)

59 per 100

61 per 100 (51 to 74)

Three more persons (14 less to 25 more)

851 participants (4 RCTs)

Very lowa,b

 Return to work (24 mo)

RR = 0.94 (0.84–1.05)

63 per 100

59 per 100

Four less persons (16 less to 5 more)

635 participants

(2 RCTs)

Moderated

W-CBT vs. UC

 Return to work

-

-

-

Two studies showed faster RTW in the intervention group compared to UC at 4 and 6 mo. One study showed no difference between the groups at 18 mo.

311 participants

(3 RCTs)

Lowa

W-CBT vs. active intervention

 Return to work (12 mo)

27 (− 8.7, 62.8) days less sick leave than control group 1,

18 (− 15.8, 52.1) days less sick leave than control group 2

-

-

No difference between the groups in number of sick days

211 participants

(1 RCTs)

Very low e,f

Problem-solving approach vs. UC

 Return to work (3 mo)

-

39 per 100

37 per 100

No difference between the groups in days on sick leave

194 participants (1 RCT)

Very lowg,h

 Return to work (6 mo)

-

62 per 100

58 per 100

No difference between the groups in days on sick leave

194 participants (1 RCT)

Very lowg,h

 Return to work (18 mo)

HR = 1.04 (0.76–1.42)

79 per 100

85 per 100

No difference between the groups in days on sick leave

194 participants (1 RCT)

Very lowg,h

Additional dialogue meeting vs. active treatment

 Return to work

MD = 48 (− 2.9, 100.8)

  

No difference between the groups in days until full RTW

60 participants (1 RCT)

Very lowf,i

 Days to first RTW

MD = − 2.02 (− 28.2, 24.2)

  

No difference between the groups in days until the first RTW

60 participants (1 RCT)

Very lowf,i

Adjuvant occupational therapy vs. UC

 Full return to work (18 mo)

HR = 0.93 (0.57–1.53)

-

-

No difference between the groups

117 participants (1 RCT)

Very lowa,f

 Partly return to work (18 mo)

HR = 0.72 (0.44–1.11)

-

-

No difference between the groups

117 participants (1 RCT)

Very lowa,f

Stress treatment program vs. UC or waiting list

 Return to work (3 mo)

OR = 8.1 (3.2–20.7)

-

-

  

Very lowf,i

MR with additional workplace meeting vs. MR

 Return to work (12 mo)

HR = 0.74 (0.48–1.16)

52 per 100

42 per 100

Sustainable RTW 42% in the intervention group vs. 52% in the control group

175 participants (1 RCT)

Very lowd,f

E-health module with collaborative occupational health care vs. UC

 Full return to work (12 mo)

-

Median days 178 (IQR 72.0–243.3)

Median days 131 (IQR 68.5–198)

Median of 47 days faster full RTW in the intervention group compared to the control group

131 participants (1 RCT)

Very lowd,f

 First return to work (12 mo)

HR = 1.39 (1.03–1.87)

77 days until the first RTW

50 days until the first RTW

27 days faster first RTW in the intervention group compared to the control group

131 participants (1 RCT)

Very lowd,f

  1. CI confidence interval, RCT randomized controlled trial, HR hazard ratio, MD mean difference, OR odds ratio, RR risk ratio, RTW return to work
  2. aDowngraded 2 levels for risk of bias (selection and performance bias)
  3. bDowngraded 1 level for inconsistency
  4. cDowngraded 1 level for imprecision (confidence interval encloses negative and positive effect)
  5. dDowngraded 1 level for risk of bias (performance bias)
  6. eDowngraded 1 level for risk of bias (reporting and performance bias)
  7. fDowngraded 2 levels for imprecision (confidence interval encloses negative and positive effect and low number of events)
  8. gDowngraded 2 levels for risk of bias (selection, reporting, and other bias)
  9. hDowngraded 1 level for imprecision (low number of events)
  10. iDowngraded 2 levels for risk of bias (selection, performance, and attrition bias)