Background
Methods
Search strategy for the identification of relevant intervention studies
Assessment of the risk of bias and level of evidence
Criteria | Rationale | Weights |
---|---|---|
Is there a control group? | Reduction of bias of the estimates of intervention effects | 0 or 1 |
Is the baseline response greater than 70% | Reduction of sampling bias | 0 or 1 |
Is the follow-up response greater than 50% | Reduction of attrition and turnover bias | 0 or 1 |
Have the authors adjusted for non-response and drop-out? | Reduction of bias of the estimates of intervention effects by adjusting for attrition and turnover bias | 0 or 1 |
Is there adequate adjustment for the majority of known confounders? | Reduction of bias of the estimates of intervention effects by adjusting for relevant confounders | 0 or 1 |
Where appropriate statistical tests used? | Reduction of bias of the estimates of intervention effects by using appropriate statistical methods | 0 or 1 |
Sample size at baseline. 0 if n is in 1st quantile, 2 if n is in the 2nd quantile, 3 if n is in the 3rd quantile | Validity of statistical inference | 0 to 3 |
Sample size at follow-up. 0 if n is in 1st quantile, 2 if n is in the 2nd quantile, 3 if n is in the 3rd quantile. | Validity of statistical inference | 0 to 3 |
Level | Definition |
---|---|
High | The study has a randomized controlled design whose quality is greater than the median study quality. |
Medium | The study is either a randomized controlled design whose quality is less than the median study quality, or the study is a quasi-experimental control study whose quality is greater than the median study quality |
Low | The study is either a quasi-experimental control trial whose quality is less than the median study quality, or the study is a quasi-experimental one-group prospective or a cross-sectional one. |
Classification of the major modifications of the working conditions implemented by the interventions
Estimation of the probability of reporting significant intervention effects
Results
Variable | Number of studies/median and range |
---|---|
EGP occupational class
| |
I-III | 18 |
I-III, VI-VII | 6 |
VI-VII | 15 |
Study design
| |
Quasi-experimental, cross-sectional | 1 |
Quasi-experimental one-group prospective | 9 |
Quasi-experimental prospective with control group | 21 |
Quasi-experimental prospective and retrospective | 2 |
Randomized controlled | 6 |
Evidence grade
| |
High | 4 |
Medium | 25 |
Low | 10 |
Significant intervention effects
| |
No | 21 |
Yes | 18 |
Intervention type
| |
Organisational intervention (OI) | 9 |
Participatory (including Participatory Ergonomics PE and Participatory Research Action PAR) | 7 |
Shift schedules | 18 |
Other | 5 |
Year of publication | 2005 [1993, 2012] |
Sample size at baseline | 300 [41, 3506] |
Sample size at follow-up | 187 [36, 2617] |
Sample size of control group | 122 [31, 347] |
Follow-up time (years) | 1 [0.25, 13] |
Study quality | 7 [1,11] |
Number of working conditions changed | 2 [1,4] |
Statistically significant? | |||
---|---|---|---|
Working condition | Number of modified working conditions | No | Yes |
Material | 1 | 1 | 1 |
Material, organisation | 2 | 4 | 6 |
Material, time, organisation | 3 | 0 | 3 |
Organisation | 1 | 10 | 6 |
Time | 1 | 3 | 2 |
Time, organisation | 2 | 2 | 1 |
OR | Lower 95% CI | Upper 95% CI | P | |
---|---|---|---|---|
Number of modified working conditions | 2.71 | 0.94 | 11.12 | 0.07 |
Study quality | 0.95 | 0.75 | 1.18 | 0.63 |
EGP class
| ||||
Reference I-III | ||||
I-III, VI-VII | 0.42 | 0.05 | 2.66 | 0.37 |
VI-VII | 0.73 | 0.14 | 3.14 | 0.68 |