A1981 Psychosocial Stressors and their impact on musculoskeletal problems: A meta-analysis of longitudinal studies

Wednesday, March 21, 2012: 15:35
Xcaret 4 (Cancun Center)
Jessica Lang, Institute for Occupational Medicine, Rwth Aachen University, Aachen, Germany
Elke Ochsmann, Insitute for Occupational Medicine, RWTH Aachen University, Aachen, Germany
Thomas Kraus, Insitute for Occupational Medicine, RWTH Aachen University, Aachen, Germany
Jonas Wb Lang, Faculty of Psychology & Neuroscience, Maastricht University, Maastricht, Netherlands
Introduction
The relationship between psychosocial workplace conditions and musculoskeletal problems has largely been studied, yet the causal impact of psychosocial workplace factors for the development of musculoskeletal problems remains unclear (e.g., Hartvigsen et al., 2004; Steenstra et al., 2005). The purpose of the present meta-analysis was to conduct a systematic review of baseline-adjusted prospective longitudinal studies estimating the lagged effect of psychosocial workplace stressors on musculoskeletal problems.

Methods
Literature review was conducted by searching the MEDLINE, EMBASE, and PsychINFO databases until August 2009. The authors independently classified studies into categories of psychological work stressors, and musculoskeletal problems. Available effect sizes were converted to Odds ratios (OR). ORs were then pooled for each stressor–problem relationship using a random-effects model. The possibility of publication bias was assessed with the Duval and Tweedie nonparametric “trim and fill” procedure.

Results
At least five effect sizes were available for 20 of the 45 possible psychosocial work stress–musculoskeletal problems relationships. For 20 of these stressor–problem relationships, pooled OR estimates were positive and significant. Pooled ORs ranged from 1.153 to 1.579 with the largest pooled OR estimating the relationship between highly monotonous work and lower back pain. The lagged effect of low social support on lower back problems was the only effect size for which the statistical test for bias was significant. However, the corrected effect size remained significant. The large majority of the primary studies in our meta-analysis did not report the reliability of the measures and dichotomized continuous independent and outcome variables. Thus the estimates, likely seriously underestimate the true lagged effects.

Discussion
Most psychosocial stressor had small but significant lagged effects on the development of musculoskeletal problems. Thus, organizational interventions to minimize these stressors may be promising to prevent musculoskeletal problems at the workplace.