Causal assessment of occupational bending or twisting and low back pain: results of a systematic review

Background contextLow back pain (LBP) is a common musculoskeletal disorder that often occurs in the working-age population. Although numerous physical activities have been implicated in its etiology, determining causation remains challenging and requires a methodologically rigorous approach.PurposeTo conduct a systematic review focused on establishing a causal relationship between occupational bending or twisting and LBP.Study designA systematic review of the literature using Medline, Embase, CINAHL, Cochrane Library, and Occupational Safety and Health database, gray literature, hand-searching occupational health journals, reference lists of included studies, and experts. Evaluation of methodological quality using a modified Newcastle-Ottawa Scale for observational studies. Summary levels of evidence for each of the Bradford-Hill criteria for causality for each category of bending or twisting and type of LBP.SampleStudies reporting an association between occupational bending or twisting and LBP.Outcome measuresNumerical association between different levels of exposure to bending or twisting and the presence or severity of LBP.MethodsA systematic review was performed to identify, evaluate, and summarize the literature related to establishing a causal relationship, according to Bradford-Hill criteria, between occupational bending or twisting and LBP.ResultsThis search yielded 2,766 citations. Ten high-quality studies reported on bending and LBP. Five were case-control studies and five were prospective cohort studies. There was conflicting evidence for association, with five studies demonstrating significant associations in the majority of their risk estimates, but no evidence for consistency. Seven studies assessed dose response, with four studies demonstrating a nonsignificant dose-response trend. Four studies were able to assess temporality, but only one demonstrated significant risk estimates. Biological plausibility was discussed by two studies. There was no available evidence for experiment. Seven high-quality studies reported on twisting and LBP. Two were case-control studies and five were prospective cohort studies. Three studies reported significant associations in the majority of their risk estimates, with no evidence for consistency. Three studies demonstrated a nonsignificant dose-response trend. Two studies were able to assess temporality, but only one study was able to demonstrate significant risk estimates. Two studies discussed biological plausibility. There was no available evidence for experiment.ConclusionsA summary of existing studies was not able to find high-quality studies that satisfied more than three of the Bradford-Hill criteria for causation for either occupational bending or twisting and LBP. Conflicting evidence in multiple criteria was identified. This suggests that specific subcategories could contribute to LBP. However, the evidence suggests that occupational bending or twisting in general is unlikely to be independently causative of LBP.