Causal assessment of workplace manual handling or assisting patients and low back pain: results of a systematic review.

BACKGROUND CONTEXT Low back pain (LBP) is a common musculoskeletal disorder associated with a considerable social and economic burden within the working-age population. Despite an unclear etiology, numerous physical activities are suspected of leading to LBP. Declaring a causal relationship between occupational activities and LBP remains challenging and requires a methodologically rigorous approach. PURPOSE To conduct a systematic review focused on assessing the potentially causal relationship between workplace manual handling or assisting patients and LBP. STUDY DESIGN Systematic review of the literature. SAMPLES Studies reporting an association between workplace manual handling or assisting patients and LBP. OUTCOME MEASURES Numerical association between different levels of exposure to manual handling or assisting patients, and the presence or severity of LBP. METHODS A systematic review was conducted 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 content experts. The methodological quality of each study was assessed using a modified Newcastle-Ottawa Scale (NOS) for observational studies. The overall level of evidence supporting various Bradford-Hill criteria for causality for each category of manual handling or assisting patients and type of LBP was then evaluated. Studies were deemed of higher quality if they received a score of five or more on the modified NOS and used appropriate statistical analysis methods. RESULTS This search yielded 2,766 citations, and 32 studies met the eligibility criteria. Three high-quality studies reported on manual handling and LBP, including two prospective cohorts and one cross-sectional design. None demonstrated a significant association in most of their multivariate risk estimates. One study was able to assess dose-response and temporality, but its results did not support these criteria. Only one study discussed the biological plausibility of this association. Four high-quality studies evaluated assisting patients and LBP, including two case-controls, one cross-sectional, and one prospective cohort design. These studies were consistent in reporting no significant association. Two studies demonstrated a nonsignificant dose-response trend, and two studies discussed the biological plausibility of this association. No studies were able to demonstrate the temporality or experiment criteria. CONCLUSIONS The studies reviewed did not support a causal association between workplace manual handling or assisting patients and LBP in a Bradford-Hill framework. Conflicting evidence in specific subcategories of assisting patients was identified, suggesting that tasks such as assisting patients with ambulation may possibly contribute to LBP. It appears unlikely that workplace manual handling or assisting patients is independently causative of LBP in the populations of workers studied.