Predictors of low back pain in a longitudinal study of Iranian nurses and office workers.

Background: A previous survey had indicated high rates of low back pain (LBP) in Iranian nurses and office workers.

Objective: To explore possible risk factors, we carried out a longitudinal study of the same subjects.

Methods: Baseline information about risk factors and recent history of LBP was collected by self-administered questionnaire from 246 nurses and 182 office workers. Approximately 12 months later, 385 (90%) answered a second questionnaire about LBP in the past month. Predictors of LBP at follow-up were assessed by Poisson regression and summarised by prevalence rate ratios (PRRs) with 95% confidence intervals (CIs).

Results: In a regression model that included all risk factors, the strongest predictor of LBP at follow-up was report of recent LBP at baseline. In addition, LBP was associated with older age, adverse beliefs about the work-relatedness of LBP (PRR 1.3, 95%CI 1.0-1.5), and incentives from piecework or bonuses (PRR 1.4, 95%CI 1.1-1.6). When baseline report of LBP was omitted from the model, associations were also observed with tendency to somatise, poor mental health and time pressures at work.

Conclusions: Our findings support the importance of psychosocial risk factors for LBP, including adverse health beliefs and working conditions that encourage higher output.