Mechanisms for reducing low back pain: a mediation analysis of a multifaceted intervention in workers in elderly care

Abstract

Purpose A multifaceted workplace intervention consisting of participatory ergonomics, physical training, and cognitive–behavioural training (CBT) has shown effectiveness for reducing low back pain (LBP). However, the mechanisms of action underlying these intervention components are not well understood. Methods This was a mediation analysis of a cluster-randomised controlled trial of a multifaceted intervention in 420 workers in elderly care. Mediation analysis was carried out via structural equation modelling. Potential mediators investigated were: fear-avoidance beliefs, perceived muscle strength, use of assistive devices at work and perceived physical exertion at work. LBP outcomes assessed were: days with LBP, LBP intensity and days with bothersome LBP. Results There were no significant indirect effects of the intervention on LBP outcomes. There were significant effects of the intervention on both fear-avoidance measures [β = − 0.63, 95% CI (1.23, 0.03); β = − 1.03, 95% CI (− 1.70, − 0.34)] and the use of assistive devices [β = − 0.55, 95% CI (− 1.04, − 0.05)], but not on perceived muscle strength [β = − 0.18, 95% CI (− 0.50, 0.13)] or physical exertion [β = − 0.05, 95% CI (− 0.40, 0.31)]. The only potential mediator with a significant effect on LBP outcomes was physical exertion, which had a significant effect on LBP intensity [β = 0.14, 95% CI (0.04, 0.23)]. Conclusions A multifaceted intervention consisting of participatory ergonomics, physical training, and CBT was able to decrease fear-avoidance beliefs and increase use of assistive devices in the workplace. However, these changes did not explain the effect of any of the intervention components on days with LBP, LBP intensity and days with bothersome LBP.