Document resource
Objective To estimate the effect of cumulative occupational lifting on the risk of lumbar spine disc herniation. To discuss different methods for assessing the effect of cumulative occupational lifting on the risk of lumbar disc herniation.Material and Methods In this nationwide register-based cohort study, we included Danes aged 18–25 years in 1995, corresponding to ~570,000 individuals with up to 31 years of follow-up. Cumulative lifting was estimated annually using job exposure matrices (JEMs) linked to occupational registers, whereas lumbar disc herniations and covariates were obtained from registers. We performed a pooled logistic regression with time-varying exposure adjusted for the following baseline and time-varying confounders: calendar year, age, sex, education, industry, family income, and marital status.Results The pooled logistic regression model yielded the following HRRs (95%CI):<250 ton Ref.250-750 ton 1.20(1.13-1.28)751-1500 ton 1.31 (1.23-1.38)1501-2500 ton 1.47(1.39-1.56)2501-4000 ton 1.59(1.49-1.69)4001-6000 ton 1.53(1.42-1.64)>6000 ton 1.59(1.46-1.72)We are currently performing a target trial emulation, where we emulate an RCT evaluating different lifting thresholds (e.g. ‘lift below 750 tons’). Each individual is cloned to seven lifting threshold arms, where the clone is censored when the observed exposure exceeds the assigned threshold. Inverse probability of censoring weights account for the selection bias introduced by censoring, and for baseline and time-varying confounding, and exposure-confounder feedback.Conclusion Preliminary results showed an increased risk of lumbar disc herniation with increasing cumulative lifting. However, these results require very strong assumptions to be considered causal and are likely influenced by exposure-confounder feedback (e.g. healthy worker biases).Funding We thank ‘the Danish Working Environment Research Fund’ and ‘NIDO | Centre for Research and Education’ for their financial support.