Using a cohort of 8183 postal workers, this study assesses the efficacy of preplacement medical examinations in defining the risk of occupational low back injuries. From this cohort, 154 subjects with occupational low back injuries between 1983 and 1988 and 942 control subjects who did not have low back injuries were identified. A multivariate logistic regression shows that a history of prior disability, odds ratio 2.90 (95% confidence interval 1,88–4.48), and a heavy lifting job, odds ratio 1.91 (1.32–276) are associated with occupational low back injuries. However, a history of previous back injury on screening examination is not associated with subsequent occupational injury. The association between a history of disability and occupational low back injury has not been previously noted and warrants further research.