Socio-economic status and incident diabetes mellitus among employees in Denmark: a prospective analysis with 10-year follow-up

Aims To contribute to the research on diabetes and social inequality by presenting national data on incident diabetes mellitus, stratified according to socio-economic status. Methods National registers were combined, linking socio-economic status with incident diabetes over a 10-year period (2001-2010). The study population comprised employees in Denmark aged 20-59 years at baseline. Poisson regression analysis was used to estimate socio-economic rate ratios. Excess fraction analysis was used to determine the proportion of cases that would not have occurred if morbidity rates in each socio-economic group had been as low as those in the reference group. Monte Carlo simulation was used to calculate 95% CIs for excess fraction estimates Results A total of 1 005 572 men and 951 039 women were included in the analysis. The follow-up yielded 43 439 cases in 9 533 199 person-years at risk among men and 29 266 cases in 9 163 405 person-years at risk among women. Using 'professionals' as a reference group, higher levels of relative risk were observed among every other socio-occupational group. The excess fraction was, 0.342 (95% CI 0.329-0.354) among men and 0.359 (95% CI 0.349-0.369) among women. Conclusions Excess fraction analysis suggests that more than a third of cases of diabetes could be prevented if all employees were exposed to the same working conditions as the reference population. Acknowledging potential confounders, the observed levels of incident diabetes among the workforce highlight the potential gains to be had by better use of the workplace as an arena for prevention. Greater integration of occupational health and general healthcare is required to achieve this.