Linking data on work,health and lifestyle to explain socio-occupational inequality in Danish register-based incidence of diabetes

Abstract

Aims: This study aimed to combine three consecutive cohorts representative of the general working population to estimate the lifestyle, work and health of the entire Danish working population and to explore the influence of these factors on socio-occupational inequality in relation to incident diabetes over ten years. Methods: This study linked the National Danish Diabetes Register with (a) three questionnaire rounds (1995, 2000, 2005) on work, health and lifestyle from the Danish Work Environment Cohort Study (DWECS) among 6823 representative Danish workers aged 30–59 years and (b) the different socio-occupational groups of the entire Danish population aged 30–59 years (n=1,833,756). The relative risk (RR) was calculated using a Poisson regression. Results: In DWECS, none of the investigated job factors was associated with incident diabetes. Lifestyle factors in terms of smoking (RR = 1.35; 95% confidence interval (CI) 1.07–1.70), high body mass index (overweight RR = 2.81; 95% CI 2.11–3.74 and obesity RR = 7.49; 95% CI 5.46–10.28) were risk factors for incident diabetes. When adjusted for health and lifestyle, there was no difference in diabetes risk among socio-occupational groups in DWECS. In the entire working population, there was a negative socio-occupational gradient for the risk of diabetes. Compared with professionals, workers in elementary occupations had the highest ten-year risk (RR = 2.06; 95% CI 2.01–2.12). Conclusions: We found a significant negative socio-occupational gradient in the risk of incident diabetes. However, lifestyle rather than work environment appears to be the most important factor and adjusting for this almost eliminated the socio-occupational inequality for the development of diabetes.