Purpose The aim of the study was to investigate the association between rotating shift work (RSW) and prevalence of diabetes and prediabetes in male employees of a large chemical company.

Methods We used cross-sectional data, surveyed in health checks between January 1^st2011 and December 31^st2014 at BASF SE in Ludwigshafen (Germany). A blood sample, physical examination and anamnesis by a medical officer as well as a written questionnaire were part of the health-check. Diabetes status (no diabetes, prediabetes, diabetes) was assessed via HbA_1c-value and information provided by employees themselves. Shift status was surveyed using company records. Partial proportional odds models for ordinal response variables were applied to assess the association of RSW with diabetes status.

Results Data on 9765 male employees (40.4% rotating shift workers) with a mean age of 44.3 years (SD: 9.6) could be used for analyses. Prevalence of diabetes and prediabetes in rotating shift workers were 6.4% and 14.2% respectively. Corresponding figures for day workers were 3.6% and 11.5% (p<0.001). Unadjusted analyses yielded an Odds Ratio of 1.85 (95%-CI: 1.53–2.22) for RSW (comparing diabetes with prediabetes/no diabetes). The effect attenuated in the fully adjusted multivariable model (aOR: 1.38; 95% CI: 1.11–1.70) but remained significant. The association between RSW and diabetes status varied by BMI and occupational status. The association was stronger in skilled/supervisory workers and respondents with a BMI <30 kg/m².

Conclusions The present study found a significant association between RSW and diabetes which differed by BMI and occupational status.

Declaration of potential conflict of interest: All authors are employees of BASF SE.