Pertussis infection in childhood and subsequent Type 1 diabetes mellitus

Aims Pertussis has been implicated but not proven as a risk for Type 1 diabetes mellitus (DM). Previous studies have investigated paediatric, but not adult-onset Type 1 DM. We investigated association of pertussis exposures and Type 1 DM with follow-up into adulthood. Methods Longitudinal analysis of 16 820 members (100 with Type 1 DM) of two nationally representative British birth cohorts (the 1970 British Cohort Study (BCS70) and the National Child Development Study (NCDS)) followed from birth to ages 30 years (BCS70) and 42 years (NCDS). Cox regression analysis with age of onset for Type 1 DM as the dependent variable investigated relationships with pertussis infection and immunization, modelled as time-dependent co-variates. Simultaneous adjustment was made for Wild measles, mumps and chickenpox infections; tetanus and smallpox immunizations; sex, parental social class and cohort. The potential confounding factors were modelled as fixed co-variates. Results Cox regression analysis produced adjusted odds ratios (ORs) (with 95% confidence intervals (Cls)) of 2.21 (1.35-3.59) and 0.73 (0.49-1.05) for Type 1 DM (with onset at any age) associated with pertussis infections and immunization (trend over number of vaccinations), respectively. Adjusted ORs from Cox regression for Type 1 DM with onset after age 10 years are 2.59 (1.56-4.30) for pertussis infection and 0.63 (0.42-0.94) for pertussis immunization. None of the other infections or immunizations are notably associated with Type 1 DM. Conclusions Some pertussis infections may be a risk for Type 1 DM and immunization may confer protection. Further research should consider delayed Type 1 DM following pertussis exposures.