Article: Epidemiology Parity and risk of diabetes in a Danish nationwide birth cohort K. V. Naver, S. Lundbye-Christensen*, A. Gorst-Rasmussen†, L. Nilas, N. J. Secher, S. Rasmussen‡ and P. Ovesen§ Department of Obstetrics and Gynaecology, Copenhagen University Hospital Hvidovre, Denmark, *Department of Cardiology, Centre for Cardiovascular Research, Aalborg Hospital Aarhus University Hospital, Aalborg, Denmark, †Department of Mathematical Sciences, Aalborg University, Aalborg, Denmark, ‡The Danish National Board of Health, Monitoring & Health Technology Assessment, Copenhagen, Denmark and §Department of Obstetrics and Gynaecology, Aarhus University Hospital, Skejby, Aarhus, Denmark Accepted 18 October 2010 Abstract Aims The purpose was to elucidate the association between parity and the incidence of diabetes using national register data. Methods The study population consisted of all Danish women with a singleton delivery in 1982 / 1983 (n = 100 669), who subsequently had 74 966 deliveries. The included women were followed up via registries until the end of 2006 for subsequent deliveries, diagnosis of diabetes and death / emigration. Results A total of 2021 cases (2.0%) were diagnosed with diabetes in connection with hospitalization or outpatient treatment during follow-up. Analyses were adjusted for fetal weight and duration of gestation, both at index pregnancy. Cox regression analysis with parity as a time-varying exposure, stratified in two age groups, showed an association between parity and risk of a diagnosis of diabetes. In women <33 years of age, parity 2, 3 and 4 + were associated with an increased risk of being diagnosed with diabetes compared with parity 1 [relative risks: 1.6 (95% confidence interval 1.1–2.3), 2.8 (1.8–4.3) and 2.5 (1.3–4.8), respectively]. Among women >33 years of age, parity 2 was associated with a significantly lower risk of diabetes diagnosis compared with parity 1, whereas parity 4 + was associated with a significantly higher risk of diabetes diagnosis compared with parity 1. Conclusions The study shows that the risk of diabetes diagnosis increases with parity in young Danish women. This may support a causal association between diabetes and parity. Diabet. Med. 28, 43–47 (2011) Keywords cohort, diabetes, multiparity, parity, pregnancy Abbreviations BMI, body mass index; CI, confidence interval; RR, relative risk Introduction The prevalence of diabetes is escalating worldwide, mainly due to an increase in Type 2 diabetes. Some risk factors are well known, including age, adiposity and gestational diabetes, whereas the relationship between parity and diabetes has been a matter of discussion for many years. Some investigations have found that parity, particularly five or more births, may be associated with a higher incidence of diabetes [1–6], whereas others have found no association [7–10]. Glucose homeostasis is altered during pregnancy. Peripheral insulin sensitivity is reduced due to the action of several diabetogenic hormones [11,12], and the increased need for insulin is met by expansion of the ß cell mass [13]. Whether this metabolic stress during pregnancy confers an increased risk of diabetes remains an unresolved question. To investigate the association between parity and the subsequent risk of diabetes, we performed a population-based study on a cohort consisting of all Danish women giving birth to a live singleton in 1982 and 1983. We linked these data to hospital discharge data, birth registries and the Danish civil registry to obtain information about parity, diabetes, vital statistics and emigration status. Correspondence to: Klara Vinsand Naver, MD, Department of Obstetrics and Gynaecology, Copenhagen University Hospital, Hvidovre, Kettega ˚ rd Alle 30, DK 2650 Hvidovre, Denmark. E-mail: kvnaver@gmail.com DIABETICMedicine DOI: 10.1111/j.1464-5491.2010.03169.x ª 2010 The Authors. Diabetic Medicine ª 2010 Diabetes UK 43