1. ¹Department of Obstetrics and Gynaecology, Medical Faculty, Division of Akershus University Hospital, 1478 Lørenskog, Norway
2. ²Department of Public Health, Faculty of Medicine, Norwegian University of Science and Technology, 7289 Trondheim, Norway
3. ³International Agency for Research on Cancer, 69372 Lyon, France
4. ⁴Division of Mental Health, Norwegian Institute of Public Health, PO Box 4404 Nydalen, 0403 Oslo, Norway

Correspondence: Dr K Lindemann, Department of Obstetrics and Gynaecology, Akershus University Hospital, Sykehusveien 27, 1478 Lørenskog, Norway. E-mail: kristina.lindemann@ahus.no

Received 2 January 2008; Revised 15 February 2008; Accepted 25 February 2008; Published online 25 March 2008.

Abstract

We examined the relationship of body mass index (BMI), diabetes and smoking to endometrial cancer risk in a cohort of 36 761 Norwegian women during 15.7 years of follow-up. In multivariable analyses of 222 incident cases of endometrial cancer, identified by linkage to the Norwegian Cancer Registry, there was a strong increase in risk with increasing BMI (P-trend <0.001). Compared to the reference (BMI 20–24 kg m^-2), the adjusted relative risk (RR) was 0.53 (95% confidence interval (CI): 0.19–1.47) for BMI<20 kg m^-2, 4.28 (95% CI: 2.58–7.09) for BMI of 35–39 kg m^-2 and 6.36 (95% CI: 3.08–13.16) for BMI40 kg m^-2. Women with known diabetes at baseline were at three-fold higher risk (RR 3.13, 95% CI: 1.92–5.11) than those without diabetes; women who reported current smoking at baseline were at reduced risk compared to never smokers (RR 0.55, 95% CI: 0.35–0.86). The strong linear positive association of BMI with endometrial cancer risk and a strongly increased risk among women with diabetes suggest that any increase in body mass in the female population will increase endometrial cancer incidence.