Association between body mass index, diabetes and breast cancer incidence: a population-based cohort study

Background: We aimed to evaluate if higher body mass index (BMI) and diabetes are associated with higher post-menopausal breast cancer (BC) incidence in a population-based cohort with a long follow-up period. Methods: The Singapore Chinese Health Study is a prospective cohort investigating impact of diet, genetics and environmental factors on cancer. The cohort was established in 1993-1998 through recruitment of 63,257 ethnic Chinese in Singapore, aged 45-74 years at baseline. A questionnaire was administered in-person to collect demographic, reproductive, comorbidity, diet and lifestyle factors. In this study, female participants who reported being menopausal and without cancer at baseline, had valid BMI readings and >1 year of follow-up were included. Breast cancer cases were identified via record linkage with National Registry of Diseases Office. A cox proportional hazard model was used to analyze BC risk after adjusting for baseline age, year of interview, educational level, number of live births, age at menarche, BMI and self-reported diabetes at baseline. Findings: Out of 19,271 post-menopausal women, 582 developed BC (168.70/100,000 person-years) over a mean follow-up time of 17.90 years. Hazard ratio (HR) (95% confidence interval [95% CI]) of BC incidence in women with BMI ≥28, 24 to <28 and 20 to <24 vs women with BMI <20 was 2.49 (1.78 – 3.48) (p < 0.0001), 1.98 (1.48 – 2.65) (p < 0.0001) and 1.80 (1.36 – 2.37) (p < 0.0001) respectively. Higher BMI was also significantly associated with BC incidence in participants without self-reported diabetes after adjustment for other aforementioned risk factors. Interpretation: BMI is a significant independent risk factor for post-menopausal BC in Chinese populations, even in those with BMI readings of 20-24 without self-reported diabetes at baseline. This study’s strength lies in a large number of participants but the small number with diabetes, which was self-reported at baseline, is a limitation. Further research into how body fat influences BC risk even in individuals with normal BMI is warranted. Funding: Nil.