Weight status and breast cancer incidence in the UK Women's Cohort Study: a survival analysis

• Published in: The Lancet (British edition) 2014-11, Vol.384, p.S53-S53
• Main Authors: Morris, Michelle A, Dr, Hulme, Claire, Prof, Clarke, Graham P, Prof, Edwards, Kimberley L, PhD, Cade, Janet E, Prof
• Format: Article
• Language: English
• Subjects: Breast cancer; Internal Medicine; Medical research; Menopause; Obesity
• ISSN: 0140-6736; 1474-547X
• DOI: 10.1016/S0140-6736(14)62179-6

Abstract Background Breast cancer is the most common type of cancer in the UK, even though it predominantly affects women. Some research suggests that obesity before menopause can be protective against breast cancer, whereas postmenopausal obesity poses increased risk. Because of a period of latency in the development of cancer, longitudinal cohort studies are essential. The aim of this research was to investigate the association between body-mass index (BMI) and breast cancer incidence. Methods The UK Women's Cohort Study was established in the 1990s to investigate patterns between diet and health. 35 000 women were recruited and followed up via cancer incidence and mortality data reports from the National Health Service information centre. Self-reported height and weight were used to calculate BMI. WHO cut-off points were used to define BMI categories. Age-adjusted Cox proportional hazards regression was applied with Stata (version IC12). Findings 1445 (4%) of 35 372 women developed breast cancer during a median time to follow-up of 14·9 years (IQR 1·4). Cases prevalent at baseline were excluded. There were significant differences between the observed and expected outcomes by BMI category (p=0·0003, log-rank test), with underweight and normal weight having lower observed than expected incidence. Overweight and obese women had higher observed than expected incidence. Test for trend of survivor functions showed a significant trend of increasing breast cancer incidence with increasing BMI for the whole cohort (p=0·0001) and the postmenopausal subgroup (p=0·0001), but not for premenopausal women. Hazard ratios in an age-adjusted model were significant in postmenopausal women: underweight 1·0 (reference); normal weight 1·6, 95% CI 0·8–3·2; overweight 2·0, 1·0–4·1; obese 2·1, 1·0–4·3. Interpretation We found a clear association between BMI and breast cancer incidence, especially in postmenopausal women. We did not see an association in premenopausal women. These results are supportive of existing scientific literature and highlight the importance of maintaining a healthy weight in adulthood, especially after the menopause. Health-care professionals should consider referral of overweight and obese women to weight loss services as a protective measure before menopause. Funding This research was funded as part of an Economic and Social Research Council/Medical Research Council interdisciplinary PhD studentship.