Ethnic differences in mammographic density and their determinants: A cross-sectional study among Caucasian and Indian women


Session type:

Constantinos Savva1, Zoe Aitken2, Isabel dos Santos Silva2
1Norfolk and Norwich University Hospital, Norwich, UK, 2London School of Hygiene & Tropical Medicine, London, UK


Mammographic density (MD) is known to be associated with breast cancer (BC) risk, but it is not clear whether the ethnic differences in BC risk are reflected in differences in MD1. The aim of this study is to assess differences in MD between Caucasian and Indian women and investigate to what extent known breast cancer risk factors account for ethnic differences in MD.


Caucasian women attending routine screening in London were recruited from Bart’s and the London Breast Cancer Screening Centre and a sample of women from Indian population were recruited through a free BC screening initiative; filling in a questionnaire on anthropometric and reproductive factors.MD was measured using Volpara software, a fully-automated method quantifying the volume of dense breast tissue from digital mammograms2. Multivariate linear regression techniques were used to investigate the association between ethnicity and MD and the extent to which this was mediated by BC risk factors.


258 women were included in the analysis aged 50 to 70. There was strong evidence that the geometric means of the MD were higher in Caucasians than in Asians after adjustment for age and body mass index (BMI). Minimally adjusted analyses for age and BMI showed strong evidence of association between all measures of MD and ethnicity. Percent MD was higher in British (1.24;95%CI 1.21 to 1.26) as well as absolute dense (1.53;95%CI 1.34 to 1.73), non-dense (1.20;95%CI 1.04 to 1.38) and total volume (1.23;95%CI 1.07 to 1.41). These associations were attenuated on adjustment for known BC risk factors. Sequentially adjusted analyses revealed that these ethnic variations were mainly explained by the higher BMI and nulliparity, fewer births,shorter breastfeeding duration, increased use of hormone replacement therapy and higher education in Caucasians.


These findings suggest that the differences in MD between British and Indian women are accounted for anthropometric and reproductive factors, and education.