An epidemiologic investigation of physical activity and breast cancer risk in Africa.

Abstract

Physical activity (PA) is modifiable and linked to decreased breast cancer risk but its impact has not been investigated among indigenous African populations. From 2011 to 2013, 558 cases and 1,014 controls were recruited into the African Breast Cancer Study in Nigeria, Cameroon, and Uganda, and completed a culturally tailored PA questionnaire that assesses habitual PA the year before diagnosis/interview. PA sub-scores (housework, occupational, and leisure PA) and a total PA score were calculated (metabolic equivalent of task, MET-hours/day). Multiple logistic regressions were performed, adjusting for age, body mass index (BMI), study sites, and menopausal status. The models were then stratified by BMI and study site, respectively. The overall PA score among controls (17.8 MET-hours/day on average) was mainly composed by housework PA and occupational PA with little leisure PA (7.0, 10.3, and 0.5 MET-hours/day, respectively). Multivariable analyses showed that PA was significantly associated with reduced breast cancer risk in both pre- and postmenopausal women (up to 60% risk reduction), with a dose-responsive relationship (Ptrend < 0.001). The inverse association was strong among lean women, less strong but still significant among overweight women, but not existing among obese women. The inverse association held for all intensity-level and domains of PA. PA of African women mainly consists of housework and work-related activities. The preliminary data show that PA may be significantly associated with reduced breast cancer risk. An inverse association between PA and breast cancer risk was observed among indigenous African women, a unique and understudied population.