Figure 1.

Hazard ratios and 95% confidence intervals for the association of recreational physical activity and all-cause mortality after first invasive breast cancer in women with and without BRCA1/2 pathogenic variants in the Prospective Family Study Cohort (n = 4610). Hazard ratios and 95% confidence intervals (given within parentheses and depicted by error bars) are adjusted for age group at diagnosis (younger than 40, 40-49, 50-59, 60 years or older), study center (Australia, Canada, kConFab, New York, Northern California, Philadelphia, Utah), decade of birth year (<1950, 1950-1959, 1960-1969, ≥1970), race and ethnicity (Asian, Hispanic, non-Hispanic Black, non-Hispanic White, other), education (high school graduate or less, some college, Bachelor degree or higher), cigarette smoking (never, former, current), alcohol consumption (never, former, current), postmenopausal hormone therapy use (never vs otherwise), body mass index (<25, 25 to <30, ≥30 kg/m²), stage at diagnosis (1, 2, 3 or 4, unknown), chemotherapy for first invasive breast cancer (yes vs otherwise), radiation treatment for first invasive breast cancer (yes vs otherwise), and surgery for first invasive breast cancer (mastectomy, lumpectomy, otherwise). Women without BRCA1 or BRCA2 pathogenic variants (PVs) include true negatives (n = 3137) and women without a genetic test result (n = 1099). The interaction term between recreational physical activity categorized as any vs none and BRCA1/2 PV status was statistically significant (2-sided Wald test P = .005). The interaction term between recreational physical activity categorized into quintiles of total metabolic equivalents hours per week (MET-hours/week) and BRCA1/2 PV status was also statistically significant (2-sided Wald test P = .007). CI = confidence interval; HR = hazard ratio.