Abstract
OBJECTIVE: There is limited evidence on the association between lifestyle factors and breast cancer (BC) incidence from Australian longitudinal studies. This study aims to investigate the influence of lifestyle factors on BC incidence over time among Australian women from mid-life to older age. DESIGN: Longitudinal study. SETTING: Data from the Australian Longitudinal Study on Women's Health (ALSWH) and linked Australian Cancer Database (ACD). PARTICIPANTS: 12 782 women from the ALSWH 1946-1951 birth cohort linked with the ACD from 1996 to 2019. MAIN OUTCOME AND MEASURES: Time to the occurrence of BC, accounting for death as a competing event. Body mass index, alcohol consumption, smoking, marital status, oral contraception and hormone replacement therapy were considered as lifestyle factors due to their impact from mid-life to older age. RESULTS: Among 12 782 women in the cohort, a total of 941 incident BC cases (7.4%) were identified between 1996 and 2019. Time-dependent analysis disclosed that a higher hazard of BC in alcohol drinkers (rarely drinks/low-risk drinkers: Subdistribution HR [sHR]=1.49, 95% CI: 1.33-1.69; risky/high-risk drinkers: sHR=1.36, 95% CI: 1.14-1.62) relative to non-drinkers and those with overweight/obesity (sHR=1.23, 95% CI: 1.14-1.32) relative to underweight/acceptable weight. Results also revealed that non-partnered women (sHR=1.32, 95% CI: 1.12-1.57) had a higher hazard of BC than those with partners. Models were adjusted for lifestyle, reproductive and demographic factors. The probability of cumulative incidence of BC for alcohol drinkers and overweight/obese women steadily increased over time. CONCLUSION: This study demonstrated that being non-partnered, overweight/obese and consuming alcohol were associated with increased hazards of BC in women's mid-life to older age.