Past recreational physical activity and risk of breast cancer

To examine the association between baseline and earlier recreational physical activity and incidence of breast cancer in postmenopausal women.

Multicenter cohort study.

Women were enrolled in the Women's Health Initiative Observational Study (WHIOS) at 40 clinical centers between October 1993 and December 1998.

Of a total of 93676 women enrolled in the WHIOS, 74171 were included in this analysis. Eligibility criteria were: aged 50 to 79 years, postmenopausal, and free of serious health conditions that might reduce survival during the following 3 years. Women were excluded if they reported a history of breast cancer or had missing physical activity data. Women of non-European extraction made up 15% of the sample. By February 28, 2002, 3.2% were lost to follow-up and 2.7% had died.

Participants completed baseline questionnaires that included information on medical and reproductive history, and health behaviors including physical activity and diet. Staff collected anthropometric data and information on use of hormone therapy. Women were asked if they usually exercised enough to work up a sweat > or =3 times per week at ages 18, 35, and 50 years. Current (baseline) walking for > or =10 minutes and participation in leisure-time activities were categorized by frequency, duration, and intensity. A current total activity variable was computed from the product of metabolic equivalent values and duration (MET h/wk).

The main outcome measure was the association of incident breast cancer with measures of physical activity during a mean period of follow-up of 4.7 years. Cases of breast cancer were ascertained by the methods of the National Cancer Institute Surveillance, Epidemiology, and End Results guidelines, and collated by physicians and cancer coders blinded to exposure data.

During follow-up, 1780 incident cases of breast cancer were documented. Women who had engaged in strenuous physical activity > or =3 times per week at age 35 had a decreased risk of breast cancer (multivariate adjusted relative risk [RR], 0.86; 95% CI, 0.78-0.95) compared with women who had not. The association was not significant for strenuous activity at age 50 or age 18 years. A greater amount of total current (baseline) physical activity was associated with a lower risk of breast cancer (P for trend, 0.03). Compared with no current physical activity, risk was reduced 18%, 11%, 17%, and 22% for women who exercised 5.1-10, 10.1-20, 20.1-40, and >40 MET h/wk. Hours of current moderate or strenuous physical activity was not significantly related to risk of breast cancer (P for trend, 0.12), although the highest duration category, >7 h/wk, compared with 0 h/wk showed a significant reduction in risk (RR, 0.79; CI, 0.63-0.99). When the women were divided by tertiles of body mass index (BMI cut points at 24.13 and 28.44 kg/m), increased total current activity was protective against breast cancer for those in the lowest tertile of BMI (P for trend, 0.03) but not for those in the upper tertiles (P for trend, 0.74 and 0.30).

More physical activity was associated with a lower risk of breast cancer in postmenopausal American women. An hour every day of moderate or strenuous activity provided most benefit.