Significantly greater reduction in breast cancer mortality from post-diagnosis running than walking.

The purpose of these analyses is to test prospectively whether post-diagnosis running and walking differ significantly in their association with breast cancer mortality. Cox proportional hazard analyses were used to compare breast cancer mortality to baseline exercise energy expenditure (METs, 1 MET-hour ≅1 km run) in 272 runners and 714 walkers previously diagnosed with breast cancer from the National Runners' and Walkers' Health Studies when adjusted for age, race, menopause, family history, breastfeeding and oral contraceptive use. Diagnosis occurred (mean ± SD) 7.9 ± 7.3 years before baseline. Forty-six women (13 runners and 33 walkers) died from breast cancer during 9.1-year mortality surveillance. For the 986 runners and walkers combined, breast cancer mortality decreased an average of 23.9% MET-hours/day [95% confidence interval (CI): 7.9-38.3%; p = 0.004]. There was a significantly greater decrease in risk for running than walking (risk per MET-hours/day run vs. walked: p = 0.03). For the 272 runners analyzed separately, breast cancer mortality decreased an average of 40.9% per MET-hours/day run (95% CI: 19.3-60.0%, p = 0.0004). When analyzed by categories of running energy expenditure, breast cancer mortality was 87.4% lower for the 1.8-3.6 MET-hours/day category (95% CI: 41.3-98.2% lower, p = 0.008) and 95.4% lower for the ≥3.6 MET-hours/day category (95% CI: 71.9-100% lower, p = 0.0004) compared to the <1.07 MET-hours/day category. In contrast, the 714 walkers showed a nonsignificant 4.6% decrease in breast cancer mortality per MET-hours/day walked (95% CI: 27.3% decreased risk to 21.3% increased risk, p = 0.71). These results suggest that post-diagnosis running is associated with significantly lower breast cancer mortality than post-diagnosis walking.