Estimating cancer incidence attributable to physical inactivity in the United States.

Abstract

Previous estimates of the number of cancers attributable to physical inactivity in the United States have typically focused on only three malignancies (colon, endometrial, and postmenopausal breast cancer). Contemporary epidemiologic evidence suggests that physical inactivity could contribute to up to 15 types of cancer, and a dose-response effect has been demonstrated for 13 of these. This study estimated the number of cancers diagnosed in the United States in 2015 due to physical inactivity for these 13 sites. Data from the 2005 National Health Interview Survey were used to estimate physical activity prevalence and, with the assumption of a 10-year latency period, 2015 cancer incidence data from the National Program of Cancer Registries and Surveillance, Epidemiology, and End Results Incidence US Cancer Statistics Public Use Database. The potential impact fraction was estimated to be 4.1%, which meant that 30,951 of 761,625 incident cancers at the 13 sites could have been prevented in the United States in 2015 if adults had increased physical activity by one category in 2005 (approximately 7.5 additional metabolic equivalent task hours per week [MET-h/week]). Theoretically, 85,415 of 761,625 incident cancers at the 13 sites (population attributable fraction, 11.2%) could have been prevented if all adults had achieved the highest level of physical activity (>30 MET-h/week). When estimates are based on updated epidemiologic evidence regarding physical inactivity and cancer risk, substantially more cancers are attributable to physical inactivity than previously reported. A greater focus on physical activity promotion is warranted for cancer control in the United States.