Abstract
ABSTRACT
Physical activity is associated with lower risk of liver cancer, but obtaining an unbiased estimate of the causal effect is challenging due to the competing risk of death due to any other cause. We performed a causal competing risk analysis to estimate the total effect of physical activity on risk of liver cancer in the presence of competing deaths. Data were included for 263,184 participants (870 liver cancers) in the NIH‐AARP Diet and Health Study. Total effects in the presence of competing deaths were estimated on the risk difference (RD) scale as the differences, by age, in standardised cause‐specific cumulative incidence functions of liver cancer had all participants engaged in ≥ 4 h/week of moderate‐ to vigorous‐intensity physical activity (MVPA) versus < 4 h/week. There were an estimated 18 per 100,000 fewer liver cancer cases for ≥ 4 h/week of MVPA versus < 4 h/week (95% confidence interval [CI]: 6.2, 29.9) by age 67 years (average age of diagnosis in the United States), and an estimated 58 per 100,000 fewer cases (95% CI: 21.1, 95.8) by age 78 years (average US life expectancy). Estimated RDs were larger in older than younger age groups (e.g., RD per 100,000 at age 85 years = −79.0, 95% CI: −133.4, −24.7; at age 55 years = −2.0, 95% CI: −4.5, 0.6). Our findings strengthen the evidence for physical activity reducing risk of liver cancer. Increasing physical activity at the population level may reduce the growing, global burden of liver cancer.