Association between Physical Activity and Reduced Mortality in Inoperable Lung Cancer

Abstract

We examined device-measured physical activity (PA) and sedentary behaviour at the time of diagnosis in people with inoperable lung cancer and investigated their associations with 12-month mortality. The people with inoperable lung cancer wore an accelerometer for seven days prior to the treatment commencement. The analysed PA/sedentary behaviour variables included light-intensity PA, moderate-to-vigorous-intensity PA (MVPA), step count, the total time spent sedentary, and the usual sedentary bout duration. The data on the disease stage, clinical covariates and 12-month mortality were extracted from medical records. Cox regression models were used to estimate the association between the PA measures and 12-month mortality, and the sedentary behaviour measures and 12-month mortality. The models were adjusted for the stage and neutrophil-to-lymphocyte ratio. All the PA and sedentary behaviour variables were dichotomised at their medians for analysis. Eighty-nine participants (70 ± 10 years; 55 [62%] males) contributed valid data. The twelve-month mortality was 30% (n = 27). Compared to the participants who spent ≤4.6 min/day in MVPA (n = 45), those who spent >4.6 min/day (n = 44) had a relative risk of 12-month mortality reduced by 60% (hazard ratio, 0.40; 95% CI, 0.16 to 0.96; 18 versus nine deaths, respectively). The other variables of PA/sedentary behaviour were not associated with 12-month mortality. Higher device-measured MVPA was associated with reduced 12-month mortality in people who were newly diagnosed with inoperable lung cancer.