Abstract
PURPOSE: To confirm that a multi-component physical activity index (PAI), with physical activity (PA)-related behaviors and performance measures, enhances PA assessment for adults with cancer history ("survivors") vs. for the general population for key health variables. METHODS: Data from 2011 to 2014 NHANES were analyzed (n = 9620 adults, including 866 survivors). We generated PAI scores by calculating subscales for activity-related behaviors (i.e., moderate-to-vigorous PA and TV viewing hours/sedentary time) (n/8) and performance (i.e., cardiorespiratory fitness and muscle strength) (n/8), then combined for an overall PAI score (n/16 points). Dependent variables (self-reported health, multimorbidity, and functional limitations) were dichotomized and tested with logistic regression to estimate associations with PAI variables. RESULTS: Survivors obtaining 22.5 + MET hours/week of PA were >5× more likely to report excellent health than non-exercisers (OR = 5.5, 95% CI 3.27-9.28). We observed a general decrease in likelihood of multimorbidity and functional limitations with increasing PA. Models with the combined score showed that higher PAI scores had larger positive associations with all three for all adults. Survivors with higher PAI scores were >30% more likely to report excellent self-rated health (OR = 1.3, 95% CI 1.2-1.4) and 20% less likely to report multimorbidity and functional limitations compared to survivors with lower PAI scores, considering covariates. CONCLUSIONS: Greater PA-related behavior and performance was beneficial for adults irrespective of cancer history. However, using a combined score elucidates unique needs and benefits for survivors vs. for general population. IMPLICATIONS FOR CANCER SURVIVORS: Using a multi-component approach to PA assessment could help in developing validated tools to plan exercise programs and interventions for survivors.