Purpose: We aimed to investigate associations of self-rated health with fruit and vegetable consumption (FVC) and physical activity (PA) among older cancer survivors.
Methods: We used the 2017 Behavioral Risk Factor Surveillance System to identify cancer survivors ≥ 65 years (N = 2663). Self-reported FVC and PA were categorized as ordinal variables to approximate quartiles. Low general health (LGH) was defined as fair or poor self-rated health. A multivariable logistic regression treating LGH as the outcome was used to calculate adjusted odd ratios (aORs) and 95% confidence intervals (CIs) for FVC and PA. Restricted cubic spline depicted non-linear dose-response curves for FVC and PA. In comparative analysis, we used the same logistic regression and dose-response model to calculate ORs of FVC and PA in 73,134 people ≥ 65 years without cancer history.
Results: Overall, 470 (17.7%) survivors had LGH. Survivors' mean age was 73.3 years (SD = 5.2), 55.1% of them were female, and 95.4% self-reported as white. In cancer survivors, FVC was not associated with LGH (≥ 28 vs. < 14 times/week: aOR = 1.02, 95% CI = 0.75-1.39, p-trend = 0.50), whereas PA was inversely associated with LGH (≥ 30 vs. < 7 MET-hours/week: aOR = 0.55, 95% CI = 0.41-0.75, p-trend < 0.01). Dose-response curves demonstrated consistent association patterns. In comparative analysis, ORs of PA did not change substantially but we observed inverse association for FVC.
Conclusions: An inverse association between PA and LGH was observed among older cancer survivors, but no significant association was obtained for FVC among them. Regular PA may maintain or indicate a favorable health in older cancer survivors, whereas impacts of FVC deserve further investigations.
Keywords: Cancer survivorship; Epidemiology; Fruit and vegetable; Gerontology; Physical activity.