Lifestyle factors and all-cause mortality in long-term cancer survivors: a population-based prospective cohort study

The association between healthy lifestyles and mortality in cancer survivors remains inconclusive with few evidence among long-term cancer survivors (LTCS, survived ≥ 5 years post-diagnosis). Our study aims to investigate the association between individual and combined healthy lifestyle factors and mortality in LTCS. We included 6,057 LTCS of breast, colorectal or prostate cancer from a multiple regions study in Germany. A healthy lifestyle score (HLS) comprising alcohol consumption, body mass index (BMI), physical activity and smoking was created and was classified into tertiles with higher tertile indicating healthier lifestyle. We used Cox proportional hazards regression to examine the associations of individual lifestyle factors and HLS with all-cause mortality among LTCS. A total of 2,015 death events occurred over a maximum follow-up period of 12.3 years. Compared with the lowest tertile, participants in the middle and highest tertile experienced a 27% and 32% lower mortality (middle [hazard ratio (HR), 0.73; 95% CI 0.65-0.83]; highest [HR, 0.68, 95% CI 0.61-0.76]). A significant dose-response relationship was observed (p_{- trend} < 0.001). These associations were consistent across different demographic and clinical characteristics. In addition, full adherence to lifestyle recommendations for smoking (HR, 0.51, 95% CI 0.44-0.59), physical activity (HR, 0.78, 95% CI 0.70-0.86) and BMI (HR, 0.87, 95% CI 0.77-0.99) were significantly related to a lower mortality, after full adjustment. Adherence to an overall healthy lifestyle was associated with significantly lower all-cause mortality in LTCS, emphasizing the importance of maintaining and promoting a healthier lifestyle among LTCS.