Lifestyle risk factors and all-cause and cause-specific mortality in the Mexico City prospective study: Assessing the influence of reverse causation

Abstract

Background: We examined the association between individual lifestyle risk factors with all-cause and cause-specific mortality.

Methods: Prospective cohort study including 155,002 participants from the Mexico City Prospective Study. Cox regression models were used to estimate the association between individual lifestyle risk factors and all-cause and cause-specific mortality. Participants with prevalent diseases at baseline and participants who died during the first 2, 5, 10, and 15 years of follow-up were excluded to account for reverse causation.

Results: 27,469 people died during 18.3 years of follow-up years. Overweight and moderate alcohol consumption were inversely associated with all-cause mortality, while low physical activity and smoking were positively associated when all participants were included, regardless of prevalent disease or duration of follow-up. The direction of the association of overweight with all-cause mortality changed from inverse to positive after excluding the first 10 years of follow-up. Compared with normal weight, the hazard ratio (95 % confidence interval) was 1.17 (1.13,1.22) for obesity after excluding those who died in the first 5 years of follow-up and 1.71 (1.59,1.84) after excluding the first 15 years of follow-up. The magnitude of the association of alcohol intake, low physical activity, and smoking with mortality attenuated, whereas for fruits and vegetables increased, after excluding longer periods of follow-up.

Limitations: The data were collected exclusively in Mexico City; lifestyle risk factors were self-reported and thus prone to misclassification bias.

Conclusions: Reverse causation may influence both the magnitude and the direction of the associations between lifestyle risk factors and mortality.

Keywords: Bias; Lifestyle risk factor; Mortality; Prospective study; Reverse causation.