Adiposity, adipose tissue distribution and mortality rates in the Canada Fitness Survey follow-up study

OBJECTIVE: To compare mortality rates across indicators of adiposity and relative adipose tissue distribution in the Canadian population. SUBJECTS: The sample included 10 323 adult participants 20-69y of age from the Canada Fitness Survey who were monitored for all-cause mortality over 13 y. METHODS: BMI, waist circumference (WC) and the sum of five skinfolds (SF5) were indicators of adiposity, and the first principal component of skinfold residuals (PC1) represented subcutaneous adipose tissue distribution. Proportional hazards regression was used to estimate relative mortality risk from mortality rates across levels of adiposity and adipose tissue distribution, controlling for the confounding effects of age, smoking status and alcohol consumption. RESULTS: Significant curvilinear (J-shaped) relationships in men and linear relationships in women were observed between BMI, WC and SF5 and all-cause mortality rates. PC1 was not related to mortality rates in either men or women. In women, the inclusion of the other indicators of adiposity and adipose tissue distribution did not significantly add to the prediction of mortality rates beyond BMI; however, combinations of BMI and both WC and SF5 produced significant models in men. CONCLUSION: The results support the hypothesis that overall level of adiposity is an important predictor of all-cause mortality, more so than the relative distribution of subcutaneous body fat, once overall level of body fatness has been accounted for.