Purpose: In order to improve population-based approaches in communities to extend healthy life expectancy of our population in the 21st century, it is essential to identify characteristics and risks thoroughly. This study assessed associations of dietary intake and health behavior with mortality from cancer, cardiovascular diseases, and all-causes in Japan at the prefectural level.
Methods: By prefecture and sex, we calculated age-adjusted means for BMI, step counts, and nutrient and food intakes, as well as age-adjusted prevalence of smoking and alcohol drinking habits, using datasets of the 2001-05 National Health and Nutrition Survey. Age-adjusted total mortality rates (per 100,000 population), as well as cancer and cardiovascular mortality rates were obtained from Vital Statistics (2007). Multivariate stepwise regression analysis was used to compute partial correlation coefficients.
Results: In regard to BMI, mortality from myocardial infarction and cerebral hemorrhage showed significant positive correlations in both males and females, but significant negative correlations with mortality from stomach cancer in both sexes. The sodium chloride equivalent (salt) was positively correlated with mortality from cerebral hemorrhage in both males and females. In males, salt was also positively correlated with total mortality. In females, salt was also positively correlated with mortality from cerebral infarction and all types of stroke. Alcohol drinking was positively correlated with mortality from esophageal cancer in both genders and from cerebral infarction in males. Several other nutrients and food groups, as well as physical activity, were also associated with mortality risk.
Conclusions: Although this study is an ecological analysis, these findings highlight some factors of public health importance.