Short-term and long-term ambient NO2 exposure, dietary effect modification, and incident cardiovascular disease in rural Japan: a population-based cohort study.

OBJECTIVES: The contributions of short- and long-term nitrogen dioxide (NO2) exposure to cardiovascular disease (CVD) in rural areas remain unclear. Dietary intake may modify these associations, but this role remains understudied. METHODS: We conducted a secondary analysis of a population-based prospective cohort study in 12 rural Japanese municipalities. Adults without CVD enrolled in 1992-1995 were followed until 2005. Daily municipality-level NO2 concentrations were estimated using generalized random forests. The primary outcome was time to first incident CVD (stroke or myocardial infarction). Short-term effects were estimated using extended Cox regression models with daily NO2 exposure, adjusted for confounders and CVD predictors. Long-term effects were estimated using conventional Cox regression models with average follow-up NO2 exposure. Interactions between NO2 exposure and baseline food intake were evaluated. RESULTS: Among 10,032 individuals with complete data (116,759 person-years), 458 incident CVD cases were identified (390 strokes and 68 myocardial infarctions). In the short-term exposure analysis, each 10-ppb increase in daily NO2 exposure was associated with increased CVD risk (hazard ratio [HR], 1.26; 95% confidence interval [CI], 1.05 to 1.51; p=0.01). In the long-term analysis, each 1-ppb increase in average NO2 exposure during follow-up was associated with increased CVD risk (HR, 3.18; 95% CI, 1.54 to 6.59). For interactions between NO2 exposure and dietary intake, no associations remained significant after controlling for multiple comparisons using the false discovery rate. CONCLUSION: This study suggests that short- and long-term NO2 exposure may contribute to CVD risk. Long-term average exposure showed a larger effect estimate than short-term exposure.