Nested case-control study on associations between lung function, smoking and mortality in Japanese population

Objectives Only a few long-term follow-up studies with a focus on the association between lung function and mortality in the Japanese population have been undertaken. In this study, we examined the associations of lung function, smoking and the results of allergy skin tests with mortality in a longitudinal study of the Japanese population. Methods Baseline measurements were performed on residents of Fukui, Japan in 1972, and a follow-up survey was conducted in 2002. By employing a nested case-control design, 596 cases (deaths) and 596 age and sex-matched controls (survivals) were selected. Lung function was assessed using forced vital capacity (FVC) expressed as the normal percent predicted (FVC %pred) and the ratio of forced expiratory volume in 1 second (FEV₁) to FVC (FEV₁/FVC). Allergy skin tests were performed with extracts of house dust, candidia and mixed fungal samples (bronchomycosis). The Brinkman index was used to assess smoking intensity. The Cox proportional hazards model was used to evaluate whether lung function was associated with mortality after adjustment for other potential confounding variables. Results Those categorized into the first- or second-lowest quartile of FVC %pred had a higher mortality [hazard ratios (HRs) and 95% confidence intervals (CIs): 2.01 (1.26–3.19) and 1.84 (1.11–3.05)], respectively. On top of these, heavy smoking (BI≥400) was associated with a higher mortality [HR and 95% CI: 1.73 (1.18–2.53)]. There were only weak of associations between the results of allergy skin tests and mortality. Conclusions These results suggest that FVC %pred of lung function and smoking can serve as long-term independent predictors of mortality.