Data AnalysisData from the subjects mentioned above were available for sex, age, smoking status, BI, FVC and FEV1, but did not include any personal identifiers such as name or address. We counted personyear of follow-up for each subject from the date of the baseline of the first health check-up to incidence of COPD or the last health check-up in the study period. The date of incidence of COPD was determined as a median date between the health check-ups with the first diagnosis of COPD and with the last diagnosis of not having COPD. Those who were not diagnosed as COPD during the follow-up period were treated as censored cases.The incidence rate of COPD by sex, age, smoking status, and follow-up period was calculated as the number of COPD incidence cases divided by the person-years of follow-up. The age was classified into 10 groups of 5-year intervals from 25-29 to 70-74 years old. The follow-up period was classified into less than 2 years and 2 years or more after baseline. The incidence rate ratios (IRRs) of age groups to 40-44 years old and those of smoking status to non-smokers by sex were estimated using Cox proportional hazard models with both variables. These analyses were performed using an SPSS ® 12.0J software package (SPSS Japan Inc.).

BACKGROUND: The Global Initiative for Chronic Obstructive Lung Disease guidelines recommended a forced expiratory volume at one second per forced vital capacity as a standard diagnostic criterion of chronic obstructive pulmonary disease (COPD). A few reports on the risk factors of COPD have used the standard diagnostic criteria. In our study, the effects of age and smoking on COPD in Japan under the standard diagnosis criteria were evaluated. METHODS: Subjects were 11,460 participants aged 25-74 years during health check-ups including spirometry at the Toyota Community Medical Center in Japan. Logistic regression analyses with or without COPD as a dependent variable and age as an independent variable were conducted among non-smokers. The ratio of the observed number of COPD cases in former and current smokers to the number expected for non-smokers with the same distribution of age (O/E) was calculated. RESULTS: The proportion of males incurring COPD significantly increased with age, and the O/E for former and current male smokers was significantly higher than one, i.e., O/E (95% confidence interval) for current smokers with a Brinkman Index of <400, 400-799, and 800+ were 3.10 (2.00-4.81), 2.78 (2.05-3.73), and 4.76 (3.65-6.19), respectively. Among females, the O/E for current smokers with a Brinkman Index of <400, and 400-799 were significantly higher than one. CONCLUSION: Age and smoking were shown to constitute strong risk factors for COPD under the standard diagnostic criteria.