RATIONALE: Weight gain is recognized as an important factor in the decline of forced expiratory volume in one second (FEVJ This study investigates the relationship between FEV(1) body mass index (BMI), and body measurements reflecting the distribution of weight in working age males. METHODS: 2745 Northern Irish men were recruited to participate in the Prospective Epidemiological Study of Myocardial Infarction (PRIME) from 1991 to 1994; 2010 were re-studied at 10 years. Demographic data, BMI, waist and hip circumference, and FEV(1), were determined at both surveys. This analysis includes the 930 men with two valid spirometry measurements. FEV(1) values at baseline and at 10 years were modelled using simple linear regression, and adjusted for age, height, alcohol intake, physical activity, educational group, and smoking history. RESULTS: Mean age at baseline was 54 years (SD 2.9). Over the 10 years, mean BMI went from 26.2 to 27.6 kg/m2 and waist circumference from 90.9 to 95.6 cm. Linear models of initial FEV(1) showed an inverse relationship with waist circumference but the association with BMI was not significant. FEV(1) at 10 yrs was reduced by 19 mls for each kg/m2 increase in BMI (p=0.041, r2=0.296). When change in waist circumference was included, FEV(1) at 10 yrs was reduced by 11 mls for each centimetre increase in waist circumference (p<0.001, r2=0.307), but BMI change was no longer significant (p=0.584). Baseline waist-tn-hip ratio was also a significant predictor of FEV(1) at 10 yrs (p<0.001). Linear models of FEV(1) better fit the data when waist and hip measurements were included, compared to models with BMI alone. CONCLUSIONS: FEV(1) appears related to the distribution of weight. In addition to weight, collection of measurements such as waist circumference and waist-hip ratios may improve understanding of changes in lung function.