To determine whether diabetes is associated with reduced lung function, we studied 421 Anglo-Celt/European subjects, representing 20.5% of all patients with type 2 diabetes identified in an urban Australian catchment area of 120097 people. In addition to collection of detailed demographic and diabetes-specific data, spirometry was performed and forced vital capacity (FVC), forced expiratory volume in 1 s (FEV1), vital capacity (VC) and peak expiratory flow (PEF) measured. When expressed as a percentage of those predicted (%pred) for age, sex and height, the means of all spirometric measures were reduced by > or =9.5%. After controlling for smoking, age and gender in a linear regression model, HbA(1c) was not associated with any measure of lung function (P>0.13) but diabetes duration was significantly associated with FEV1(%pred) and PEF(%pred) (P< or =0.04) and had borderline associations with FVC(%pred) and VC(%pred) (P< or =0.064). In separate analyses controlling for smoking alone, age, body mass index (BMI), coronary heart disease (CHD) and retinopathy were independently and inversely associated with FVC(%pred), FEV1(%pred) and VC(%pred) (P<0.05). In sub-group analyses, these three spirometric measures were associated with BMI, CHD and diabetes duration in males, and age and BMI in females. Pulmonary function is reduced in type 2 diabetes. Diabetes duration seems a more important influence than glycaemic control, but obesity and vascular disease may also contribute.