The objective of the present study was to analyse the association between cigarette smoking and progression of retinopathy and nephropathy, respectively, in a prospective multicentre study including 636 people with Type 1 diabetes: 81% of the original cohort of consecutively referred patients, aged 15 to 40 years and free of severe late diabetic complications. At baseline, all patients had participated in a 5-day in-patient group treatment and teaching programme for intensification of insulin therapy. Patients were examined at recruitment, and after 1, 2, 3 and 6 years including assessment of smoking status, blood pressure, metabolic control, and degree of nephropathy. Degree of retinopathy was assessed by ophthalmoscopy or fundus photography at baseline and after 6 years. Several logistic regression analyses were performed by describing the responses retinopathy and nephropathy, respectively, either as progression yes/no or as actual status at the 6-year follow-up and by using different measures for smoking. Adjustments for important covariables were made. While significant associations between smoking, and retinopathy and nephropathy respectively, were found, the relations were variable depending on the statistical model used. The results show that the real associations between smoking and retinopathy and nephropathy are complex and that more emphasis should be put on the complete description of the response variables and the statistical models used in clinical and epidemiological research.