BACKGROUND: Passive smoking has been found to be a respiratory health hazard in humans. The present study describes the calculation of a reference interval for urinary nicotine metabolites calculated as cotinine equivalents on the basis of 72 non-smokers exposed to tobacco smoke less than 25% of the day. METHODS: Twenty subjects (passive smokers) exposed to tobacco smoke more than 25% of the day (subjectively assessed) and 32 smokers were used to validate the estimated reference interval. Urine samples were collected three times during the day approximately at 06.30, 17.00 and 22.45 h. RESULTS: Within-subject variation was found to be 89.4, 72.6, and 79.2% and between-subject variation was found to be 64.5, 64.2, and 36.1%. No gender difference could be demonstrated. In general all subjects showed increased concentrations in the afternoon and evening samples compared to the morning samples. Parametric reference interval for excretion of nicotine metabolites in urine from non-smokers was established according to International Union of Pure and Applied Chemistry (IUPAC) and International Federation for Clinical Chemistry (IFCC) for use of risk assessment of exposure to tobacco smoke. The reference interval for urinary cotinine was estimated to be 1.1-90.0 micromol/mol creatinine in morning samples from non-smokers. An intercomparison between the radioimmunoassay (RIA) method used for determination of nicotine metabolites and a gas chromatography-mass spectrometry (GC-MS) method for determination of cotinine was carried out on 27 samples from non-smokers and smokers. Results obtained from the RIA method showed 2.84 [confidence interval (CI): 2.50; 3.18] times higher results compared to the GC-MS method. A linear correlation between the two methods was demonstrated (rho=0.96). CONCLUSION: The RIA method is rapid and adequate for clinical use in the assessment of exposure to tobacco smoke, i.e. ratio between CV(a)/CV(ti) was<0.50.