In 1998, the German Environmental Survey (GerES III) recruited approximately 5000 adults between the ages of 18 and 69 years. The study population for these analyses consisted of 1580 smokers (34% of the total population) and 3126 nonsmokers. Nicotine and cotinine concentrations in urine were determined by HPLC methods with UV-detection and corrected for creatinine. Nicotine and cotinine concentrations differed between smokers and nonsmokers by factors of 10-100. The multiple linear regression models used for the analyses of nicotine detection in the urine of smokers explained 43.2% and 42.3% of the total volume-specific and creatinine-specific variances, respectively. Cigarette smoking was the major factor responsible for 41% of the total variance. The explained variances of the cotinine results were larger, 51.0% and 49.3% of the total variance were volume-specific and creatinine-specific, respectively. More than 20% of nonsmokers in GerES III were exposed to environmental tobacco smoke at home, at work or in other places. The logistic regression analysis approach used for the group of nonsmokers showed the greatest effects for those exposed to tobacco smoke at home (adjusted OR varied between 4 and 6). These results were seen for nicotine as well as for cotinine excretion. Exposure to tobacco smoke in the workplace doubled the risk for the detection of nicotine and cotinine in urine. When other risk factors such as age, sex, social status, community size, season of urine collection, and the consumption of food containing nicotine such as potatoes, cabbage, tea were included, the effect estimates for tobacco smoke exposure remained unchanged. A new federal bill to diminish environmental tobacco smoke (ETS) exposure in the workplace was recently passed in Germany, but protection of nonsmokers from smoking family members at home needs more attention.