Smoking among medical students has been found to vary strongly between European countries. Few studies have addressed factors associated with smoking among medical students within countries. In this study, we assessed the association of parental smoking and sociodemographic factors with smoking habits of medical students at the University of Ulm, Germany, Students who entered the 1st, 3rd and 5th year of medical school in fall 1992 were asked to complete a self-administered questionnaire. Questionnaires were returned by 696 (85.2%) of 817 eligible students. Overall, 23.7% of students were current smokers, and 11.9% were former smokers. Smoking habits were related to maternal smoking: Odds ratios for the association of maternal smoking with ever or current smoking of students were 2.11 (95% CI: 1.48-3.03) and 2.01 (95% CI: 1.35-3.01), respectively, after adjustment for potential confounders in multiple logistic regression. In contrast, no association was found between paternal smoking and students' smoking status. Male students were more likely to smoke than female students, and living in a large city during secondary school was also associated with ever smoking. No association was found between students' smoking habits and educational achievement of their mothers and fathers. These results suggest a key role of maternal smoking for smoking among medical students in this society.

Physicians can play a key role in promoting abstinence from tobacco. In many European countries, however, a large proportion of physicians are still smoking themselves. To assess smoking habits of future physicians, a cross-sectional study was conducted in 1992/1993 among 817 students enrolled in the first, third, and fifth years of medical school at the university of Ulm, Germany, using a self-administered questionnaire. The overall participation rate was 85.2%. Prevalence of current smoking was 17.6% among female participants and 29.2% among male participants. Among male students, smoking prevalence varied between 22.9% in the first year and 34.6% in the fifth year of medical school. Among female students, there were only minor differences in smoking prevalence between the first, third and fifth years at medical school. Most of the smokers had started to smoke prior to entry into medical school. Among third and fifth year students, slightly more students quit smoking than started to smoke during medical school, and smokers in the fifth year of medical school were more frequently willing to quit than smokers in the first and third year. Factors associated with regular smoking of medical students, after adjustment for potential confounders in multivariable analysis, were age, sex, and maternal smoking. While smoking prevalence was somewhat lower in the present study than in previous studies from Germany, further efforts are needed to reduce smoking among future physicians in this country.