Introduction Smokers typically experience more sleep problems than non-smokers, so sleep problems are frequently reported by smokers who are attempting to quit and these can be expected to develop also after initiation of smoking. However, little is known about the effects of starting smoking on sleep problems. Some research suggests smoking reduces stress, which indicates that some aspects of smoking could benefit sleep. This study aims to contribute to the understanding of the asymmetry of the relationship between smoking and sleep problems. Methods To test the asymmetry of the relationship between smoking and sleep problems, a logistic regression analysis was conducted via GEE (generalized estimating equation). GEEs are an extension of Generalized Linear Models (GLMs) which allow for longitudinal data and correlation among observations of the same unit (person). This technique was recently proposed to test the asymmetry of causal effects based on panel data. The data used were from the LISS panel, a representative sample of Dutch individuals. The sample (n=1477) contained 259 individuals that changed their smoking status at least once. When conducting the analysis, we controlled for time-varying variables such as: subjective health, BMI, partner status, and parental status. Results The asymmetric effects of smoking on sleep problems were statistically significant, suggesting that starting smoking and quitting smoking are not similarly related to sleep problems. Starting smoking was not significantly related to sleep problems, but quitting smoking increased the odds of sleep problems by 23%. No statistically significant relationship was found between sleeping problems and changes in BMI and parental status. Subjective health was negatively associated with sleeping problems. Stopping to live with a partner increased the likelihood of sleep problems by 41%. Conclusions Quitting smoking can have a different impact on sleep problems than starting smoking. In this study, starting smoking did not result in sleep ...