Please cite this article in press as: Baumeister, S.E. et al., Effects of smoking cessation on health care use: Is elevated risk of hospitalization among former smokers attributable to smoking-related morbidity?, Drug Alcohol Dependence (2006), doi:10.1016/j.drugalcdep.2006.10.015 ARTICLE IN PRESS +Model DAD-2904; No. of Pages 7 Drug and Alcohol Dependence xxx (2006) xxx–xxx Effects of smoking cessation on health care use: Is elevated risk of hospitalization among former smokers attributable to smoking-related morbidity? Sebastian E. Baumeister ∗ , Anja Schumann, Christian Meyer, Ulrich John, Henry V ¨ olzke, Dietrich Alte Institute of Epidemiology and Social Medicine, Medical School, University of Greifswald, Walther-Rathenau-Str. 48, 17487 Greifswald, Germany Received 24 July 2006; received in revised form 16 October 2006; accepted 18 October 2006 Abstract Background: Although the association between smoking status and health services use is well established, this relation is not well-studied for the comparison of current and former smokers. Some studies showed higher utilization of health services among former smokers compared to continuing smokers. This study investigates the relation between smoking status, time since smoking cessation and hospitalization in a general population sample. We hypothesized that elevated risk of hospitalization among former smokers compared with continuing smokers is related to higher smoking-related morbidity among former smokers. Methods: Data from a cross-sectional sample of 4310 adults aged 20–79 in Pomerania, Germany was used (response proportion 68.8%). Smoking status, time since smoking cessation (in years), and date of diagnosis of smoking-related diseases were determined from self-reports. We used fractional polynomials to determine the dose–response relation of time since cessation and risk of hospitalization. Confounding was investigated allowing for different sets of confounding variables. Results: We found that the probability of hospitalization was highest among those who quit 1–3 years ago and decreased thereafter. Adjustment for health status and socio-economic variables revealed that this association is attenuated by current diagnosis of smoking-related diseases. Conclusion: Short-term excess health care utilization among former smokers might result from smoking-related conditions that may have led to smoking cessation. Findings suggest that smoking cessation has long-term health benefits resulting in lower health care needs. © 2006 Elsevier Ireland Ltd. All rights reserved. Keywords: Smoking cessation; Health care utilization; Hospital use; General population; Confounding 1. Introduction Research to date has focused extensively on the effects of smoking on health (Department of Health and Human Services, 2004) and costs (Barendregt et al., 1997), but has paid lit- tle attention to the effects of smoking and quitting on health care use. Most evidence comes from studies, which distin- guished smokers, former smokers, and never smokers, and assessed their utilization of various types of health services. While some studies found that smokers use more health ser- vices than those who never smoked (Ashford, 1973; Freeborn ∗ Corresponding author. Tel.: +49 3834 867704; fax: +49 3834 866681. E-mail address: sebastian.baumeister@uni-greifswald.de (S.E. Baumeister). et al., 1990; Haapanen-Niemi et al., 1999; Izumi et al., 2001; Jee et al., 1993), other studies showed conflicting results. In a nationally representative sample from Spain, female smok- ers did not differ from female never smokers in their rates of hospitalizations (Rodriguez Artalejo et al., 2000). A commu- nity study among elderly participants in the United States (U.S.) revealed decreased use of outpatient services among smokers and increased use of hospital services among former smokers (Kaplan et al., 1992). One study that used data from a U.S. Health Maintenance Organization (HMO) revealed that while smokers used more inpatient services than never smokers, outpatient use was similar. Former smokers used less hospital services but more outpatient services than current smokers (Vogt and Schweitzer, 1985). These findings suggest that smoking cessation might influ- ence health care use. Data from a HMO-based smoking cessation 0376-8716/$ – see front matter © 2006 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.drugalcdep.2006.10.015