目標︰吸菸是許多導致過早死亡的原因中能事先予以預防的。群體的吸菸盛行率是反映菸害程度與鑑別高危險群的重要指標。本研究目的在呈現台灣地區菸害防制法實施前的成人吸菸盛行狀況，並探討教育及職業與吸菸的關係。方法：自1993年，1994年及1996年分別完成的全台灣地區抽樣調查資料檔，各挑出年齡、性別、教育程度、職業等級與吸菸行為等五個變項進行統計分析。結果：不論年度，成年男性之吸菸率以25-44歲、國中程度者及技術性工人較高。成年女性吸菸率則是65歲以上較高。經年齡標準化處理後，1993年，1994年及1996年成年男性之吸菸率分別為50.07%、 51.59%和51.89%；女性則分別為2.82%、6.35%和4.63%。控制性別與年齡的影響後，有吸菸的可能性為大專及以上教育程度者顯著地低於小學及以下程度者；技術性工人顯著地高於專業或行政人員。在戒菸的可能性方面，則為大專及以上程度者顯著地高於小學及以下程度者；技術性工人和半技術工人及無業者則顯著地低於專業或行政人員。結論：為監測國人吸菸盛行狀況，定期舉辦且採統一定義的盛行率調查是必須的。教育程度及職業等級與吸菸行為的顯著關係，可供規劃菸害防制介入策略與界定高危險群的參考。

　　　　　Objectives: Smoking has been recognized as the single most important preventable causes of premature death. Prevalence of tobacco use in a population is an important measure of the magnitude of smoking problem. The purpose of this study is to understand Taiwanese adults' smoking rates and the associated risk factors for smoking before the implementation of the Tobacco Hazards Control Act in 1997. Methods: Data were obtained from three nationwide surveys. Variables including age, sex, education, occupation, and smoking behavior were selected for analysis. Results: Major findings were: (1)For male adults, the highest current smoking rates were found in the subgroups of those aged 25-44 years, technical workers and those with an educational level of junior high school. For female adults, the highest current smoking rate was found in those 65 years and older. The age-adjusted smoking rates for males who currently smoke in 1993, 1994, and 1996 were 50.07%, 51.59%, and 51.89%, respectively. The age-adjusted smoki ng rates for females who currently smoke for the same years were 2.82%, 6.35% and 4.63%, respectively. (2)After controlling for sex and age, the smoking rate for individuals who currently smoke and for those who have ever smoked were significantly lower for the subgroup of college graduates than for the subgroup of individuals with only an elementary education. Compared to professionals, technical workers had higher smoking rates. (3)Smokers with higher educational levels and those working as professionals had the highest cessation rates. Conclusions: In order to monitor the smoking prevalence in Taiwan, it is imperative to establish a surveillance system. Since education and occupation have been proved to play important roles, researchers and practitioners should use these findings to identify and develop effective strategies for high risk groups.