This study estimated the burden of premature death due to smoking in Korea between 1990 and 1999 using the years of life lost (YLL) due to premature death method. To implement this study, age-group-specific YLL due to premature death were calculated by employing the standard expected years of life lost method. YLL due to smoking were calculated based on assumptions and methods developed by the Global Burden of Disease Study Group. The burden of premature death due to smoking was estimated by multiplying the population attributable risk by the YLL of smoking-related diseases. In 1999, the burden of premature death due to smoking was 57.7% in males and 11.4% in females in Korea. The burden of premature death due to smoking increased from 1643 person years per 100,000 in 1990 to 1888 person years in 1999 for males, and increased from 151 person years in 1990 to 225 person years in 1999 for females in Korea. Our results suggest that the method employed in this study, generated in quantified terms, enabled the burden of premature death due to smoking to be obtained comparably with methods used by other international studies in this field, and thus can provide a rational basis for national health policy planning regarding premature death from smoking and the related risk factors in Korea.