Objectives : There have been few studies examining the differences in health care utilization across social classes during the last year of life. Therefore, in this study we analyzed the quantitative and qualitative differences in health care utilization among cancer patients across educational classes in their last year of life, and derived from it implications for policy. Methods : To evaluate health care utilization by cancer patients in the last year of life, Death certificate data from 2004 were merged with National Health Insurance data (n=60,088). In order to use educational level as a social class index, we selected the individuals aged 40 and over as study subjects (n=57,484). We analyzed the differences in the medical expenditures, admission days, and rates of admission experience across educational classes descriptively. Multiple regression analysis was conducted to evaluate the association between medical expenditures and independent variables such as sex, age, education class, site of death and type of cancer. Results : The upper educational class spent much more on medical expenditures in the last one year of life, particularly during the last month of life, than the lower educational class did. The ratio of monthly medical expenditures per capita between the college class and no education class was 2.5 in the last 6-12 months of life, but the ratio was 1.6 in the last 1 month. Also, the lower the educational class, the higher the proportion of medical expenditures during the last one month of life, compared to total medical expenditures in the last one year of life. The college educational class had a much higher rate of admission experiences in tertiary hospitals within Seoul than the other education classes did. Conclusions : This study shows that the lower educational classes had qualitative and quantitative disadvantages in utilizing health care services for cancer in the last year of life.