이 연구는 가계소득 중 일정기준치(10-30%) 이상의 의료비를 부담하는 경우를 과부담의료비 지출(Catastrophic Health Expenditure)로 정의한 후, 제8차 (1998년~2002년) 도시가계조사 원자료를 이용하여 과부담의료비 가구들의 의료비가 전체의료비에서 차지하는 비중과 소득계층별 가구당 월평균 의료비(연간 3개월 이상 보고한 경우) 및 구성요소별 비중, 그리고 소득계층별 과부담의료비 지출 가구의 발생율을 연도별로 파악하였다. 또한 연도별 자료를 통합하여 일반화추정방정식(GEEs)에 의해 소득계층간 그 발생률 및 관련 요인을 분석하였다. 분석결과, 전체가구 대비 과부담의료비 지출 가구 의 규모(2.0-11.8%)에 비해 이들 가구들이 지출한 의료비가 전체의료비에서 차지하는비중(10.9-39.2%)이 상대적으로 높았으며, 동일한 과부담의료비 지출이지만 가구당 월평균 의료비의 크기는 저소득층이 고소득층 보다 통계적으로 유의하게 낮았다(p<0.05). 또한, 1998년에서 2002년사이 과부담의료비 지출가구의 발생률은 최하위소득계층(I)에서 1.5~l.9배로 다른 상위소득계층들에 비해 상대적으로 큰 폭으로 증가하였다 소득계층간 과부담의료비 지출 가구의 발생위험비는 가구소득 최하위소득계층(I)이 최상위소득계층(Ⅴ)에 비해 관련 요인 보정 전(1.39~2.02배), 보정 후(1.13~1.84배)모두 통계적으로 유의하게 높았다(p<0.05). 과부담의료비지출 가구의 발생은 가구원이적고 노인가구원이 많은 가구에서 유의하게 높았다(p

This study defined `Catastrophic Health Expenditure (CHE)` as spending more than some pre-specified percent (10?30%) of an urbanite`s household income on health care. Their total expenditure and the monthly mean expenditure outlay per household among income groups were analyzed together with their incidence during the period between 1998 and 2002. Additionally, we have identified influencing factors on their frequency of healthcare requirements by using the generalized estimating equations (GEEs) to account for the correlation between observations in generalized linear models. The data that we are using were taken from the 1998 -- 2002 Urban Household Expenditure Surveys. The results of this study show that the ratio (10.9?39.2%) of health care costs of the households with catastrophic health care expenditures to the total households` healthcare expenditures were higher than the proportions (2.0--11.8%) of total households They were the same cases, however the monthly mean health expenditure per household was significantly less in the poor(p<0.05). The incidence rates in the lowest income group were increased 1.5?1.9 times between 1998 and 2002. But, on the other hand, the other groups did not increase to that extent. Odds ratios of its incidence rate were significantly higher in only the lowest income group compared to the highest income group in both before (1.39?2.02, p<0.05 ) and just after that (1.13?1.84, p<0.05). The less family members there were with more of them being elderly, the higher the odds ratios of their incidence rates. Incidences of catastrophic health care expenditure may be potential things that contribute to the deficit of the national budget for health security. Additionally it is a financial burden that causes the poor to live in impoverishment. Consequently incidences of catastrophic health care expenditures need to be thoroughly investigated and more closely managed. As a priority, health security policies need to be developed that guarantee health coverage for the lowest income groups that have more elderly family members.