중증도지표 (APACHE II)를 이용한 중환자실 입원환자의 사망률 분석

Abstract

[한글]

중환자의 중증도 분포를 조사하고 중증도에 따라 각 병원을 비교하여 보다 정확한 사망

률을 측정하고 중환자실 구조와 치료과정의 잠재적 차이가 중환자 치료결과에 영향을 주

는지를 알아보고자 하였다. 3개 대학 부속병원 중환자실 환자 493명을 대상으로 의무기록

자료를 이용하여 APACHE Ⅱ로 중증도를 조절하고 사망률을 분석하였다.

1. 중환자실 환자의 APACHE Ⅱ 중증도 점수는 0부터 51점으로 다양하였고 중증도의 분

포는 병원간 차이가 있었다.

2. APACHE Ⅱ절수에 따른 사망률 분포는 각 병원에서 유의한 차이가 있었다. 중증도에

따른 사망률은 중증도가 높은 그룹에서 낮은 그룹보다 유의하게 높았다. 동일한 점수 그

룹에서도 병원간 사망률은 차이가 있었다.

3. 병원간 실제사망률과 기대사망률의 비는 큰 차이가 없었다.

4. 환자의 치료적 중재수준(TISS)은 병원간 유의한 차이가 없었다. 본 연구에서 대상은

비슷한 수준의 대학 부속병원으로 중증도 분포에 따라 치료력 중재수준이나 중환자실 시

설이 비슷하여 결과의 차이가 적었다. 그러므로 중증도 지표를 이용하여 중환자를 그룹화

하고 표준화된 방법으로 자료를 수집하여 분석 연구함으로서 중환자실 자료정보화와 효율

적인 질관리 방안을 마련할 수 있다.

[영문]

To contrast mortality for groups of ICU patients treated in different hospitals,

Ⅰ retrospectively studied 493 patients in intensive care units at 3 university

hospitals. By using the information of the APACHE system, on other factors such as

diagnosis, indication for admission, and surgical status, Ⅰ tried to compare

treatment courses and outcomes of ICU patients in each hospital. Ⅰ categorized

each patients by APACHE Ⅱ score which summarized, and measured the risk

factors-acute physiologic disorders, chronic health status and age. Ⅰ compared

actual death rates with predicted, rates using APACHE Ⅱ multiple logistic

regression equation.

The results of the study is as follows:

1. The APACHE Ⅱ score had a wide distributions ranging from 0 to 51. The average

APACHE Ⅱ score of each hospital was 12.9, 14.6, 9.9 points, proved to be a

significant, variation. Distribution of mortality by APACHE Ⅱ score proved to be a

significant variation in each hospital

2. Hospital mortality rate by severity score was higher in higher groups than

lower APACHE Ⅱ points groups. Mortality rate between hospitals proved to be a

significant variation in similar groups of serverity score.

3. There were substantial differences in severity of acute illness anong the

hospitals which accounted for most of the variation in death rates. The ratio

between observed death rate and predicted death rate to be a low variation each

hospital.

4. Average therapeutic intervention points given, when adjusted for the type and

severity of illness of patients treated, was similar at 3 hospitals.

Three hospitals had similar technical capabilities in their units but differed in

organization, staffing, research and eduation. Ⅰ examine whether these substantial

differences in the structure and process of intensive care influnce effectiveness

of care, as measured by hospital motality.

In this study, substantial differences in the structure did not influnce

effectiveness of care, in the three university hospitals. These findings suggest

that the use of a general severity of illness index and multivariate statistical

techniques could improve the outcome of intensive care unit's patients.

These findings suggest that we could improve the effectiveness of ICU management

through analyzing the variation of ICU death rates and applying this study by using

standarized methods for improving the sevenity of illness index.