Acute Physiology and Chronic Health Evaluation (APACHE ) II評估表的發展已經有超過20年的歷史，但仍是目前用途最多樣化也是最廣為接受的疾病量化表。各醫院的政策、設備與加護病房病患的特色，都會有差異，所以各個加護病房依據APACHE II分數來制定醫療政策時，都應該先分析本身的特質與數據，才能依據顯著影響死亡率的因子，訂定適合的治療或者是侵入性的監測，以期能降低死亡率。
本研究收集中部某醫學中心外科加護病房，自2008年1月份到2008年9月份，共552名患者，排除45名患者後，有507名患者，分成未接受手術、接受緊急手術與接受選擇性手術組。比較APACHE II的分數、手術、年齡、性別及加護病房住院天數與患者的死亡率之間的相關性。分析顯著影響死亡率的分界點，再以線性迴歸分析各個變項的勝算比(Odds ratio)。
結果發現年齡與死亡率成正相關，年齡大於80歲的患者死亡率顯著增加(Odds ratio = 4.07)，APACHE II分數超過20分(Odds ratio = 4.56)、接受緊急手術(Odds ratio = 1.75)與加護病房住院天數7天(Odds ratio = 2.56)以上，都會增加死亡率。
依據該外科加護病房的數據分析，年齡大於80歲、APACHE II分數超過20分、接受緊急手術與加護病房住院天數7天以上的患者，屬於高危險性的患者，因此在併發症調查與預防，應該更加積極，可能存在有危險性的侵入性監測與治療，也更可合理的建議使用於這一些高死亡率的患者，以降低死亡率。
Acute Physiology and Chronic Health Evaluation (APACHE ) II score developed for more than 20 years. Currently, the most popular scale to quantify disease and predict various outcomes is still APACHE II score. Every intensive care unit should set up a treatment modality and goal according to the characteristics of patients which were different from the individualized hospital policies and facilities. We believe the APACHE II score and the parameters will help us to know the risk factors of mortality after statistically analysis. An adequate or even more invasive monitoring or treatment policies could be made after calibration by these factors to improve the high mortality rate condition.
This study collected 552 patients in a surgical intensive care unit of a medical center in mid-Taiwan. 507 patients were included after 45 patients excluded. The collecting period was from January 2008 to October 2008. The APACHE II score of these patients were collected within 24 hours after they admitted in the intensive care unit. These 507 patients were divided into 3 groups: non-operative, emergency-operation and elective-operation. We statistically analysis the correlations between mortality rate and APACHE II score, operation, age, gender and the length of stay (LOS) in intensive care unit. We used t-test, χ2 test, analysis of variance (ANOVA). Logistic regression was used to calculate the odds ratio of each variable.
The odds ratios of mortality rate by each variable was 4.07 of patient’s age more than 80 years； 4.56 of APACHE II score more than 20 points; 1.75 of received emergency operation and 2.56 of LOS in intensive care unit more than 7 days.
According to the odds ratios of each variable, the active survey and prevention of complications should be indicated for these patients that age more than 80 years or APACHE II score more than 20 points or received emergency operation or LOS in intensive care unit for more than 7 days in this surgical intensive care unit. Potentially harmful invasive monitoring or treatments were also more indicated in the higher mortality risk groups in order to improve the mortality rate.
