Choi Mona, Lee Hyeong Suk
Nursing Policy Research Institute, College of Nursing, Yonsei University, Seoul, Korea.
Seoul Women's College of Nursing, Seoul, Korea.
Nurs Crit Care. 2016 Jul;21(4):206-13. doi: 10.1111/nicc.12223. Epub 2016 Feb 3.
The CPSCS was developed to assess the nursing care demands of patients in intensive care units (ICUs).
This study aimed to examine the Critical Patient Severity Classification System (CPSCS) score as an independent predictor of patient hospital outcomes.
This study was a secondary analysis.
Data from 6380 cases were extracted from the electronic medical records in ICUs at a tertiary hospital in Korea during 2010-2012. To examine the association of the CPSCS score with 30-day ICU mortality, the Cox proportional hazards model and Kaplan-Meier survival curves were used, and generalized linear regression models of gamma distribution were developed for ICU length of stay (LOS).
More patients were admitted to surgical ICUs than medical ICUs (4664 versus 1716) during the study period. Medical ICU patients had longer ICU LOS, higher 30-day ICU mortality and a higher mean CPSCS score than surgical ICU patients. Cox analysis indicated that the mid and high CPSCS score groups had 1·687 and 2·913 times higher mortality risk, respectively, than the low CPSCS score group after adjusting for age, sex and primary diagnosis. The CPSCS score significantly predicted ICU mortality in both medical and surgical ICUs. Multivariate generalized linear regression indicated that CPSCS score was a significant predictor of ICU LOS after adjusting for other covariates.
The CPSCS score can be used to efficiently predict ICU mortality and LOS in patients admitted to the medical and surgical ICUs, although only the high CPSCS score group had significantly high mortality than the low CPSCS score group in the medical ICU.
The findings of this study contribute to valuable evidence that nursing-related factors have an impact on patient outcomes such as ICU mortality and LOS and that they have implications for hospital management, clinical practice and future research.
危重症患者严重程度分类系统（CPSCS）旨在评估重症监护病房（ICU）患者的护理需求。
本研究旨在探讨危重症患者严重程度分类系统（CPSCS）评分作为患者医院结局独立预测指标的情况。
本研究为二次分析。
2010 - 2012年期间，从韩国一家三级医院ICU的电子病历中提取6380例患者的数据。为检验CPSCS评分与30天ICU死亡率的关联，使用Cox比例风险模型和Kaplan - Meier生存曲线，并针对ICU住院时间（LOS）建立伽马分布的广义线性回归模型。
研究期间，入住外科ICU的患者多于内科ICU（4664例对1716例）。内科ICU患者的ICU住院时间更长，30天ICU死亡率更高，且CPSCS平均评分高于外科ICU患者。Cox分析表明，在调整年龄、性别和主要诊断后，CPSCS评分中、高分组的死亡风险分别是低分组的1.687倍和2.913倍。CPSCS评分在内科和外科ICU中均能显著预测ICU死亡率。多变量广义线性回归表明，在调整其他协变量后，CPSCS评分是ICU住院时间的显著预测指标。
CPSCS评分可有效预测入住内科和外科ICU患者的ICU死亡率和住院时间，尽管在内科ICU中只有CPSCS高分组的死亡率显著高于低分组。
本研究结果提供了有价值的证据，表明护理相关因素会影响ICU死亡率和住院时间等患者结局，对医院管理、临床实践和未来研究具有启示意义。
