Sanagou M, Leder K, Cheng A C, Pilcher D, Reid C M, Wolfe R
Department of Epidemiology and Preventive Medicine,Faculty of Medicine,Nursing and Health Sciences,Monash University,Melbourne,VIC,Australia.
Department of Intensive Care Medicine,The Alfred Hospital,Melbourne,VIC,Australia.
Epidemiol Infect. 2016 Apr;144(5):1065-74. doi: 10.1017/S0950268815002307. Epub 2015 Oct 9.
To identify hospital-level factors associated with post-cardiac surgical pneumonia for assessing their impact on standardized infection rates (SIRs), we studied 43 691 patients in a cardiac surgery registry (2001-2011) in 16 hospitals. In a logistic regression model for pneumonia following cardiac surgery, associations with hospital characteristics were quantified with adjustment for patient characteristics while allowing for clustering of patients by hospital. Pneumonia rates varied from 0·7% to 12·4% across hospitals. Seventy percent of variability in the pneumonia rate was attributable to differences in hospitals in their long-term rates with the remainder attributable to within-hospital differences in rates over time. After adjusting for patient characteristics, the pneumonia rate was found to be higher in hospitals with more registered nurses (RNs)/100 intensive-care unit (ICU) admissions [adjusted odds ratio (aOR) 1·2, P = 0·006] and more RNs/available ICU beds (aOR 1·4, P < 0·001). Other hospital characteristics had no significant association with pneumonia. SIRs calculated on the basis of patient characteristics alone differed substantially from the same rates calculated on the basis of patient characteristics and the hospital characteristic of RNs/100 ICU admissions. Since SIRs using patient case-mix information are important for comparing rates between hospitals, the additional allowance for hospital characteristics can impact significantly on how hospitals compare.
为了确定与心脏手术后肺炎相关的医院层面因素，以评估其对标准化感染率（SIR）的影响，我们对16家医院心脏手术登记处（2001 - 2011年）的43691例患者进行了研究。在心脏手术后肺炎的逻辑回归模型中，在对患者特征进行调整的同时，考虑到患者按医院聚类的情况，对医院特征的关联性进行了量化。各医院的肺炎发生率从0.7%到12.4%不等。肺炎发生率70%的变异性可归因于医院长期发生率的差异，其余部分可归因于医院内部发生率随时间的差异。在对患者特征进行调整后，发现每100例重症监护病房（ICU）入院患者中注册护士（RN）较多的医院，肺炎发生率较高[调整优势比（aOR）为1.2，P = 0.006]，每可用ICU床位的RN较多的医院也是如此（aOR为1.4，P < 0.001）。其他医院特征与肺炎无显著关联。仅根据患者特征计算的SIR与根据患者特征以及每100例ICU入院患者中RN的医院特征计算的相同率有很大差异。由于使用患者病例组合信息的SIR对于比较医院之间的发生率很重要，因此考虑医院特征会对医院之间的比较产生显著影响。
