结直肠癌患者术后医院感染影响因素及与按病种分值付费支付差额的中介效应

Post-operative healthcare-associated infection influencing factors and mediating effect of diagnosis-intervention packet payment differentials in colorectal cancer patients

摘要:

目的探讨结直肠癌患者术后医院感染的潜在影响因素及与按病种分值付费(DIP)支付差额间的中介效应关系。方法回顾性分析广州市某三级甲等肿瘤医院结直肠癌手术患者病历资料,按医院感染情况分为感染组和非感染组。采用秩和检验或卡方检验比较两组患者人口学基本信息和不同组别患者DIP支付差额的差异,采用logistic回归分析各潜在因素对医院感染发生的影响。采用bootstrap法进行中介分析,探讨医院感染和住院总日数对各因素对DIP支付差额影响的中介效应。结果共纳入350例患者,其中感染组50例,非感染组300例,医院感染发病率为14.29%。logistic回归分析结果显示,中心静脉置管日数≥10.00 d的患者发生医院感染风险是<10.00 d的13.558倍(P<0.001);导尿管插管≥3.00 d的患者发生医院感染风险是<3.00 d的2.388倍(P=0.022)。不同年龄、预后营养指数(PNI)、医院感染情况、住院总日数、手术时长、中心静脉置管日数、导尿管插管日数的患者DIP支付差额比较,差异均有统计学意义(均P<0.05)。中介分析结果显示,发生医院感染通过影响住院总日数引起DIP支付差额发生变化,住院总日数的中介效应值为0.038,中介效应占总效应的35.68%。结论医疗机构应重视结直肠癌手术患者因中心静脉置管和导尿管插管时间延长所导致的医院感染,减少患者住院总日数,从而降低DIP支付差额增加所致的住院费用超支。

Abstract:

Objective To explore the potential influencing factors of post-operative healthcare-associated infection (HAI) in colorectal cancer patients, as well as the mediating effect relationship between the influencing factors and the diagnosis-intervention packet (DIP) payment differentials. Methods Medical data of patients who underwent colorectal cancer surgery in a tertiary first-class cancer hospital in Guangzhou were retrospectively analyzed. According to HAI status, patients were divided into infection group and non-infection group. Baseline demographic information and differences in DIP payment differentials between two groups of patients were compared by rank sum test or chi-square test. The influence of each potential factor on the occurrence of HAI was analyzed by logistic regression. Mediating analysis was preformed by bootstrap method, and mediating effect of HAI and total hospitalization days on DIP payment differentials was evaluated. Results A total of 350 patients were included in analysis, 50 were in the infection group and 300 in the non-infection group. The incidence of HAI was 14.29%. Logistic regression analysis result showed that risk of HAI in patients with central venous catheterization ≥10.00 days was 13.558 times higher than that <10.00 days (P<0.001); risk of HAI in patients with urinary catheterization ≥3.00 days was 2.388 times higher than that <3.00 days (P=0.022). There were all statistically significant differences in DIP payment differentials among patients with different ages, prognostic nutritional index (PNI), HAI status, total length of hospitalization stay, duration of surgery, central venous catheterization days, and catheterization days (all P<0.05). The mediating analysis results showed that the occurrence of HAI resulted a change in DIP payment di-fferentials by affecting the total number of hospitalization days. The mediating effect value of total hospitalization days was 0.038, accounting for 35.68% of the total effect. Conclusion Medical institutions should pay attention to HAI resulting from prolonged central venous and urinary catheterization in patients underwent surgery for colorectal cancer, reducing the total length of hospital stay, thus reducing the overruns associated with the increased DIP payment differentials.