Changing trend and multivariate analysis of prevalence rates of healthcare-associated infection in a tertiary hospital for 10 years
10.12138/j.issn.1671-9638.20245142
- VernacularTitle:某三级医院10年医院感染现患率变化趋势及多因素分析
- Author:
Zi-Quan ZHOU
1
;
Jin-Ai QIN
;
Juan HUANG
;
Yong-Kang YE
;
Qun-Xiu GUO
;
Lan-Lan LI
;
Jing-Jing YA
;
Ying-Hua ZHANG
Author Information
1. 广西医科大学第一附属医院医院感染管理科,广西 南宁 530021
- Keywords:
healthcare-associated infection;
prevalence rate;
trend analysis;
logistic regression;
multilayer per-ceptron
- From:
Chinese Journal of Infection Control
2024;23(9):1135-1141
- CountryChina
- Language:Chinese
-
Abstract:
Objective To describe the changing trend and related factors of prevalence rates of healthcare-associa-ted infection(HAI)in a tertiary hospital in the past 10 years,and analyze the influencing factors for HAI.Methods A cross-sectional survey on HAI was conducted for 10 consecutive years from 2013 to 2022(one day was selected as the survey day each year),data were collected.The distribution and related factors of prevalence rates of HAI were analyzed by trend-x2 test and Pearson correlation coefficient.Multivariate logistic regression and multilayer percep-tron(MLP)models were constructed to analyze the independent effect and significance of factors.Results From 2013 to 2022,the prevalence rates of HAI ranged from 4.66％to 8.07％in this hospital,showing a linear upward trend.The proportions of ICU patients and utilization rate of central venous catheters within 2 days before the sur-vey showed linear upward trends,while the proportion of patients with urinary catheters within 2 days before the survey and proportion of patients undergoing surgery within 30 days before the survey decreased.The MLP model revealed that the top 3 important factors for HAI were length of hospital stay＞10 days,admission in ICU,and in-dwelling central venous catheters within 2 days before the survey.Multivariate logistic regression model indicated that length of hospital stay＞10 days,indwelling central venous catheters or urinary catheters within 2 days before the survey,surgery within 30 days before the survey,and admission in ICU were independent influencing factors for HAI.Conclusion The incidence of HAI in this hospital presents a linear increase in recent 10 years,the causes should be further analyzed and the direction of intervention should be determined through targeted surveillance.Adopting trend test statistical analysis method,logistic regression,MLP multi-factor model can further explore the data value of HAI prevalence survey.
