Research On Dynamic Correlation And Factors Of Nosocomial Infection Rate In A Third Grade Hospitalbased On Copula

ObjectivesHAI is serious.There have been several outbreaks of nosocomial infections in China in recent years.It not only brings serious mental and physical pain to patients and their families.It also caused medical institutions to suffer losses in economy and reputation.It leads to escalating conflicts between doctors and patients.Now,Researchers generally conduct microscopic studies on the influencing factors or factors of nosocomial infection in a single department or a certain disease.It is rarely studied from the correlation between multiple departments or diseases.So,We hope to understand the status of hospital infection through this study.And the dynamic correlation degree and influencing factors of infection rate in some departments are analyzed based on Copula theory.To provide a new idea and means for monitoring and prevention of nosocomial infection.To prepare for the establishment of an early warning model of nosocomial infection outbreak at the hospital level.Methods1.This is a retrospective research from 2010 to 2016.Relevant data of inpatients were investigated and entered into the EXCEL form one by one by the full-time staff of the Department of nosocomial infection control using the nosocomial infection management and surveillance system.All data were entered by three full-time personnel and checked by the fourth person.We used EXCEL tables to record monthly infection rates and infection cases according to departments and infection sites,and the annual ranking of data,ranking of infection sites.And we ranked each year’s data and ranked the part of infection.2.We collected the basic information and nosocomial infection-related information of inpatients in three departments of hepatobiliary surgery,neurosurgery and thoracic surgery in 2015.We use SPSS20.0 for all data.We used logistic regression analysis to optimize the core risk factors for HAI of them.3.We collected monthly nosocomial infection rates in thoracic,hepatobiliary and neurosurgical inpatients from January 2010 to June 2017.We used Copula theory to study the dynamic correlation of nosocomial infection rate in departments.First,the suitable marginal distribution function and Copula function were selected.Secondly,we estimated the parameters of the selected Copula function and characterize the correlation between the two departments and analyze the possible influencing factors of the correlation.Results1.We studied 663331 patients in the hospital from 2010 to 2016.The HAI rate was 3.21%.The total HAI rate was 4.04%.The annual hospital infection rate fluctuated at 2.66%~3.61%.ICU,Department of Hematology,Department of thoracic surgery,and rehabilitation departments ranked the top 5 in the past 7 years.The first is ICU in 6 consecutive years.In surgical departments,Department of thoracic surgery,Department of neurosurgery and hospital for 7 consecutive years,they ranked in the top 15.And their hospital infection rates were higher than the average annual level.Lung infection,urinary tract infection and blood infection rank the top 5.Lung infection was first in 7 years.And it accounted for more than 35% of the total hospital infection cases.For postoperative HAI,department of thoracic surgery,department of neurosurgery and surgery for 7 consecutive years after the hospital infection rate ranked the top 5.And lung infection,incision infection and abdominal infection were ranked the top 5 in hospital infection after surgery.Lung infection was first in 7 years,and it accounts for more than 35% of the total HAI cases.2.The hospital infection rate in Department of hepatobiliary surgery was 6.25% in 2015.The infection site of the Department is mainly abdominal cavity,lung,blood flow and incision.The postoperative HAI rate was 10%.The site of infection was mainly in the 4 parts above.Multivariate logistic analysis showed the key risk factors were antimicrobial use,number of ventilator use and combination of antimicrobial agents.The hospital infection rate in Department of neurosurgery was 8.57% in 2015.The infection sites in this department are mainly lungs and blood.The postoperative HAI rate was 16.19%.The site of infection was mainly in the 2 parts above.Multivariate logistic analysis showed the key risk factors were whether to use special class antibiotics,whether to use antibiotics in combination with ventilators and so on.The hospital infection rate in Department of thoracic surgery was 9.61% in 2015.The infection sites in this department are mainly lungs and blood.The postoperative HAI rate was 26.88%.The site of infection was mainly in the 2 parts above.Multivariate logistic analysis showed the key risk factors were antibiotics use,diabetes mellitus and malignant tumor.3.Estimation of distribution function by KS test and Q-Q graph,the efect test showed the nonparametric kernel density function is used to describe the better staining rate in Department of thoracic surgery,Department of hepatobiliary surgery and hospital,and the results were checked by uniform distribution,and they conformed the requirement of Copula function.We estimate the parameters of dynamic and static Copula functions,it showed that dynamic Copula was more suitable than static Copula.Selection criteria based on maximum likelihood function and AIC minimum value,time-varying SJC Copula was selected to characterize the interdependent structure of infection rates between hepatobiliary surgery and thoracic surgery,neurosurgery and hepatobiliary surgery,thoracic surgery and neurosurgery,respectively.The results showed that the correlation between 22 infection rates of three departments was far higher than their high correlation.Only Department of hepatobiliary surgery and Department of thoracic surgery had a high correlation when infection rate was high,but the correlation was high when infection rate was low.ConclusionsThe overall situation of hospital infection was relatively stable.It also conformed the requirements of the state for less than 10% of the infection rate in hospitals with more than 500 beds.Hospitals should focus on the control of key departments including ICU,hematology,thoracic surgery,neurosurgery and hepatobiliary surgery.Prevention and control of pulmonary infection should be strengthened.Prevention and control of internal and external factors of hospital infection should be strengthened in hospitals.By using Copula,The results showed that Copula had better fitting rate for the infection rate of three departments.The influencing factors of two or two infection rates in the three departments were the distribution characteristics of disease types,environmental factors,operation,nursing,antibiotic management and infection control intervention.By the method,we build a macroscopic model to study the correlation between nosocomial infection rates.It provides a new working mode for nosocomial infection management.It also provides a new idea for hospital infection management.It provides an empirical basis for establishing an early warning model of nosocomial infection outbreak in a general hospital.