Application of propensity score matching in the evaluation of economic burden of healthcareassociated infection

Abstract:

ObjectiveTo improve the balance of each covariate using propensity score matching(PSM), provide a new method for evaluating economic burden of healthcare associated infection(HAI).Methods158 976 patients who were discharged from a large general hospital between January 1, 2015 and December 31, 2016 were selected as the studied subjects, a 1∶1 PSM matching of patients with HAI and without HAI was performed by SPSS 22.0 software, balanced database after matching was analyzed to evaluate the effect of HAI on patients’ length of hospital stay and hospitalization expense.ResultsA total of 2 388 pairs were successfully matched by using PSM, the successful matching rate was 97.75%, there was no statistical difference in covariates between HAI group and nonHAI group after matching (all P>0.05). The median of length of hospital stay in HAI group and nonHAI group were 21 and 10 days respectively, difference in median was 11 days, difference between two groups was statistically significant (P<0.05). The median of total hospitalization expenses in HAI group and nonHAI group were 33 618.38 and 15 004.93 Yuan respectively, difference in median was 18 613.45 Yuan, difference between two groups was statistically significant (P<0.05). Fees for bed, western medicine, traditional Chinese medicine, therapy, examination, laboratory testing, diagnostic check, nursing, blood transfusion, operation, material, and anesthesia in HAI group were all higher than nonHAI group (all P<0.05), difference in expense of western medicine was the largest, which was 6 987.00 Yuan, the other fees were not significantly different.ConclusionPSM results showed that HAI causes serious financial burden, PSM can provide a new method for evaluating the economic burden of HAI.