Abstract:Objective To evaluate the risk factors for post-operative pneumonia (POP) in patients undergoing hip joint surgery, evaluate the application value of propensity score matching (PSM) in processing of retrospective data. Methods Data of 1 141 patients undergoing hip joint surgery from January 1, 2018 to December 31, 2019 in the department of orthopedics and joint surgery of a hospital were collected, data of patients after hip joint surgery were analyzed by propensity score. Results There were 102 cases of POP among 1 141 hip joint surgery patients, incidence was 8.94%. Before PSM, there was no significant difference in POP incidence between fracture patients and non-fracture patients (P>0.05). Before PSM, there were significant differences in 8 covariates, including gender, age, anesthesia, delirium and dementia, ASA classification, malignant tumor, diabetes and cardiovascular disease between fracture patients and non-fracture patients (all P < 0.05). After PSM, distribution of the above 8 covaria-tes reached equilibrium between two groups (all P>0.05). Logistic multivariate analysis showed that age>65 years old, history of chronic obstructive pulmonary disease and fracture were independent risk factors for POP in hip joint surgery patients (all P < 0.05); compared with general anesthesia, intravertebral anesthesia was an indepen-dent protective factor for POP in hip joint surgery patients (P < 0.05). Conclusion PSM method can effectively reduce the confounding bias among groups in observational study. Fracture is one of the risk factors for POP in hip joint surgery patients.