摘要:Introduction. Hip replacement (HR) operations are increasing. Short term mortality isan indicator of quality; few studies include risk adjustment models to predict HR outcomes.We evaluated in-hospital and 30-day mortality in hospitalized patients for HRand compared the performance of two risk adjustment algorithms.Materials and methods. A retrospective cohort study on hospital discharge records ofpatients undergoing HR from 2000 to 2005 in Tuscany Region, Italy, applied All-PatientRefined Diagnosis Related Groups (APR-DRG) and Elixhauser Index (EI) risk adjustmentmodels to predict outcomes. Logistic regression was used to analyse the performanceof the two models; C statistic (C) was used to define their discriminating ability.Results. 25 850 hospital discharge records were studied. In-hospital and 30-day crudemortality were 1.3% and 3%, respectively. Female gender was a significant (p < 0.001)protective factor under both models and had the following Odds Ratios (OR): 0.64 forin-hospital and 0.51 for 30-day mortality using APR-DRG and 0.55 and 0.48, respectively,with EI. Among EI comorbidities, heart failure and liver disease were associatedwith in-hospital (OR 9.29 and 5.60; p < 0.001) and 30-day (OR 6.36 and 3.26; p < 0.001)mortality. Increasing age and APR-DRG risk class were predictive of all the outcomes.Discriminating ability for in-hospital and 30-day mortality was reasonable with EI (C0.79 and 0.68) and good with APR-DRG (C 0.86 and 0.82).Conclusions. Our study found that gender, age, EI comorbidities and APR-DRG riskof death are predictive factors of in-hospital and 30-day mortality outcomes in patientsundergoing HR. At least one risk adjustment algorithm should always be implementedin patient management.
关键词:hip replacement; mortality; risk adjustment; APR-DRG; Elixhauser index