Impact of the age-adjusted Charlson comorbidity index on in-hospital mortality in long-living people with community-acquired pneumonia and a predictive mortality model / 中华老年医学杂志

ABSTRACT

Objective:To investigate the relationship between the age-adjusted Charlson comorbidity index(aCCI)and the risk of in-hospital death for people aged ≥ 90 years with community-acquired pneumonia(CAP), and to construct a novel scoring model for predicting in-hospital mortality.Methods:Basic personal and medical data about sex, age, hospitalization days, hospitalization expenses, in-hospital outcomes and discharge/admitting diagnosis of CAP patients aged ≥ 90 years hospitalized in Peking University Third Hospital between 2010 and 2019 were collected retrospectively.Multivariate Logistic regression analysis was conducted to examine the association between aCCI or other complications and in-hospital death.The receiver operating characteristic curve(ROC)was used to assess the value of aCCI and a new scoring model in predicting in-hospital death of CAP in people aged ≥ 90 years.Results:A total of 274 CAP patients aged ≥ 90 years were included in this study, of whom 85 died in hospital.Multivariate Logistic regression analysis showed that malnutrition( OR=2.21, 95% CI: 1.05-4.67, P<0.05), respiratory failure( OR=18.91, 95% CI: 9.34-38.25, P<0.001)and aCCI( OR=1.51, 95% CI: 1.23-1.85, P<0.001)were prognostic factors for in-hospital death in CAP patients aged ≥ 90 years.Based on the above results, a novel scoring model, MRC(malnutrition, respiratory failure, aCCI)was established.The area under the ROC curve of the aCCI score for predicting the risk of in-hospital death in CAP patients aged ≥ 90 years was 0.743(95% CI: 0.684-0.802). The area under the ROC curve of the MRC score was 0.891(95% CI: 0.848-0.933), indicating a higher predictive value than that of the aCCI score alone( Z=6.337, P<0.001). Conclusions:The MRC score model can be used to evaluate and predict the risk of in-hospital death in long-living CAP patients.