A Prospective Evaluation of the Charlson Comorbidity Index for Use in Long‐Term Care Patients

Abstract

OBJECTIVES: Because of the high prevalence of coexisting medical conditions in frail older adults, clinical investigators often need to adjust for comorbidity when assessing the effect of long‐term care (LTC) on patient outcomes. This study examined the prognostic value of the Charlson Comorbidity Index (CCI) in predicting 3‐year mortality and functional decline in the LTC setting and compared its prognostic value to that of two data‐derived comorbidity indices. DESIGN: Longitudinal cohort study. SETTING: Eighty‐eight residential care facilities from Quebec, Canada. PARTICIPANTS: Two hundred ninety‐one dependent older adults aged 65 and older. MEASUREMENTS: Subjects' functional abilities were assessed at baseline and 3 years later with the revised version of the Functional Autonomy Measurement System (SMAF). Comorbidity data and the exact date of death for those who had died were collected retrospectively from the subjects' medical files. Subjects were classified as functional decliners if they died or gained 5 points or more on the SMAF between the two assessments. RESULTS: Multivariate Cox and logistic regressions were used to derive two new comorbidity indices, one for predicting mortality and the other for identifying functional decliners. Although the CCI performed well in predicting these two outcomes, its performance was generally inferior to that of the two newly proposed indices. CONCLUSIONS: Findings suggest that the CCI can be improved upon when used to measure comorbidity in LTC patients. J Am Geriatr Soc 50:740–745, 2002.