Abstract
OBJECTIVES: To design a proxy to the Charlson comorbidity index (PrCCI) for use in studies with pre-existing data sources and to analyse its influence on mortality risk in elderly people. DESIGN: Longitudinal study. SETTING: Population base, From "Aging in Leganes" cohort. PARTICIPANTS: All individuals with complete information (n=978, aged 65 years old and older) were included in the construction of the proxy (PrCCI). MEASUREMENTS: A proxy (PrCCI) was created based on the original Charlson Comorbidity Index (CCI) and the available pre-existing data from the "Aging in Leganes" study. The relationship between PrCCI and mortality was assessed using a Survival analysis. Cox proportional Hazard Models were adjusted for possible confusion factors. RESULTS: The PrCCI varied from 0 to 11 points (median=2 points, RI25-75=1-3). It was higher in women than in men until 79 years old, but no differences were found for both men and women from 80 years old. Individuals who scored four and more points in the new index had a higher mortality risk after 5 and 15 years of follow up, even after controlling for sociodemographic, health behaviours and health status related covariables (HR: 3.69, 95% CI: 1.52-8.96 and 2.14, 95% CI: 1.42-3.21, respectively). CONCLUSION: The proxy to the Charlson comorbidity index is easy and useful to measure comorbidity in studies in population from 65 to 75 years old, which use pre-existing data bases and data available in the primary care setting. The association between PrCCI and mortality shows that it is useful as comorbidity index.