Abstract
BACKGROUND: Few studies examine the link between measured process of care and outcome. OBJECTIVE: To evaluate the relationship of claims-based assessment of process of care to subsequent function and survival. RESEARCH DESIGN: Retrospective cohort study using claims from 1999 to assess performance on 41 quality indicators (QIs) from the Assessing Care of Vulnerable Elders (ACOVE) measurement set on functional decline and death in 2000. SETTING: Community-dwelling individuals. SUBJECTS: All persons ≥75 years enrolled in Medicare and Medicaid in 19 California counties in 1998 and 1999 who received In Home Supportive Services. MEASURES: Quality of care index, activities of daily living, and instrumental activities of daily living (IADL) need indices, mortality. RESULTS: Total 21,310 persons were eligible for a mean of 7.1 QIs; and received 46% of recommended care. The ADL index increased from 8.1 to 11.6 between baseline and follow-up. The IADL index increased from 13.6 to 14.1. Fifteen percent of the cohort died in 2000. After accounting for number of QIs triggered, baseline function and other covariates, better quality was associated with better function at follow-up. Ten percent better quality was associated at follow-up with 0.21 lower ADL need score [95% confidence interval (CI), 0.25-0.17], 0.022 lower IADL need score (95% CI, 0.032-0.013), and lower odds of death (0.91; 95% CI, 0.89 to 0.93). CONCLUSIONS: Routinely collected data implementing ACOVE measures for community vulnerable elders generate quality scores that are directly related to patient functional and survival outcomes. These findings suggest that population-based assessment of care is feasible for vulnerable older persons.