Risk adjustment methods for Home Care Quality Indicators (HCQIs) based on the minimum data set for home care

TitleRisk adjustment methods for Home Care Quality Indicators (HCQIs) based on the minimum data set for home care
Publication TypeJournal Article
Year of Publication2005
AuthorsDalby D.M, Hirdes J.P, Fries B.E
JournalBMC Health Services Research
Volume5
Issue1
Pagination7
Date PublishedJan 18
Accession Number15656901
Keywords*Home Care Agencies/st [Standards], *Quality Indicators, Health Care/sn [Statistics & Numerical Data], *Risk Adjustment/mt [Methods], Adolescent, Adult, Age Distribution, Aged, Aged, 80 and over, Female, Health Services Research, Home Care Agencies/ut [Utilization], Humans, Male, Manitoba, Middle Aged, Ontario, Sex Distribution
Abstract

BACKGROUND: There has been increasing interest in enhancing accountability in health care. As such, several methods have been developed to compare the quality of home care services. These comparisons can be problematic if client populations vary across providers and no adjustment is made to account for these differences. The current paper explores the effects of risk adjustment for a set of home care quality indicators (HCQIs) based on the Minimum Data Set for Home Care (MDS-HC). METHODS: A total of 22 home care providers in Ontario and the Winnipeg Regional Health Authority (WRHA) in Manitoba, Canada, gathered data on their clients using the MDS-HC. These assessment data were used to generate HCQIs for each agency and for the two regions. Three types of risk adjustment methods were contrasted: a) client covariates only; b) client covariates plus an "Agency Intake Profile" (AIP) to adjust for ascertainment and selection bias by the agency; and c) client covariates plus the intake Case Mix Index (CMI). RESULTS: The mean age and gender distribution in the two populations was very similar. Across the 19 risk-adjusted HCQIs, Ontario CCACs had a significantly higher AIP adjustment value for eight HCQIs, indicating a greater propensity to trigger on these quality issues on admission. On average, Ontario had unadjusted rates that were 0.3% higher than the WRHA. Following risk adjustment with the AIP covariate, Ontario rates were, on average, 1.5% lower than the WRHA. In the WRHA, individual agencies were likely to experience a decline in their standing, whereby they were more likely to be ranked among the worst performers following risk adjustment. The opposite was true for sites in Ontario. CONCLUSIONS: Risk adjustment is essential when comparing quality of care across providers when home care agencies provide services to populations with different characteristics. While such adjustment had a relatively small effect for the two regions, it did substantially affect the ranking of many individual home care providers.

Short TitleBMC Health Serv ResBMC Health Serv Res
Alternate JournalBMC Health Serv Res