19
Quality and Relative Resource Use for Patients with Diabetes
Study Design: Retrospective cohort study employing data from Healthcare Effectiveness Data and Information Set (HEDIS) for 407 commercial health plans during the 2009-2011 period. The RRU measures are based on standardized prices, which allows for the analysis to abstract from price differences across plans or geographic regions. The measures are also risk-adjusted and are categorized into medical services and ambulatory pharmacy. Quality is measured as an unweighted composite of the ten individual components that make up the HEDIS Comprehensive Diabetes Care measure. In addition to levels, first differences are analyzed to provide insight into how changes in quality are related to changes in RRU.
Methods: Correlations between quality and various relative resource use measures are estimated. In addition to overall estimates, the sample is disaggregated by year, plan type, and geographic region. Correlations are also estimated for year-to-year changes in quality and relative resource use.
Results: The overall results are generally consistent with previous findings of a mostly negative relationship between quality and relative resource use. However, the disaggregated results indicate nuanced associations. For instance, the relationship appears to vary by year. Also, while there is a positive relationship between quality and evaluation and management services for Preferred Provider Organizations, the relationship is negative for the inpatient quality measure. For ambulatory pharmacy RRU measure, the correlations for HMO and HMO/POS combined are negative.
The correlations disaggregated by geographic region indicate that there are relatively large negative correlations for the total medical RRU measure in the Chicago and New York regions. In Atlanta and Boston there is a positive correlation for the outpatient aspect of procedure and surgery services. The analysis of year-to-year changes indicates that changes in quality are negatively related to inpatient costs.
Conclusions: The findings suggest that, given quality and relative resource use are not necessarily positively related, it may be possible to improve diabetes care quality at minimal cost. However, the disaggregated results suggest that the potential to increase quality can vary significantly by resource type, plan type, and geographic region. Thus, the most effective policies to improve quality may vary by plan. Further research is needed to explore the causes of these differences and to more thoroughly control for the many factors that may influence the relationship between quality and cost.