The aim of this study was to create new measures of quality that combine individual service measures. Using an all-or-none approach, we identify 5 levels of care reflecting the extent to which optimal patterns of service were obtained by patients with asthma, diabetes, and heart failure. We also assess the feasibility of these levels-of-care measures and their potential value in quality improvement efforts. The study was designed to analyze claims data to reflect patterns of services used in a single metropolitan market of about 1 million residents in the northeastern United States. More than 80,000 patients insured over 4 years (1994–1997) had claims with 1 or more of 3 chronic conditions. The analysis showed that the measures discriminated effectively among groups of patients with the 3 chronic conditions and highlighted areas to target quality improvement efforts. Although the numbers vary by year, for two of the diagnoses, most patients were in the lowest categories (59%–75%), and for the third, 40% were in these categories. Few were in the highest category. Most patients were in the same category from one year to the next. The levels-of-care approach to quality measurement can help caregivers and policy makers find methods for avoiding unnecessary utilization and expenditures while raising—not lowering—the probability that utilization patterns will conform to condition-specific recommended care.
From the Strategy and Policy Department and Health Sector Management Program (Dr Davidson) and the Operations and Technology Management Department and Health Sector Management Program (Dr Shwartz), Boston University School of Management, Boston, Mass; and the Stanford Prevention Research Center, Stanford University School of Medicine, Stanford, Calif (Dr Stafford).
This project was funded by a grant from The Commonwealth Fund to John Snow Research and Training Institute, Boston, Mass. Earlier work on the data was supported by a grant from The Robert Wood Johnson Foundation to Beth Israel Deaconess Medical Center, Boston, Mass; Harriet Davidson, PhD, Principal Investigator. We appreciate the support and assistance of Anne-Marie Audet, Stephen C. Schoenbaum, Jim Knickman, Harriet Davidson, and Mayer B. Davidson. A version of this article was presented at the Annual Research Meeting of the Academy for Health Services Research and Health Policy; June 7, 2004; San Diego, Calif.
Corresponding author: Stephen M. Davidson, PhD, Boston University School of Management. 595 Commonwealth Ave, Boston, MA 02215 (e-mail: firstname.lastname@example.org).