Toward a 21st century quality-measurement system for managed-care organizations

Health Care Financing Review, Summer, 1995 by Rodney C. Armstead, Paul Elstein, John Gorman

Delmarva Contract

In order to move away from individual case review and toward analysis of patterns of care provided by HMOs, in September 1993, HCFA contracted competitively with the Delmarva Foundation for Medical Care, Inc. (Although Delmarva is the PRO for Maryland and the District of Columbia, it was not working as a PRO for the purposes of this contract.) Dr. R. Heather Palmer of the Harvard School of Public Health was the principal investigator for the project.

HCFA asked Delmarva to collaborate with a panel of QA experts from the managed-care industry and academia to develop a set of performance measures or quality indicators to identify the minimum data needed for the indicators and to develop a new review methodology for external (medical third-party) review.

The contract was intended to help HCFA shift from the current mode of HMO oversight to one based on measurement and improving the mainstream of care for entire populations. HCFA asked the contractor to survey performance indicators that are already in use, so that HCFA could build on the efforts of others and move expeditiously to a more state-of-the-art review system. Further, in the interest of minimizing burdens on managed-care plans, HCFA asked the contractor to use a panel of experts to evaluate which of the indicators currently in practice were most likely to be of value in quality management for Medicare risk-contracting HMOs.

In the team's final report of August 1994, Delmarva recommended three core measures to be drawn from the HMOs' administrative data. These core measures, which would apply to all Medicare enrollees in the HMO, included access to services (defined as one or more services from plan practitioners), annual influenza vaccination, and screening mammography for women.

Delmarva also recommended adopting two diagnostically related measure sets (DRMs) to measure clinical performance. A DRMS uses both administrative and medical record information to measure performance of care for patients with specific diagnoses or conditions. For the initial DRMs, Delmarva recommended diabetes mellitus and ischemic heart disease/hypertension. These conditions are common among Medicare beneficiaries. Among the measures that Delmarva suggested be included in those DRMS were leg and foot examinations for beneficiaries with diabetes, and blood pressure screening for patients with ischemic heart disease.

The measures were drawn from a number of existing data sets. The most prevalent were the Health Plan Employer Data and Information Set (HEDIS) and the Develop and Evaluate Methods to Promote Ambulatory Care Quality (DEMPAQ) project, an earlier HCFA-funded project to develop an approach to review fee-for-service care provided in physicians' offices.

Following release of the Delmarva report, HCFA asked PROs and HMOs whether they would participate in a pilot test of the proposed methodology. Because of the need to have a sufficient number of beneficiaries from which to draw our sample, HCFA required that each HMO have at least 5,000 Medicare enrollees in 1994. It was estimated that this number would provide the desired sample of 300 beneficiaries in each DRMS estimated for the project. In addition, HCFA asked that any PRO that wished to be considered for the pilot have at least two such HMOs in their State willing to participate. This limited the potential field of PROs to 17; of those, 10 came in with proposals meeting the criteria for consideration.