Addressing Racial and Ethnic Disparities in the Context of
Medicaid Managed Care: A Six-State Demonstration Project

Methodology

This demonstration project was designed to determine whether health plans can use data on race/ethnicity, not only to identify disparities in quality of care, but to organize QI initiatives, working in coordination with the state Medicaid agency and with other health plans, to reduce or eliminate disparities and assess the impact of those initiatives.

The project started in the fall of 2002. Presentations about the project were made to a joint meeting of the Medicaid Managed Care and Quality of Care Technical Advisory Groups (TAGs) in Baltimore, MD, in August. Letters of invitation that included a project description were sent to the Medicaid directors in all 50 States and the District of Columbia, with responses requested by December 1, 2002. A stipend of $4,000 was offered to each participating Medicaid program and each participating managed care plan to help offset costs of data analysis and QI project development.

Medicaid programs in six States (Washington, Oregon, Montana, Texas, Michigan, and Virginia) accepted the invitation to participate, and they in turn invited managed care plans in each State to participate. The Medicaid programs were free to invite managed care plans on the basis of whatever criteria they wished to use. We did, though, provide a list of suggested criteria in areas like size of plan membership, racial/ethnic diversity of plan membership, data analysis capability, QI project capability, and history of successful work with the State or with other managed care plans on collaborative projects. The list of plans participating in each state is presented in Table 1. A total of 12 health plans (plus the Primary Care Case Management program in Montana) were involved. The full list of project participants is included in Appendix A.

Table 1. States and Managed Care Plans Participating in the Demonstration Project

Medicaid program staff in each State provided data on race/ethnicity from eligibility/enrollment files to participating managed care plans in January of 2003. Health plans merged these data with membership files or HEDIS data files to produce quality of care reports stratified by race/ethnicity. In some plans, this process took longer than in others, and one or two of the participating plans had actually done a similar analysis in 2001 and/or 2002, so they had a base of experience from which to work.

For some health plans, the denominator populations in specific HEDIS Effectiveness of Care Measures (e.g., beta blocker use after heart attack) were too small to permit meaningful comparison of performance rates across specific racial/ethnic minority groups. NCQA guidelines for minimum sample size requirements (generally at least 30 patients in a denominator population) were used to determine which rates would be calculated and/or included in subsequent analyses. Some plans, then, were able to examine virtually all HEDIS measures to identify possible disparities; other plans were able to examine only those measures for which a large fraction of the plan membership was part of the denominator (e.g., childhood immunizations, diabetes care).

Plans were required to choose one of the areas in which disparities were found and organize at least one QI intervention designed to reduce or eliminate that disparity. State Medicaid program staff were involved in this decision, and plans within a given state were encouraged to choose the same measure of disparity if possible (e.g., glycemic control in diabetes) so that they could take advantage of opportunities for collaboration. When possible, plans were also encouraged to choose clinical areas and measures already identified in the State as a high-priority area for quality improvement.

Some plans that had already made a commitment to large QI or DM initiatives in areas like asthma care or diabetes management focused their analysis of HEDIS measures on those specific clinical areas. If disparities were found in those areas, then plans were encouraged to link a disparities QI or DM initiative to the larger QI or DM programs already ongoing in those areas.

We did not impose any specific quality improvement model or specific parameters on the QI projects. Projects could include patient education, provider education, case or disease management, process improvement, or other general types of interventions. The choice of approach in each plan was driven by a combination of local experience, local knowledge of underlying reasons for disparities in quality, availability of staff and resources, opportunities for collaboration, suggestions from project staff and consultants, and opportunities for community involvement and partnership.

After a two-day training session co-sponsored and funded by the Center for Health Care Strategies (CHCS) in June of 2003, QI projects were organized and implemented in most plans starting in September of 2003. The project funding ended on June 30, 2004 and all plans were asked to provide reports on data analyses to assess project effectiveness as of that date. (We have been told that these QI projects and other disparity-related initiatives will continue in many of the participating health plans beyond that date, so some additional evaluation data will become available in the future.)

Monthly conference calls were organized by core project staff, and plans had the opportunity to provide updates, ask questions, and identify problems or barriers during those calls. The calls provided an opportunity for staff in the participating plans to help each other with suggestions on how to organize projects, identify members of key target populations, gather data for analysis, and make decisions about project success. Although we did not formally evaluate the value of these calls to plans during the project, participation was generally good and plans frequently reported that ideas shared, or resources identified, were helpful to them in organizing their QI projects.

Project staff and consultants were available as needed to participating States and health plans. Staff and consultants reviewed draft written materials, reviewed preliminary data analyses, identified sources of relevant patient education or provider education materials, and offered suggestions about potential choices of target areas of QI work when more than one disparity in quality was found. In one instance, project staff helped a participating plan apply for additional funding for its DM/QI initiative.

(Back to: Table of Contents)