Prepared by Suzanne Felt-Lisk, Rebecca Dodge, Megan McHugh, Mathematica Policy Research, Inc.

November 2001

Executive Summary

From 1997 through 1999, Medicaid managed care enrollment continued to rise, even in the face of growing concern about the willingness of health plans to participate in state Medicaid programs. Overall, Medicaid enrollment in full-risk managed care organizations rose during the period by 21.3 percent. In 1999, 316 full-risk managed care plans served 11.4 million Medicaid enrollees in fully capitated programs in 45 states.

The characteristics of health plans serving Medicaid beneficiaries in full-risk programs vary considerably. At one end of the spectrum are large, commercial plans affiliated with national managed care companies. At the other are very small local plans often owned by hospitals or health centers that serve Medicaid beneficiaries almost exclusively. As an increasing number of commercial plans began to exit from the Medicaid market in 1997, concerns surfaced about where this new trend would lead:

• Would the departures continue and grow, or were they a temporary adjustment in the market?
• Would declining participation by commercial plans lead to the contraction or collapse of full-risk Medicaid managed care? Or would Medicaid-dominated plans fill the gap left by commercial plans exiting, and if so, what would this mean for beneficiaries?

This paper addresses these issues, building on previous work that describes the trends in commercial health plan participation in Medicaid managed care and the characteristics and roles of Medicaid-dominated plans through June 1997 (Felt-Lisk 1999; Felt-Lisk 2000). In addition to updating previously published information, this paper includes new analyses on the performance of Medicaid-dominated and commercial plans on selected HEDIS2 measures for Medicaid on the effectiveness of care and access to care, and on the extent to which health plans are restricting their Medicaid service areas at the county level within states as well as exiting from Medicaid in all counties in a state.

In brief, we find that during 199799, about the same number of commercial plans exited from state Medicaid markets in each year but at a much higher rate than in the years prior, with 15 to 17 percent of participating plans exiting annually. In 1997, commercial plan exits were concentrated in certain states and driven by the decisions of a few national managed care organizations (MCOs). In 1998 and 1999, the departures occurred more broadly nationwide. In addition, commercial plans rarely entered the Medicaid market in 1998, a major shift from previous years. By June 1999, only 37 percent of plans that are large and affiliated with a national MCO firm participated in Medicaid, compared with 56 percent in 1996. Similarly, only 41 percent of BlueCross/BlueShield plans participated compared with 59 percent in 1996. Our analysis of 21 states that account for almost 90 percent of full-risk Medicaid managed care enrollment suggests that the exits continued at a high pace, though with some abatement, through June 2000.

A total of 1.2 million Medicaid enrollees in commercial plans that exited during 199799 experienced, at best, the burden of selecting a new health plan and at worst, discontinuity of care if they were required to change providers. About an additional 600,000 enrollees were in plans that were consolidated with another plan that served Medicaid; we cannot assess whether the transitions during the consolidation period were seamless or not from the enrollees point of view. The total number of enrollees in commercial health plans that exited or consolidated increased by 140 percent from 1997 to 1998 (from 360,000 to 874,000), then dropped some, but remained at a relatively high level for 1999 (606,000).

Despite the large number of exits, the structure of Medicaid managed care remains relatively intact:

• Full-risk Medicaid managed care enrollment continued to grow nationally, from 9.4 million in 1997 to 11.4 million in 1999. The number of states with any full-risk enrollment rose from 43 to 45 during this period.
• Commercial plans throughout the nation still serve a majority of full-risk Medicaid managed care enrollees (58 percent in 1999). Our analysis of 21 high-volume Medicaid managed care states suggests that this figure dropped somewhat by mid-2000, but that it was still over 50 percent.
• The number of Medicaid-dominated plans and the size of their enrollment has grown since 1997, offering additional options for enrollees in commercial plans that have exited. As a group, these Medicaid-dominated plans also appear to have been financially stronger in 1999 than they were in 1997.
• Few counties that were part of a full-risk Medicaid managed care program in mid-1998 dropped out completely by mid-2000.

Policymakers have historically been somewhat reluctant to encourage the development of Medicaid-dominated plans because of quality of care concerns associated with the absence of a large commercial population that might enhance the pressure to perform well. However, an initial analysis of selected HEDIS performance measures for the Medicaid population does not show a difference between Medicaid-dominated and commercial plans on a majority, though not on all, of the indicators reviewed. However, additional analysis of quality of care is needed to confirm this finding, since data for the initial analysis were limited in several ways; for example, data did not cover a broad spectrum of care (most of the indicators available pertained to women and children's health, and focused on preventive services). (See Moreno et al., 2001 for another analysis of this topic using survey data.)

The study's findings suggest that the exodus of commercial plans from full-risk Medicaid managed care first identified in 1997 has expanded into a national phenomenon, continuing at least through mid-2000. For the hundreds of thousands of Medicaid beneficiaries enrolled in these plans, the exodus represents a disruption in care. However, possibly because of the efforts by states and many other organizations to respond to the effects of the exits, they do not appear to have crippled full risk Medicaid managed care programs. Our analysis of 12 high-volume Medicaid managed care states suggests that most beneficiaries still have at least one commercial plan option in their county.

Further, Medicaid-dominated plans which are inherently vulnerable because they depend almost entirely on revenue from the Medicaid program appear to be surviving and growing in general, and were at least breaking even financially during the study period.