Medicaid managed care is not fulfilling its promise

The Expanding Role of Managed Care in the Medicaid Program

By Kyle J. Caswell, PhD, Sharon K. Long, PhD, The Urban Institute, Washington, DC
Inquiry, April 16, 2015


States increasingly use managed care for Medicaid enrollees, yet evidence of its impact on health care outcomes is mixed. This research studies county-level Medicaid managed care (MMC) penetration and health care outcomes among nonelderly disabled and nondisabled enrollees. Results for nondisabled adults show that increased penetration is associated with increased probability of an emergency department visit, difficulty seeing a specialist, and unmet need for prescription drugs, and is not associated with reduced expenditures. We find no association between penetration and health care outcomes for disabled adults. This suggests that the primary gains from MMC may be administrative simplicity and budget predictability for states rather than reduced expenditures or improved access for individuals.

From the Discussion

This study finds that increased MMC penetration in a county is associated with an increase in the probability of an ED visit, and reported difficulty seeing a specialist and unmet need for prescription drugs among nonelderly non-SSI Medicaid adults. Furthermore, we find no evidence of reduced expenditures associated with increased MMC penetration for the non-SSI population. For nonelderly SSI Medicaid adults, we observe no consistent evidence that MMC penetration influences medical care access, use, or expenditures. We do not, however, interpret this as evidence that MMC has no impact on the outcomes examined. Our SSI sample is small, raising concerns about the precision of our estimates for this subpopulation, a limitation of the MEPS sample size.

Overall these results, especially among non-SSI Medicaid enrollees, seemingly contradict conventional theories on the expenditures and benefits of managed care. That is, managed care is generally intended to provide access to appropriate care in a timely, efficient, and cost-effective way by shifting the locus of care from higher cost settings to primary care. If successful, one would expect higher levels of primary care use, lower levels of specialist use that could be provided in primary care settings, lower levels of unmet need for care, and lower levels of inpatient stays and ED visits for ambulatory care sensitive conditions, and, as a result, reduced expenditures associated with those changes. Nonetheless, incentive structures created by managed care organizations may not always result in intended/desired outcomes.

There are characteristics of states’ Medicaid programs that may lead to outcomes that are inconsistent with conventional managed care theory. The largest factor, perhaps, is the programs’ low reimbursement rates. Low rates, combined with incentives to reduce expenditures within managed care organizations, could lead to reductions in appropriate access to care and/or quality of care should it limit the number and/or quality of providers available (e.g., narrow limits on number or type of prescription drugs or specialist care). Furthermore, reductions in appropriate care could eventually lead to more high-cost care (e.g., inpatient and/or ED care).

Our findings for the non-SSI population will help to inform the trend toward medical homes and accountable care organizations (ACOs). That is, to the extent that medical homes and ACOs in Medicaid are based on the MMC model, the benefits generated from these alternatives may also be limited. Silow-Carroll et al., however, highlight that the ACO model with its focus on greater integration and coordination of care, and greater emphasis on high-risk individuals, could offer greater improvements in care than have been generated by the traditional MMC model to date. These authors also note that many managed care organizations, including MMC organizations, are shifting toward the ACO model. Consequently, if the ACO model is more successful in achieving its intended benefits than the current MMC model, such a move could improve MMC. Going forward, it will be important to study whether ACOs and medical homes, whether stand-alone models or models developed as part of MMC plans, can produce the improvements over fee-for-service Medicaid that to date have largely proven elusive with MMC.



By Don McCanne, M.D.

Most states are moving most of their Medicaid patients into Medicaid managed care programs. As of 2011, 74% of Medicaid patients were in Medicaid managed care organizations (CMS data). The states contend that this shift will improve care coordination while reducing costs. What does this report from The Urban Institute show us?

The study was limited to nonelderly Medicaid beneficiaries. The study was not powered to determine the impact on the disabled, but conclusions can be drawn for the nonelderly nondisabled. For this group, increased Medicaid managed care penetration is “associated with increased probability of an emergency department visit, difficulty seeing a specialist, and unmet need for prescription drugs, and is not associated with reduced expenditures.” Based on these criteria, care was less coordinated and costs were not reduced. Some would consider this massive experiment to be a failure.

Although conclusions from this study cannot be applied to the elderly nor to the disabled, it seems intuitive that a program that is not working for younger, healthier individuals would not work any better for the elderly and disabled. However, the authors do suggest that greater attention to integration and coordination through medical home or ACO models could improve care for high-risk patients. Integration and coordination theoretically should be beneficial.

The management of care has been turned over to 267 mostly for-profit Medicaid managed care organizations (2014). These organizations are driven by the bottom line so the additional costs of improving integration and coordination remain targets for reducing overhead. They are businesses first, providing a market product - health care.

How could a single payer system improve on this? Imagine federally qualified health centers functioning as medical homes or ACOs. Integrating and coordinating care is part of their mission. Other private primary care groups in the community also serve as coordinators of care, ensuring access to specialists, hospitals, and other facilities and services. These are service organizations first, providing a crucial community service - health care.

These entities would be providing care for everyone on the same basis; the Medicaid designation would disappear. That is, payments would be going directly to the health care delivery system rather than to the wasteful tangled mess of intermediary money managers.

Kip Sullivan provides an excellent discussion of this topic at the PNHP Blog: