The Future of Children, Princeton - Brookings: Providing research and analysis to promote effective policies and programs for children.

Though the evidence does not suggest that managed care has been very successful in achieving its goals of improving health care at less cost, this does not mean that managed care cannot be successful. Rather, more study and experimentation may be necessary and, in particular, more attention will have to be paid to the variety of operational details which may influence the relative success or failure of different managed care arrangements. In addition, it may be necessary to make changes in the larger health care environment within which managed care plans function (for example, moving to a managed competition model) in order to realize the full potential of managed care.⁵⁸

Among the espoused advantages of managed care is that it improves access to primary care physicians, especially for individuals whose access is limited, while providing greater continuity of care for people who already have access. Greater access and continuity are believed to lead eventually to more healthful behaviors, more efficient use of the health care system, and lower costs in the long run.⁵⁹ These benefits may be realized for certain people who do not have access to physicians and join a managed care plan to improve their access. Others may not join because they do not want to leave a physician they already like. This phenomenon may, however, be diminishing because the newer IPA and PPO plans often allow people to “roll over” their choice of physician, that is, to stay with the same physician whom they had under fee-for-service.

Being able to roll over also has implications for selection bias and, therefore, for the outcomes of managed care. Individuals who are part of an established doctor-patient relationship may be sicker than those without a regular source of care. The ability to stay with the same physician may make joining a managed care plan more attractive, mitigating the effects of preferred selection. In a recent study of employees who were required to select from among HMO, PPO, and fee-for-service plans underwritten by the same insurance company, favorable selection was strongest for the HMO and less for the plans that permitted rolling over.⁶⁰ In the Rochester HMO study, children who were high utilizers of acute care and had established referrals to subspecialists were more likely to join the IPA which permitted rolling over than to remain in fee-for-service.²⁶ Individuals in the MCD who rolled over reported significantly higher rates of satisfaction with care than did individuals who had no identifiable usual source of care prior to the demonstration. They also used the emergency room less.^{61, 33} In another MCD substudy, children in Jackson County, Missouri, who stayed with the same physician by selecting a fee-for-service case management program had 62% fewer visits to specialists and 60% fewer visits to hospital emergency departments.³³ Thus, it appears that there may be important benefits to health reform plans which do not break existing doctor-patient relationships.

Quite apart from the issue of whether preferred or adverse selection pertains to the managed care versus fee-for-service choice, there is the question of whether managed care performs better when enrollment is mandatory or when it is voluntary.

Thus far, the evidence is quite limited. In the RAND HIE, although no significant differences in physiological outcomes were observed between children assigned to the HMO and those in the fee-for-service plan, children in the fee-for-service plan were perceived by their parents to be in better health.²⁴Although the general health of the children randomly assigned to the HMO (a mandatory enrollment group) and of a comparison group of children already enrolled in the HMO (a voluntary enrollment group) were the same, parents in the mandatory group expressed more concern about the health of their children than did parents in the voluntary group.²⁴ These findings suggest that parents may obtain some benefit from being able to choose between fee-for-service and managed care. In addition, there is the evidence from the Medicaid evaluations, previously mentioned, that satisfaction with care, while still high, is lower for people in managed care than for those in fee-for-service plans.³⁴ This fact suggests that satisfaction may be somewhat lower in a mandatory system.

Second, with regard to health care costs and the use of services, there appear to be no important differences attributable to enrollment status in the 25 evaluations reviewed by HFP.¹⁸ This finding contradicts an underlying tenet of the managed competition reform model which holds that the effects of managed care on cost and use will be stronger in markets where the penetration of managed care plans is greater.⁵⁸

Although there may now be insufficient evidence to judge the relative advantages of voluntary or mandatory enrollment, the conditions of enrollment may be an important issue in any health care reform plan which relies heavily on a system of managed care plans.

One of the most contentious issues in managed Medicaid regards the formulas that states use to set capitation rates.^{31, 62} A capitation rate that is set improperly can adversely affect the ability of managed care to save money and the operation of the health care marketplace itself.

Capitation rates should be set so as to be adequate to cover the costs of a plan that takes good care of its patients efficiently and to allow such plans a little surplus. With the caveat that it has been nearly impossible to observe or calculate what constitutes an efficient health plan or to observe plan profits, it is always true that plans have incentives to enroll primarily the good risks, those individuals whose treatment expenses will be minimal. The challenge, then, is to set the capitation rate in a way that compensates plans for enrolling poor risks. This can be especially important when considering how children and pregnant women fare under managed care. For example, children who are chronically ill will be expensive to treat. Managed care plans will try to avoid enrolling them, or may undertreat them, unless the capitation rate is substantial enough to cover the costs of treatment.

It is theoretically possible to adjust a capitation rate to reflect the health status of the population that a plan enrolls in order to mitigate both selection effects and the potential to underprescribe care. Actuaries, for example, could determine what additional monies might be necessary to compensate HMOs for children with severe chronic illness.⁶³ Some of the most relevant research on rate setting has been done in the context of the Medicare managed care program. It has focused primarily on ways to use enrollees' demographic characteristics and history of prior use of health services to adjust capitation rates paid to Medicare HMOs. There are still a number of technical issues to address in developing these adjustments, and in particular, procedures that work well for the Medicare population may not work as well for children and pregnant women, whose health care needs are frequently quite different from those of the elderly.⁶⁴