Duke University Press
Karen Davis - Can the Market Ensure Quality without Government? - Journal of Health Politics, Policy and Law 24:5 Journal of Health Politics, Policy and Law 24.5 (1999) 1127-1135

Can the Market Ensure Quality Work without Government?

Karen Davis

The debate about the quality of care in managed care plans is poorly framed. The question typically posed is whether quality is better or worse in managed care than in fee-for-service care. The more productive question is why quality varies across managed care plans, and how quality performance can be improved. Central to that question is whether market forces, professional standards, and industry self-regulation are sufficient to assure and improve quality or whether government has a role in setting a market framework for managed care.

The Clinton health reform proposal was designed to move the U.S. health system toward a system of competing managed care plans within a framework of government rules on plan information, choices, quality, and premiums. In addition, the Clinton proposal provided assurances of universal health insurance coverage and support for public goods such as research, health professions education, specialized care, and essential community providers. For better or worse the failure to reach consensus on the appropriate role for government was followed by rapid growth of enrollment in managed care plans--without an agreed-upon policy framework for the evolution of the industry.

Taking stock of managed care market trends is especially instructive now. The current public policy debate is focused on a federal government role in assuring managed care consumer protection and rights. In addition, the federal government in its purchaser role is increasingly proactive in setting managed care information and quality standards. [End Page 1127]

Quality Performance

Syntheses of research on whether managed care leads to better or worse quality of care yield mixed results. The most recent review (Miller and Luft 1997) of thirty-seven studies published in peer-reviewed journals between 1993 and 1996 included studies focusing on clinical quality of care and patient-reported experiences with care. Studies were equally divided between those finding clinical quality of care to be significantly better and those finding it to be worse in health maintenance organizations (HMOs) than in non-HMO plans. Two research projects representing a total of seven studies with exceptionally strong methodologies found worse results for chronically ill Medicare patients. All of the studies were based on data from 1993 or earlier--before the most recent rapid growth in managed care plans and enrollment.

Studies focused on patient-reported experiences typically have more current data and also tend to find that overall satisfaction is lower among managed care enrollees. HMO enrollees are more dissatisfied with aspects of care but more satisfied with financial aspects of coverage. Studies of Medicaid managed care enrollees are more mixed, with some studies finding Medicaid fee-for-service care ranked below that of Medicaid managed care (ibid.).

Surveys of patients and physicians supported by the Commonwealth Fund over the last five years also yield systematic evidence of poorer experiences with managed care. The 1995 Commonwealth Fund Survey of Physicians' Experiences with Managed Care, with a national sample of 1,700 practicing physicians, found respondents with more extensive managed care patients were more likely to be dissatisfied with the practice of medicine, dissatisfied with their ability to make the right care decisions for patients, their ability to get approval for care, and limits on referrals to specialists of their choices (Collins, Schoen, and Sandman 1997). Interestingly, these results did not apply to physicians practicing in group or staff model HMOs whose experiences were more similar to fee-for-service physicians with low managed care patient loads.

The 1994 Commonwealth Fund Survey of Patient Experiences with Managed Care of 3,347 working adults in Boston, Miami, and Los Angeles found higher levels of patient dissatisfaction and lower ratings of quality of care for those enrolled in managed care than those enrolled in fee-for-service plans (Davis, Collins, and Schoen 1995). Patients were particularly dissatisfied about waiting time for a routine appointment, access to specialists, access to emergency care, choice of doctors, time [End Page 1128] physician spent with them, physician being accessible by phone, and physicians' listening to their concerns. Patients with a family member in fair or poor health, a serious illness in the past year, or a chronic condition were particularly likely to rate their plans as fair or poor in providing access to specialists (Davis and Schoen 1996). Again, results varied widely across managed care plans. Group and staff model HMOs are rated more highly than individual practice association (IPA) or network model HMOs and preferred provider organizations (PPOs) (Schoen and Davidson 1996). Nonprofit HMOs are rated more highly than for-profit HMOs. Kaiser/Commonwealth Fund surveys of 10,000 low income adults in five states similarly yielded poorer results for managed care enrollees in both Medicaid and private insurance coverage compared with fee-for-service coverage (Lillie-Blanton and Lyons 1998).

The literature on variations in quality of care and patient-reported experiences with care across managed care plans is still in its infancy. The National Committee for Quality Assurance's annual report on managed care quality (NCQA 1998a) found that the immunization rate for children was 83 percent for plans at the 90th percentile, and 47 percent for plans at the 10th percentile. Use of beta-blockers following a heart attack ranged from 93 percent for the 90th percentile plan to 53 percent for the 10th percentile plan. The NCQA report found systematically better results for accredited plans than nonaccredited plans, and for plans willing to make their data publicly available. Plans in New England generally outperform other areas of the country.

While the research is in an early state, the finding of poorer performance of for-profit plans compared to nonprofit plans and IPA/network HMO/PPO plans compared to group/staff model HMOs is disturbing. It is also disturbing that plans are increasingly reluctant to make data publicly available. Twenty out of 270 plans reporting in 1996 declined to make information available on a plan-identifiable basis. In 1998, 155 out of 447 plans declined to make information publicly available (ibid.).

Market Trends

Managed care is often interchangeably referred to as "consumer choice" and "market competition." Yet, it is ironic how little choice and competition actually exists. Fee-for-service care is the ultimate in choice and typically open to all licensed physicians and accredited hospitals. When families changed from one indemnity health insurance plan to another, it did not affect where they went to get care. HMOs and PPOs, by contrast, [End Page 1129] have a defined list of providers, and patients must either pay a portion or all of the cost of care obtained outside of this list. Nearly 80 percent of those changing plans do so involuntarily either because they change jobs or employers change plans (Davis and Schoen 1998). About one-third of those changing plans reported having to change physicians as well.

But even the argument that families will have a choice between managed care and fee-for-service plans or among a selection of managed care plans is largely fiction. Only 40 percent of American workers are given a choice of two or more plans by their employers, according to the 1997 Kaiser/Commonwealth Fund National Survey of Health Insurance (ibid.). Employees of smaller firms and lower wageworkers are less likely to have a choice. Even when employers offer a fee-for-service choice, employee premiums are often prohibitive. As a result, lower-income working families are more likely to join managed care plans than higher-income families (Davis, Collins, and Schoen 1995).

Choice is a critical factor in patient experiences with managed care. Patients with a choice of plan are much more likely to rate plans high on quality, choice of physicians, access to care, and physician responsiveness to their concerns. Those in managed care without a choice of an alternate plan were nearly twice as likely to be dissatisfied with access to specialty care, availability of emergency services, and waiting time for appointments (ibid.).

To date, concern with quality does not appear to be shaping the nature of the managed care industry. Only selected large employers, such as General Motors, Ford, Xerox, GTE, and IBM, refer to reports on plan quality when selecting plans. Other employers give little weight to quality in selecting the plan or plans offered to employees. Only 9 percent of employers with two hundred or more employees require NCQA accreditation, and only 6 percent use Health Plan Employer Data and Information Set (HEDIS) data in selecting managed care plans (Gabel, Hunt, and Hurst 1998).

Employees also are unlikely to make decisions based on quality when choices are available. Only 1 percent of medium- to large-size employers provides HEDIS data to their employees to assist them in plan selection (ibid.). Even when employees are given report cards with systematic quality information, their impact is limited. An evaluation of report card initiatives in Denver and St. Louis found that only half remembered seeing the report card and approximately one-quarter of employees read most or all of the report cards provided to them (NCQA 1998b). However, [End Page 1130] of those who used the report cards, large majorities found them to be trustworthy, helpful, and understandable.

Perhaps it is not surprising, therefore, that plans rated highest on quality indicators are not among the most rapidly growing plans. Between 1988 and 1997, membership in for-profit HMOs increased from 47 percent of all enrollees to 62 percent (Gabel 1997; InterStudy 1997). This is not because they are more efficient or have lower premiums. Rather, they are more aggressive in marketing, building provider networks, and require less capital (Gabel 1997). Similarly, the most rapidly growing forms of HMOs were network, IPA, and mixed model HMOs. The share of enrollment in group and staff model HMOs declined from 43 percent in 1988 to 13 percent in 1997.

Competition among plans is also much more limited than commonly believed. Significant mergers and acquisitions have occurred in the last decade. Between 1987 and 1997, there were 162 mergers or acquisitions involving HMOs (Srinivasan, Levitt, and Lundy 1998). By 1998 the ten largest managed care companies accounted for almost two-thirds of all enrollment nationally, and three-fourths of Medicare managed care enrollment (MedPAC 1998). Four firms accounted for 47 percent of all managed care enrollment and 41 percent of Medicare managed care enrollment.

Government Role

Without a system in place to hold managed care plans accountable for quality health care, a market-driven environment may not appropriately balance cost and quality. To ensure that the market works for the health and quality of life for all Americans, it is important to develop new "rules of the game" that reward plans that act in the patient's and public's interest. As first steps toward holding managed care plans accountable, policy makers will need to innovate in at least seven basic areas:

Minimum Quality Standards

The United States does not leave issues of public safety solely to the market. Airlines, automobiles, and food, for example, must meet safety standards. We do not assume that information and consumer choice in these markets will be sufficient to avoid the loss of human life. Yet, there are no national standards on quality of care applicable to managed care plans. Accreditation by an approved private accrediting body is voluntary [End Page 1131] for managed care plans. Approximately 330 managed care plans out of over 650 plans are accredited by the National Committee for Quality Assurance (NCQA) or are in various stages of review and approval.

Information Reporting and Disclosure

While the art of measuring quality of care is still evolving, the NCQA HEDIS quality indicators and the Consumer Assessment of Health Plan Survey (CAHPS) now offer standardized leading measures of both clinical quality and patient experiences with care. Audited, timely information on these measures by name of plan should be available to patients and to purchasers such as employers and public programs such as Medicare, Medicaid, and the federal employees health benefit program. Information on plan performance, including quality information, participating physicians, hospitals, and other providers should be publicly available. Public disclosure of grievances, complaints, appeals and arbitration results, and administrative fines also would encourage plans to avoid serious complaints and to expedite the resolution of problems. The creation of the National Forum for Health Care Quality Measurement and Reporting and the Advisory Council for Health Care Quality are important first steps in developing a national framework for quality measurement and reporting, endorsing core sets of measures, planning methods for how quality data should be publicly reported, and setting a research and development agenda for the nation.

Access to Essential Care

Since the need for health care can be serious and time-sensitive, patients should have the right to seek care in an emergency without prior approval. Patients should also have the right to appeal denied services promptly, including a right to expedited review for life- or health-threatening conditions. This right should include the ability to go outside the plan to a neutral body. Physicians should be able to advocate for necessary care for patients without fear of reprisal.

Government as Purchaser

The government is a major purchaser of health care. Seventy million Americans are covered by Medicare or Medicaid. Almost half of all health spending (46 percent) comes from federal, state, or local governments. [End Page 1132] In its role as purchaser, government has a special responsibility to set quality standards on participating plans, provide choices among plans and information on quality to beneficiaries, ensure fair marketing practices, and provide financial incentives to plans that reward rather than penalize high quality care, especially care to seriously or chronically ill patients. Medicare has been a leader in requiring plans to collect and report quality information, auditing quality data, making quality information publicly available, and assuring beneficiaries choices among plans.

Public Goods

Recent research suggests that managed care squeezes cross-subsidies out of the health care system that have traditionally supported care to the uninsured and public goods such as research, education, and specialized care (Weissman et al. 1999; Cunningham et al. 1999). Either managed care must contribute to these public goods, or direct methods of financing must be developed.


The U.S. Agency for Health Care Policy and Research has been a leader in the development of quality measurement. But a more extensive investment in research on the determinants of quality variation, benchmarking, best practices, and effective strategies for quality improvement is needed.


Given the high and increasing degree of market concentration in the managed care industry, attention needs to be given to the potential for anticompetitive behavior. A few national managed care plans account for a significant share of all enrollment. Both the market and political implications of this concentration in a one trillion-dollar industry merit attention.

The jury is still out on the nation's experiment with managed care. Its promise includes the potential for improving the efficiency and effectiveness of care including slowing the growth in health care costs, improving the coordination of care and management of disease, redressing the imbalance between primary and specialty care, and encouraging prevention and health promotion. [End Page 1133]

At its worst, however, managed care could become simply a tool for cutting costs regardless of access and quality concerns. The financial incentives associated with managed care encourage providers to deny services and to trim corners, such as foregoing specialist care even when necessary. The pressure to lower costs, either from employers, government programs, or from shareholders in publicly held managed care plans, can lead to substandard care for patients.

Without a policy framework for competition within the managed care market place, quality of care and access to needed care are central concerns, especially for those who are most vulnerable: the poor, the elderly, and the seriously ill. Although the risks of market-driven changes are real, the advantage of managed care is its ability to be held accountable for care. But the market is unlikely to do so without a significant role for government in setting the rules of the game.

The Commonwealth Fund

Karen Davis is president of the Commonwealth Fund, a national philanthropy engaged in independent research on health and social policy issues. She has a Ph.D. in economics from Rice University and is the author of numerous books and articles on health care financing and delivery. Davis is a member of the Council of the Institute of Medicine and the Kaiser Commission on Medicaid and the Uninsured. She serves on the Board of Directors of the Mount-Sinai-NYU Health System and the Mount Sinai Medical School.


Collins, K. Scott, C. Schoen, and D. Sandman. 1997. The Commonwealth Fund Survey of Physician Experiences with Managed Care. New York: Commonwealth Fund, March.

Cunningham, P., J. M. Grossman, R. F. St. Peter, and C. S. Lesser. 1999. Managed Care and Physicians' Provision of Charity Care. Journal of the American Medical Association 281(12):1087-1092.

Davis, K., K. Scott Collins, and C. Schoen. 1995. Choice Matters: Enrollee Views of Managed Care and Fee-for-Service Plans in Three Cities. Health Affairs 14(2):99-112.

Davis, K., and C. Schoen. 1996. Changing Health Care Systems and Access to Care for the Chronically Ill. Changing Health Care Systems and Rheumatic Disease. Washington, DC: National Academy Press.

------. 1998. Assuring Quality, Information, and Choice in Managed Care. Inquiry 35(2):104-114.

Gabel, J. K. 1997. Ten Ways HMOs Have Changed during the 1990s. Health Affairs 16(3):134-145.

Gabel, J. K., K. Hunt, and K. Hurst. 1998. When Employers Choose Health Plans: Do NCQA Accreditation and HEDIS Data Count? New York: Commonwealth Fund, September.

InterStudy Competitive Edge. 1997. HMO Industry Report 7.2. St. Paul, MN: Decision Resources, September.

Lillie-Blanton, M., and B. Lyons. 1998. Managed Care and Low-Income Populations: Recent State Experiences. Health Affairs 17(3):238-247.

Medicare Payment Advisory Commission (MedPAC). 1998. Health Care Spending and the Medicare Program: A Data Book. Washington, DC: MedPAC, July.

Miller, R. H., and H. S. Luft. 1997. Does Managed Care Lead to Better or Worse Quality of Care? Health Affairs 16(5):7-25.

National Committee for Quality Assurance (NCQA). 1998a. The State of Managed Care Quality.Washington: NCQA, http://www.ncqa.org.

------. 1998b. Do Consumers Use Health Plan Report Cards? Washington: NCQA, December.

Schoen, C., and P. Davidson. 1996. Image and Reality: Managed Care Experiences by Type of Plan. Bulletin of the New York Academy of Medicine (suppl.) 73(2):506-

Srinivasan, S., L. Levitt, and J. Lundy. 1998. Wall Street's Love Affair with Health Care. Health Affairs 17(4):126-131.

Weissman, J. S., D. Saglam, E. G. Campbell, N. Causino, and D. Blumenthal. 1999. Market Forces and Unsponsored Research in Academic Health Centers. Journal of the American Medical Association 281(12):1093-1098.