The End of an Era: What Became of the “Managed Care Revolution” in 2001?

Abstract

Objective

To describe how the organization and dynamics of health systems changed between 1999 and 2001, in the context of expectations from the mid-1990s when managed care was in ascendance, and assess the implications for consumers and policymakers.

Data Sources/Study Setting

Data are from the Community Tracking Study site visits to 12 communities that were randomly selected to be nationally representative of metropolitan areas with 200,000 people or more. The Community Tracking Study is an ongoing effort that began in 1996 and is fielded every two years.

Study Design

Semistructured interviews were conducted with 50–90 stakeholders and observers of the local health care market in each of the 12 communities every two years. Respondents include leaders of local hospitals, health plans, and physician organizations and representatives of major employers, state and local governments, and consumer groups. First round interviews were conducted in 1996–1997 and subsequent rounds of interviews were conducted in 1998–1999 and 2000–2001. A total of 1,690 interviews were conducted between 1996 and 2001.

Data Analysis Methods

Interview information was stored and coded in qualitative data analysis software. Data were analyzed to identify patterns and themes within and across study sites and conclusions were verified by triangulating responses from different respondent types, examining outliers, searching for disconfirming evidence, and testing rival explanations.

Principal Findings

Since the mid-1990s, managed care has developed differently than expected in local health care markets nationally. Three key developments shaped health care markets between 1999 and 2001: (1) unprecedented, sustained economic growth that resulted in extremely tight labor markets and made employers highly responsive to employee demands for even fewer restrictions on access to care; (2) health plans increasingly moved away from core strategies in the “managed care toolbox”; and (3) providers gained leverage relative to managed care plans and reverted to more traditional strategies of competing for patients based on services and amenities.

Conclusions

Changes in local health care markets have contributed to rising costs and created new access problems for consumers. Moreover, the trajectory of change promises to make the goals of cost-control and quality improvement more difficult to achieve in the future.

As health care inflation reached double-digit rates of annual increase in the early 1990s, many touted the promise of managed care to foster a more efficient and effective delivery system. While many private employers moved their employees into managed care plans, the Clinton health plan of 1993 acted as a catalyst for a “managed care revolution,” raising expectations for rapid diffusion. Although by 1994 it became clear that the legislation would not pass, many in the health care industry quickly embraced the concept that markets would evolve into highly integrated delivery systems, and many of the changes anticipated under the Clinton plan were, in fact, set in motion (Ginsburg 1996; Shortell and Hull 1996).

Between 1990 and 1995, the number of public and privately insured enrolled in a health maintenance organization (HMO) grew from 36.5 million to 58.2 million, and by 1995, the majority of Americans with employer-based health insurance were enrolled in some form of managed care plan (Gabel 1997). These plans relied on selective provider networks and negotiated payment rates to help infuse greater cost-consciousness on the part of providers (Zwanziger, Melnick, and Bamezai 2000; Mobley 1998; Bindman et al. 1998). Some used risk-based payment arrangements to give providers financial incentives to manage service utilization (Conrad et al. 1998; Kralewski et al. 2000; Hillman, Pauly, and Kerstein 1989; Hellinger 1996; Kohn 2000; Grossman 2000). And many actively managed patients' access to care, using gatekeepers and prior authorization requirements (Rask et al. 1999; Kravitz et al. 1998; Meyer et al. 1996; Hurley, Fruend, and Gage 1991).

However, unlike the Jackson Hole Group model that served as the intellectual underpinning for much of the Clinton plan (Enthoven and Singer 1994), very few employers offered their employees a choice of plans with a fixed contribution. Many employers simply replaced traditional indemnity offerings with only managed care options, in some cases offering only a single option for coverage (Trude 2000). This approach disenfranchised consumers and contributed to the growth of intense consumer backlash that has sent managed care into retreat (Ginsburg and Lesser 1999; Lesser and Ginsburg 1999; Blendon et al. 1998; Robinson 2001). It also diminished incentives for health plans to compete on the basis of quality—one of the major promises of the managed competition theory (Hibbard et al. 1997; Maxwell, Temin, and Watts 2001).

Since 1996, the Center for Studying Health System Change's Community Tracking Study (CTS) has been following how the managed care revolution and the associated backlash have transformed local health care markets nationally. This paper provides an overview of major changes that have occurred in the most recent stage of the managed care era—1999 to 2001—and reflects on these changes in the context of expectations in the mid-1990s when managed care was in ascendance. The paper highlights conclusions of numerous other analyses based on the CTS data from 2000–2001 that detail changes in particular aspects of the health care system, such as how health plans' approaches to managing care have changed or how employers have responded to rising premiums (Draper et al. 2002 and Trude et al. 2002). Many of these more detailed analyses are included in this special issue and some have been published elsewhere.1 This paper takes a broad view of the health care market as a whole, discussing changes by purchasers, health plans, providers, and in public policy to identify the major ways in which local health care markets changed during this period. Ultimately, we conclude that although the managed care revolution has evolved quite differently than expected, it continues to play a major role in shaping local health care markets, although it does so with diminished potential to control costs and promote quality and in the process it creates some new access problems for consumers.

CONCEPTUAL FRAMEWORK

The American health care system relies on a mix of market forces, government regulation, professional norms, and organizational missions to shape access, cost, and quality of health care. Consumers' role in the health care system is complicated by a number of factors. First, the nature of health care services tends to leave consumers in a poor position to moderate their demand for services. Consumers rely primarily on physicians to inform them of the services they need; there is often limited information on the effectiveness of diagnostics and treatments; and there is general uncertainty about an individual's health care needs over time. The latter problem creates a market for health insurance, which in turn helps insulate consumers from the actual cost of care. Moreover, health insurance has evolved to be an important benefit employers offer to attract and retain employees, and employers commonly pay a significant proportion of employees' health insurance premiums, which further distances consumers from health care costs.

As a result, employers and the health insurance benefits they offer largely shape consumers' health care purchasing power. With advice from benefits consultants, interactions with brokers, and often directly with health plans, employers determine the array of services included in employees' benefit packages and the associated cost-sharing requirements, shaping what services consumers can afford. Employers make decisions about health benefits in the context of labor market conditions and the resources available to devote to employee compensation (Christianson and Trude 2003).

Government also plays a key role in shaping the structure and dynamics of the health care market, acting as both a purchaser and a regulator. Although government purchasing for government employees and public programs such as Medicare and Medicaid is analogous to private employers' purchase of health insurance, it also is shaped by a variety of distinct forces, including legal requirements governing social insurance programs, politics, and public accountability for the use of taxpayers' funds. The fact that government represents far more lives than any single employer typically covers also shapes its role as a purchaser (Watts et al. forthcoming).

As a regulator, federal, state, and local governments define the “rules of the game” for the health care market, addressing issues such as the number and type of providers allowed to operate in a certain market (i.e., certificate of need policy), the degree of consolidation in the market (i.e., antitrust policy), or the types of services or grievance procedures consumers are entitled to under their health insurance arrangements (i.e., patient protection policy). Government policy also ensures that health care organizations meet certain standards, such as financial solvency standards for entities that accept risk for health care services and licensure requirements for providers of care. Finally, government provides direct subsidies to certain providers to serve as a safety net for those unable to pay for medical care.2

Health plans shape the workings of health care markets in their role as intermediaries between public and private purchasers and consumers on the one hand and health care providers on the other. Health plans provide an insurance function for purchasers and consumers by forming risk pools to manage the costs of care across a large group of people. In addition, health plans act as bulk purchasers of services for their pool of covered lives. They are motivated by a desire to demonstrate value to their customers (employers, government programs, and by extension consumers) with strategies to manage the financial risk associated with unpredictable needs for medical care. Although traditional indemnity insurers approached this business primarily as claims payer, contemporary health plans have adopted a more proactive approach that involves negotiated payment, selective provider networks, and hands-on utilization management.

The providers of health care services—hospitals, physicians, and other caregivers—operate in a market shaped by these myriad forces. Health care providers are driven by a mission to provide necessary and appropriate care to their patients (sometimes defined by regulation, sometimes by professional standards and/or plan or purchaser performance standards, or combinations of these) and to generate revenue to support this mission.3 Typically, providers compete for patients by expanding the array of services they offer, keeping up with technology to support the ability to offer cutting-edge and high-quality medical care and increasing the visibility of the organization in the community.

METHODS

Data for this paper are from the Community Tracking Study (CTS), a longitudinal study that tracks changes in local health care systems nationwide. Researchers conduct site visits to 12 randomly selected and nationally representative communities every two years to interview leaders of the local health care system about changes in the organization, delivery, and financing of health care, and the impact of these changes on people.4 First round site visits were conducted in 1996–1997 and provide a baseline for tracking change. Subsequent visits were conducted in 1998–1999 and 2000–2001. A total of 1,690 interviews have been conducted in the 12 markets since 1996, with 895 interviews conducted in the most recent round.5

Depending on the size of the community, each visit consists of 50 to 90 interviews with representatives of all of the major stakeholders in each market, including health plans, hospitals, physician organizations, employers, state and local governments, and consumer groups. The sample includes representatives of the largest and most important organizations or groups in the market as well as “outliers” (e.g., small organizations or highly specialized organizations likely to have a different perspective on market developments). Such diverse samples are particularly useful for developing a balanced and nuanced perspective on market developments and their impact on particular groups and people (Miles and Huberman 1994; Patton 1990).

Semistructured interview protocols were developed to ensure data comparability across sites and over time. Protocols included a core set of questions used consistently over all the three rounds of site visits as well as questions for special study topics. The protocols primarily relied on open-ended questions that did not restrict the respondents' answers to a pre-specified set of categories, with some closed-ended questions for basic information (unavailable through other secondary sources) about the organization or group.

Researchers involved in developing the study topics and analyzing the data conducted all the interviews. A “matrix model” was used in the most recent round of visits in which each researcher served as a member of both a substantive research team focusing on a particular sector (i.e., policy, purchasers, health plans, providers) and a site visit team. Each site visit team included at least one researcher from each of the four substantive research teams, and this researcher conducted the majority of interviews with key respondents in their sector (e.g., plan team member conducted most interviews with health plan respondents). Most researchers visited three markets that varied on a number of important characteristics (e.g., population size, managed care penetration), facilitating cross-site comparison.

With the help of a note-taker who accompanied them during the interview researchers typed up notes from the interviews, and, for each site, wrote a detailed synthesis of the findings related to their research team's study topics. Both the research team and site team reviewed the syntheses for quality control.

All interview notes and syntheses were stored, coded, and analyzed in ATLAS.ti, a computer software program designed to support qualitative data management and analysis (Weitzman 1999; Muhr 1997).6ATLAS.ti was customized for the project with a detailed coding structure to track and analyze the data by question, respondent type, market characteristics, and the content of the interviewees' responses. ATLAS.ti allowed researchers to search the data using these codes or any combination of codes. For example, researchers could construct a query on all health plan executives' discussions of quality improvement efforts or all responses to a certain question or question series that involved a discussion of quality improvement. Researchers could then conduct these queries on important subsets of the data (e.g., markets with relatively high or low capitation) to see if respondents' answers were similar or different.

A variety of well-established techniques were used to draw and verify conclusions from the data (Devers 1999; Miles and Huberman 1994). Output from ATLAS.ti searches were used to identify patterns and themes, to contrast and compare by respondent type and site, to count response frequencies, to identify more abstract factors or concepts of importance, and to build a logical chain of evidence. To verify conclusions, we used a variety of techniques to assess data quality. For example, throughout data analysis we assessed the convergence or divergence of responses by different respondent types (e.g., employers compared to plans) and qualitative and quantitative data sources (e.g., respondents' view of HMO enrollment trends to quantitative data from InterStudy). We also assessed the strength of the patterns and themes by examining outliers (e.g., smaller plans, plans reporting significant enhancements to their quality improvement efforts), searching for negative or disconfirming evidence, and testing rival explanations.

RESULTS

Since the beginning of the CTS, managed care has not evolved as anticipated. In 1996–1997, HMO enrollment was on the rise, but pressures for looser forms of managed care were already evident (Christianson 1998, Grossman 2000), and some began to question if markets would evolve into the vertically integrated, tightly managed systems once envisioned (Burns et al. 1997; Robinson 1999). Two years later, growing anti-managed care sentiment intensified these pressures, sending managed care into retreat (Lesser and Ginsburg 1999). By 2000–2001, health care markets moved even further in this direction as three key developments dominated the landscape: (1) unprecedented, sustained economic growth and extremely tight labor markets made employers highly responsive to employee demands for even fewer restrictions on access to care; (2) health plans increasingly moved away from core strategies in the “managed care toolbox”; and (3) providers gained leverage relative to managed care plans and reverted to more traditional strategies of competing for patients based on services and amenities.

Employers Respond to Tight Labor Market

The economic boom of the late 1990s and early 2000s was still strong at the time of our 2000–2001 site visits. Corporate profits were soaring and the stock market value had reached an all-time high in March 2000. Perhaps most importantly from the perspective of the health care market, unemployment was just 3.9 percent nationally in October 2000, the midpoint of our field period.7 Notably, the phenomenon of tight labor markets was observed universally across our 12 study communities, with only a very narrow range of variation. The highest unemployment rate was observed in Miami, with a rate of 5.4 percent, while the lowest was just 2.4 percent in Orange County, California.8

As employers increasingly focused on attracting and retaining employees in a tight labor market, they continued to drive the market away from closed-model HMOs in favor of more loosely managed preferred provider organizations (PPOs), point-of-service (POS) products, and open-access HMOs (i.e., with no gatekeeper) that offered enrollees broad provider networks and limited restrictions on access to services (Christianson and Trude 2003; Heffler 2001; Gabel et al. 2001). Indeed, by 2000, HMO enrollment growth slowed significantly or began to decline in many of the CTS markets,9 and, given the changes occurring in the HMO product itself, the price differential between HMO and PPO/POS products diminished (Draper et al. 2002; Mays, Hurley and Grossman 2003).

At the same time, the health insurance underwriting cycle had bottomed out in 1998–1999 and plans began to raise premiums in 2000 and 2001 in excess of anticipated cost trends. In light of the tight labor market—and strong corporate earnings—employers proved willing to accept sizeable premium increases during this period without making major changes to cut costs (Christianson and Trude 2003; Trude et al. 2002). On average, employers nationally saw their premiums increase by 11.0 percent between 2000 and 2001, the highest rate of increase since 1993 (Draper et al. 2002; Strunk, Ginsburg, and Gabel 2001; Gabel et al. 2001).

In response, employers made only modest increases to employee cost-sharing requirements. Across communities, there was rapid diffusion of three-tier pharmacy benefits that require consumers to pay increasingly higher copays for preferred and other brand-name drugs (Mays, Hurley, and Grossman 2001). Some employers also adopted increased cost-sharing requirements for office visits, emergency department visits, and inpatient stays. But it was uncommon for employers to make any change to their premium contribution strategies during this period, and most maintained responsibility for a high percentage of employees' premiums (Trude et al. 2002).

Meanwhile, plans felt little pressure from purchasers to compete on the basis of clinical quality. Although many large employers had adopted value-based purchasing principles,10 lack of adequate information on clinical quality and administrative barriers to switching plans prevented employers from bringing clinical quality into the purchasing equation (Hargraves and Trude 2003). Moreover, many employers viewed the availability of broad provider networks as shifting the responsibility for selecting high-quality providers back on employees (Christianson and Trude 2003).

Continued Retreat from Tightly Managed Care

While employers continued to push health plans away from tightly managed products, the state and federal policy environment kept sustained pressure on the HMO product during this period, further driving plans away from core strategies in the managed care toolbox. States enacted extensive HMO regulation in the late 1990s, and although few new laws were passed in 2000–2001, consequences of the many new mandates were still just being realized during this period (Hurley and Draper forthcoming) while further regulation loomed with ongoing debate over a Patients' Bill of Rights at the federal level.

Selective provider networks became less and less viable in an environ-ment where broad provider choice was so highly valued and, by 2000–2001, few plans were actively constraining the size of their provider networks. In fact, differences between HMO and PPO provider networks steadily dimi-nished in most markets between 1999 and 2001 (Mays, Hurley, and Grossman 2003). To manage costs associated with such broad provider networks, some health plans were considering tiered provider networks, with higher cost-sharing requirements for certain providers (Mays, Hurley, and Grossman 2003).

Risk contracting arrangements—once viewed as a key mechanism to make providers more cost-conscious and promote quality improvement—declined in prevalence or scope in the majority of the study sites due to poor experience and diminishing feasibility given the way managed care products were changing. Hospital risk arrangements dropped most precipitously, with evidence of decline in all markets, including more advanced managed care markets such as Orange County (Mays, Hurley, and Grossman 2003). In contrast, physician risk contracting remained in use in 8 of the 12 study sites, although many eliminated risk for pharmacy or other services viewed as “uncontrollable” and a declining number of health plan members were covered under these arrangements (Hurley et al. 2002). At the same time, health plans had begun to experiment with new fee-for-service based payment arrangements that incorporated incentives for meeting certain utilization targets or clinical quality measures, such as benchmarks for surgical complication rates or for the provision of preventive and screening services (Mays, Hurley, and Grossman 2003; Felt-Lisk and Mays 2002).

Meanwhile, health plans increasingly reduced their emphasis on gate-keeping and utilization management during the study period. Seventeen of the 48 plans interviewed in 2000–2001 introduced new direct-access HMO pro-ducts that allow self-referral to specialists between 1998–1999 and 2000–2001 (Mays, Hurley, and Grossman 2003). And across all 12 study sites, most plans moved to eliminate or relax efforts to prospectively manage patients' use of services, abandoning pre-authorization requirements for hospital admissions, outpatient tests and procedures, and referrals to specialists. Instead, health plans focused on more targeted efforts, stationing utilization review nurses in hospitals to encourage expeditious discharges (Seattle, Washington, Orange County, California, and northern New Jersey) and expanding or refining disease management programs that target the highest-cost members (most markets) (Felt-Lisk and Mays 2002).

Meanwhile, with the turn in the underwriting cycle in 1998–1999, plans sharpened their focus on profitability. Continuing a trend that began in 1998–1999, plans dropped out of Medicare and Medicaid, as well as certain unprofitable commercial lines of business, such as the small group and individual markets (Draper et al. 2002). Like the experience nationally, plan exits in our study sites left consumers with fewer options (Gold and Justh 2000). Moreover, plans that remained in the Medicare market reduced benefits and increased cost-sharing requirements, increasing the financial burden on enrollees (Hurley, Grossman, and Strunk 2003).

Hospitals Exercise Leverage and Revert to Traditional Competitive Strategies

With managed care in retreat, hospitals improved their position considerably between 1998–1999 and 2000–2001, securing better contract terms from health plans and returning to an environment that allowed for more traditional strategies to compete for patients through services and amenities. Sustained demand from purchasers and consumers for broad provider networks diminished the threat of exclusion and gave hospitals greater clout in negotiations with health plans. In addition, years of mergers and acquisitions produced extensive consolidation among hospitals in our study sites and nationally (Irving Levin Associates 1999; Lesser and Brewster 2001), yielding significant potential negotiating power. Finally, the move toward less restrictive management of care increased demand for providers' services, and after years of downsizing excess capacity and accepting deep discounts to secure patient volume, many hospitals gained leverage due to newfound capacity constraints, although this also left many communities with overcrowded emergency rooms (Brewster, Rudell, and Lesser 2001; Devers, Brewster, and Casalino 2003).

Hospitals actively tested their new potential leverage in 2000–2001 as they struggled with rapidly rising underlying costs due to a severe labor shortage among allied health professionals (nurses in particular) and pharmaceutical and technological developments (Strunk, Ginsburg, and Gabel 2001). At the same time, implementation of the Balanced Budget Act (BBA) seriously strained hospital revenues. In this context, many hospitals were highly motivated to press health plans for better payment rates and better contract terms in aggressive negotiations. While disagreements about contracting arrangements are hardly new, the disputes observed in 2000–2001 were notable in two respects. Providers demonstrated a willingness to take their disputes to the brink, threatening to walk away from health plan networks if their demands were not met. Moreover, the organizations involved in these disputes often were among the largest and most prominent providers in the market, which meant that these decisions would affect a significant number of people in the community. Although providers tended to prevail in these disputes, concerns about network stability lingered in many markets (Short, Mays, and Lake 2001; Strunk, Devers and Hurley 2001; Devers, Brewster, and Casalino 2003).

Meanwhile, hospitals focused on expanding services and capacity to meet changing market conditions. Most de-emphasized physician integration activities pursued in anticipation of managed care growth (Lake et al. 2003) and reverted to a strategy of attracting patients and physicians (particularly specialists) by offering key services and amenities. There has been extensive activity to expand high-margin services such as cardiac, orthopedic, and oncology services and to establish niche services or centers of excellence, with a great deal of mimicking behavior within markets that suggests the potential re-emergence of a medical arms race (Devers, Brewster, and Casalino 2003). Relatively little attention focused on significantly improving quality during this period, other than some initial steps to address medical errors (Devers 2002).

Traditional safety net providers for the low-income and uninsured also expanded services during this period, although these providers tended to emphasize expansions of primary and preventive care. Hospitals added outpatient facilities to serve the low-income and uninsured, and the number and capacity of community health centers grew. Some communities also were successful in expanding the pool of providers available to the low-income and uninsured with new strategies for distributing funds for charity care. These developments, along with expanded public insurance options, helped the safety net remain stable or grow stronger in most of our study sites during the six-year period between 1996–1997 and 2000–2001 (Felland et al. 2003).

DISCUSSION

Changes occurring in local health care markets in 2000–2001 were quite different than those predicted when the “managed care revolution” began in the early 1990s. Health plans continued to shy away from aggressive utilization management, providers appeared to return to a medical arms race mentality, and premiums resumed double-digit annual rates of increase.

But while managed care remains in retreat, it continues to be a defining feature of the landscape. Risk contracting, for example, has fallen out of favor, but in most markets it is being modified rather than rejected in total. Similarly, efforts are underway to replace the initial blunt instruments for managing care with more sophisticated strategies to control utilization and manage chronic disease. And despite increasing breadth, provider networks continue to exist and in general prices are being set through negotiated contracts rather than the “usual, customary, and reasonable” fee schedules of the pre-managed care era. These activities suggest that managed care is not going away; rather it is evol-ving in response to prior experience and adapting to changing market demands.

For consumers, changes in health care markets between 1998–1999 and 2000–2001 have had mixed effects. In the short-term, consumers appear to have benefited enormously from the changes occurring in the health system. People with employer-based coverage enjoyed the relatively generous coverage that employers adopted in light of the tight labor market over the past few years. Although increased copays and deductibles left those who use services paying more, consumers remained largely sheltered from the growing cost of health insurance, while enjoying fewer restrictions on access to services. Indeed, the low-income and uninsured also generally fared well during this period when flush economic conditions helped expand public insurance and boost support for local safety net providers.

Yet, on the other hand, changes in the structure and dynamics of local health care markets resulted in some new types of access problems for consumers. Contentious contract negotiations became increasingly common, threatening much higher cost-sharing for consumers whose providers drop out of networks. Plans' continued exits from Medicare and Medicaid left consumers covered by these programs with fewer options. At the same time, consumers faced new access problems due to inpatient capacity constraints, and CTS survey data from 2000–2001 suggest that physician capacity constraints have begun to emerge as well (Ginsburg 2002a).

Moreover, recent market developments hold some potentially negative long-term consequences for consumers. Underlying health care costs have begun to rise rapidly again, increasing at rates not seen since 1990 in the pre-managed care era. Increased hospital spending was the leading contributor to annual cost increases in 2000 and 2001, due mainly to large increases in wage rates (presumably in response to labor shortages) and hospitals' better contract rates and contract terms (Strunk, Ginsburg, and Gabel 2001; Ginsburg 2002b). At the same time, the retreat from tightly managed products eased restrictions on utilization, freeing up consumers to access hospital services more liberally.

In addition, recent market developments hold potentially negative consequences for consumers with respect to the pursuit of clinical quality. Recent Institute of Medicine reports and the development of the Leapfrog group have introduced new pressures on providers to address patient safety concerns, and new payment arrangements that reward physicians for achieving certain clinical quality targets may help to promote quality improvements in the future. But the retreat from tightly managed products and providers' return to traditional competitive strategies have diminished incentives for the integration activities that many have argued are critical to develop the infrastructure for quality improvement (Devers 2002; Institute of Medicine 2001). Furthermore, increased consolidation among providers has strengthened their ability to withstand pressure to demonstrate quality and contributed to employers' waning interest in driving quality improvement through value-based purchasing (Christianson and Trude 2003; Hargraves and Trude 2002).

These developments leave purchasers and policymakers with some pressing problems. Managed care continues to dominate health care markets, but it is evolving in ways that have diminished its potential to control costs and promote quality. As health care premiums continue to rise and labor markets slacken, consumers are likely to face growing financial burden for the cost of care, particularly if current trends in health benefits continue. Although these developments could have the positive effect of making consumers more cost-conscious in their use of health care services, they will also result in increased disparities in the health system, with further stratification based on the ability to pay and greater burden on the chronically ill for the cost of care. Moreover, as coverage becomes more costly, employer offer rates and employee take-up rates will likely decline, resulting in growing uninsurance, and public insurance programs, with already strained budgets, will unlikely be able to fill this void. More aggressive purchasing or policy solutions will be needed to prevent such scenarios from unfolding.

At the same time, the experience in health care markets over the past several years points to the limitations of market-led efforts to promote quality improvement. Although a dedicated community of quality experts continues to push for progress, the average health care market has made only limited advancements to date. As plans have waning influence over provider behavior and hospitals revert to more traditional strategies to compete for patients, it is unlikely that broad-based market pressure for quality improvement will take hold. These developments suggest this may be an area for state and/or federal policymakers and private foundations to play a more active role in particularly as more responsibility for the cost of care is shifted onto individual consumers.