|
FROM THE FIELDBack To The Drawing Board: New Directions In Health Plans’ Care Management Strategies
The backlash against managed care has pressured health plans to reexamine their approaches to controlling utilization and managing their members’ health care needs, but how much has really changed? Interviews with health plans and others in twelve nationally representative markets suggest that the changes are significant. New and refined disease management programs are improving the care experience of participants with certain prevalent chronic illnesses, while utilization management changes are reducing the administrative burden for providers. Still, disease management programs will need to greatly expand in scope and scale if plans are to succeed in addressing the complex health care needs of aging populations and those with chronic diseases.
As managed care enrollment grew during the 1990s, so did expectations for the ability of care management strategies to control costs and improve quality of care.1 Evidence that health maintenance organizations (HMOs) have achieved some success in changing health care delivery—such as shifting care from specialists to primary care providers and improving the use of clinical preventive services—has fueled interest in care management strategies and raised questions about their adaptability to non-HMO settings.2 Evidence on the effectiveness of specific care management tools is stronger for utilization management (UM) tools, such as primary care gate keeping and prior-approval requirements, and weaker to date for disease management strategies.3 Overall, plans’ implementation of these strategies has varied considerably across markets and has been slower than expected for disease management.4 Most recently health plans—particularly HMOs—have faced pressure to change their care management strategies and redefine themselves on several fronts. Consumers have voiced concerns about undue restrictions on choice and unnecessary administrative hassles in seeking care and have begun migrating to less restrictive health insurance products.5 Facing highly competitive labor markets, employers have adjusted their health plan offerings to reflect the demand for less restrictive products.6 Meanwhile, providers have complained about interference with medical decision making and about burdensome administrative costs of complying with plans’ care management policies; they have begun to resist these policies or withdraw from plans that use them.7 In many states policymakers have responded to the backlash by exploring regulatory restrictions on plans’ care management policies.8 This paper examines how health plans’ care management practices changed from 1998 to 2000 and how these changes affect managed care’s ability to control costs and improve quality. We focus on two types of strategies that together account for much of the care management activity contributed by health plans: (1) UM strategies used to prevent unnecessary and inappropriate health care and to ensure that care is provided in the lowest-cost appropriate setting, and (2) disease and health management programs designed to improve the health of persons with specific conditions and to reduce use and costs associated with avoidable complications such as emergency visits and hospitalizations.9 These programs often focus on improving patients’ compliance with appropriate treatment and self-care strategies and on coordinating services for persons with multiple health care needs. In the context of the backlash against managed care, some argue that these programs have the potential to serve as more consumer- and provider-friendly substitutes for traditional UM practices.10
This research is based on the Community Tracking Study (CTS) site visits, conducted in twelve nationally representative communities every two years.11 This paper draws primarily upon semistructured interviews during site visits completed between June 2000 and March 2001. Approximately 220 interviews were conducted with administrators from forty-eight plans: the largest national health plan, the largest locally owned health plan, and the largest Blue Cross Blue Shield plan in each community. We focused on the largest plans because changes in them have the most widespread impact on people in the market. In each of these plans we attempted to interview the chief executive officer (CEO), medical director, UM director, marketing executive, network development executive, and pharmacy benefit manager. To ensure adequate coverage of the major health plan competitors, we interviewed marketing executives at up to two additional health plans in each community.12 In all, we interviewed thirty-three medical directors and twenty-one UM or quality management executives, as well as 156 other senior health plan executives. These interviews asked specifically about care management practices, changes made since 1998, and the rationale and perceived impact of these changes. Interviews with CEOs and marketing directors provided general information about care management strategies, while interviews with medical directors and UM directors provided specific information about the structure and operation of these activities. To confirm and expand upon this information, we also inquired about health plans’ care management practices during interviews with hospitals, physician organizations, and employers.13
Twenty-one of the plans examined in this study (44 percent) had made major changes to their UM activities since 1998 (Exhibit 1  ). Instead of pulling back from these activities altogether, many of these plans loosened some utilization controls while strengthening others.
Fewer requirements for advance approval. Health plans in all twelve communities reported reviewing fewer physician decisions in advance of treatment in 2000 than they did in 1998. Common changes included eliminating advance-approval requirements for hospital admissions, for outpatient tests and procedures, and for referrals to specialists (Exhibit 1  ). Some plans eliminated referral requirements altogether—allowing patients direct access to specialty care—while others retained these requirements but discontinued the policy of reviewing physicians’ referrals for appropriateness. Other plans streamlined the process of reviewing referrals, making approvals automatic and instantaneous via telephone or the Internet for most types of specialists.
Strengthened concurrent and retrospective review.
More than half of the plans that eliminated selected advance approval requirements since 1998 offset these changes by strengthening other type of UM practices (Exhibit 1 Other plans have enhanced their processes for retrospective utilization review and provider profiling. In Boston, Greenville, and Miami, for example, several plans expanded the types of comparative information provided to physicians to include detailed measures of quality of care and service use, adjusted for demographic differences in physicians’ patient populations. Some of these plans have begun to provide in-person consultation to physicians who show measures consistently outside the norm of their peers. Additionally, plans in several markets introduced financial bonuses tied to measures of quality.14 Effects. Health plans uniformly indicated that recent changes in the use of traditional UM practices allowed them to reduce their administrative costs of operation and improve relationships with consumers and providers. However, expectations about the impact of these changes on health care use and overall costs varied considerably across plans. Most plans expected to lower administrative costs by discontinuing advance-approval requirements for high-volume services that were costly to review but rarely denied. However, several plans acknowledged the possibility of major utilization increases and consequential cost growth as a result of the sentinel effects that these requirements have on clinical practice patterns. One large plan in Syracuse, for example, initially eliminated nearly all of its preauthorization requirements but then reintroduced them for selected specialties such as dermatology and orthopedics after noting a rise in use. Other plans expected little if any change in aggregate use and cost—either because these plans compensated for the loss of prospective utilization controls by reinforcing concurrent and retrospective controls or because plans eliminated only controls that had not proved to be effective in containing use.
Over the past few years plans in most of the markets expanded or refined their approaches to disease management (Exhibit 1 ). Approaches. In most markets plans added programs targeting specific conditions so that by the year 2000 each offered a set of three to five disease management programs, which commonly included diabetes, congestive heart failure, asthma, and pregnancy or high-risk pregnancy.15 Other conditions addressed by multiple plans in the study included cardiac care, oncology, lower back pain, depression, and chronic obstructive pulmonary disease. Most of the large plans with multiple product lines had not yet implemented disease management programs across all of their non-HMO products at the time of our visit but were in some stage of doing so. Most of the new and expanded disease and health management strategies were designed to work with members directly.16 The strategies varied in their comprehensiveness but most often included educational efforts aimed at all members with the condition along with telephonic case management for members with the condition identified as higher-risk.17 Although the predominant trend was to develop or refine disease management programs, some plans we visited were also pursuing care management approaches that were not organized to address specific diseases. The rationale was that high-risk patients often have multiple conditions that interact with each other, and therefore it may be most effective to address a person’s health problems and service needs as a set, rather than focusing solely on a targeted condition.18 At the United Healthcare plans we visited, each discharge from a hospital triggered a follow-up call from a nurse, to ensure that the enrollee had obtained prescriptions, scheduled necessary follow-up appointments, and so on, and to identify any gaps in care.19 Effects on enrollees. Most plans pursued disease management programs to save money and improve health outcomes but had little evidence of such effects to date. Many of the plans’ programs were new and thus had not had time to produce much measurable effect. Other plans noted the difficulty of documenting the effects of their programs or cited information-system limitations when asked about effects. However, a few plans had made specific estimates of various types. For example, one plan reported that its preterm labor prevention program had been successful: "We see a $2.70 return on $1 investment on that one, through decreased [neonatal intensive care unit] admissions, increased birth-weights, and an increased number of babies born after 33 weeks." Despite a lack of hard evidence, many respondents nevertheless expressed their belief that at least some of the disease management programs were effective. Some referred to studies of other programs that they viewed as similar, or the same program evaluated elsewhere, as evidence that their programs should be working. Of note, plans were far more certain that the disease management programs improved the satisfaction of members who participated in them than they were about the effects of the programs on health outcomes or costs. As one plan noted, "The case management efforts are improving consumer experiences within the health care system and improving quality of life for those with catastrophic and chronic illnesses. Purchasers are most likely getting better worker productivity and less work time lost due to illness, but this is very hard to document." In contrast to the favorable response among participating members, plans reported that physicians’ reactions to the programs generally ranged from skepticism to limited support. This is consistent with sentiments expressed in interviews with providers.20 Effects on communities. Even if, as plans believe, the programs are improving the quality of care for the participating members, their impact on persons with the conditions in the community as a whole was still small at the time of our visits, largely because of the small number of persons in many of them.21 Some programs’ small size appeared to be due to limited resources allocated for the programs.22 The reluctance to invest in these programs likely reflects several factors, foremost of which is the lack of evidence to date that the programs reduce the plans’ costs. Another consideration for plans may be their ability to reap public relations benefits from having a program, even if it is very small; few consumers or employers look beyond the presence or absence of such programs, we were told. Plans also cited employers’ focus on price and the high turnover rates of members in the market as reasons for lack of greater investment in disease management programs.23
Although there was no evidence of formal collaboration among plans, competitive pressures have led health plans to change their disease management initiatives and UM processes in roughly similar ways and points in time within markets. Most health plans were not eager to be viewed as market leaders in disease management, probably because this could lead to adverse selection as sicker people gravitate toward the plan. At the same time, they did not want to be viewed as behind the market in responding to the growing demand for less restrictive health insurance products by relaxing utilization controls. Plans in some markets appeared more likely to change their UM processes in response to a perceived market leader. In Miami and Greenville, for example, local plans began to develop less restrictive HMO products only after such products were introduced successfully by national insurers. In other markets such as Phoenix and northern New Jersey, however, health plans began to loosen their utilization controls at about the same time in response to growing demand for less restrictive products. These findings appear consistent with other studies of innovation diffusion showing that change within a local system can be stimulated both by innovators operating partially outside the system and by pressures within the system.24 The direction and pace of change in health plans’ care management strategies have not been uniform, as some markets have experienced considerably more change than others. One obvious driver of change relates to the presence of HMOs and other tightly managed health insurance products in the market. In markets with historically low levels of HMO penetration, such as Greenville and Syracuse, stringent utilization controls have never come into widespread use, and disease management programs have been slow to develop without other forces stimulating these activities. By contrast, in markets where both HMOs and less restrictive products are strong competitors—such as Seattle and northern New Jersey—plans have been more proactive in loosening utilization controls for HMOs and in expanding disease and health management efforts to non-HMO products. The presence of national plans also affected the degree of change in each market. Disease management strategies in nine of the twenty-three plans with new programs had been shaped by the national programs of United Healthcare and CIGNA.25 Also, efforts to reduce stringent utilization controls were particularly widespread in markets where local affiliates of national plans such as United and Aetna U.S. Healthcare have a strong presence and led the introduction of less restrictive products. For disease and health management practices, the pace of change appeared to depend in part on the history of these practices within the market, with more change occurring in markets that had significant disease management activity prior to 1998 (Boston, Miami, northern New Jersey, and Phoenix). Several additional market characteristics appeared to stimulate change in health plans’ care management strategies in selected communities. Large employers’ demand for disease management programs and for health plans with National Committee for Quality Assurance (NCQA) accreditation played important roles in health plans’ decisions to establish such programs in Cleveland, Lansing, and Little Rock.26 In markets where full-risk contracting between health plans and providers was declining, plans strengthened both their UM and disease management activities (Seattle and Phoenix). The large enrollment of senior populations in Medicare+Choice influenced development of disease management programs in Boston and Miami. And in Miami increased negotiating leverage among providers led several plans to discontinue their practices of retroactively denying payments for services deemed unnecessary. These findings suggest directions for further research to examine the cumulative effects of market forces on care management activities.
Health plans’ approaches to care management have changed substantively; the changes we observed are more than just posturing by a tarnished industry. Overall, the changes shift the "management" of care away from management of all or many enrollees toward management of smaller groups of sicker enrollees. This shift may be largely a positive development. Policymakers and the public at large should applaud the elimination of UM practices that were costly to administer, created "hoops" for enrollees and providers, and ultimately saved little money. We found consensus among plans that the growth of disease management has improved the care experience of participants, and many respondents believed that their programs were improving quality of care. However, several issues remain. Because the UM that remains in place is focused on smaller groups of enrollees, barriers to care may be less visible to policymakers and the general public, thus relieving political pressure on plans. Nevertheless, the potential for adverse impact on these patients may be at least as great as it was before. Also, the overall effect of care management changes on costs remains an open question. As more and more plans loosen their UM processes and providers become accustomed to the changes, costs could increase more than if just a few plans made such changes. In the context of cost-increasing pressures from other sources as well, this potential should be a concern for public and private purchasers alike and represents a compelling target for further research. Developing better data on the impact of disease management programs on both costs and quality appears to be another critical next step, particularly as public-sector purchasers consider these programs for containing costs and improving health outside of managed care arrangements as well as within them. On a more positive note, the fact that most plans are moving in similar directions to implement disease management programs largely focused on the same diseases could have a large positive effect on care for persons with those diseases nationally—perhaps far greater than any individual plan’s program alone. However, while the growth in disease management programs since 1998 represents a step forward for consumers with certain prevalent chronic illnesses, the programs will need to expand greatly in scope and scale if plans are to address the complex health care needs of aging populations and those with chronic diseases. Both public and private purchasers can stimulate such expansions by requiring plans to document their successes in reaching members who can benefit from these programs and by pressing plans to develop clinically and culturally appropriate programs for a broader array of population groups and health conditions. Persistently low enrollment in disease management programs is an obvious marker of health plans that offer the illusion of care management without the reality.
Suzanne Felt-Lisk is a senior health researcher and Glen Mays, a health researcher, at Mathematica Policy Research, Inc. in Washington, D.C. This research was conducted as part of the Center for Studying Health System Change’s Community Tracking Study, which is funded by the Robert Wood Johnson Foundation. The authors are grateful for the guidance and support provided by members of the Community Tracking Study research team, including Cara Lesser, Paul Ginsburg, Joy Grossman, Robert Hurley, and Debra Draper.
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Data And Methods
Changes In Traditional...
