Federal Employees Health Benefits Coverage-As with health insurance available to State and local public employees and to workers in the private sector, substance abuse and mental health benefits available to Federal employees under the Federal Employees Health Benefits Program are generally subject to special limitations. According to the Checkbook's Guide to 1990 Health Insurance Plans for Federal Employees and the American Psychiatric Association's Coverage Catalog, virtually all plans available to Federal employees treat inpatient mental health care, including, in most cases, inpatient treatment for alcoholism or drug abuse, differently from other hospital care. Coverage limitations include fewer days of hospitalization covered in full and a lower ceiling on covered inpatient expenses than for other illnesses. Similar limitations apply to coverage for alcohol and drug abuse inpatient treatment as well as for outpatient services for all three conditions. 16 ISSUES IN CONSIDERING A BENEFIT PACKAGE FOR SUBSTANCE ABUSE AND MENTAL HEALTH SERVICES Mental illness and substance abuse are often chronic disorders requiring periodic use of treatment services over a long period of years. At the present time, it is clear that there is no agreement on treatment modalities and settings for these disorders and little uniformity in benefit structures among third-party payers for substance abuse and mental health treatment services. Persons suffering from these disorders, or with family members suffering from such disorders, who have access through their employment to health insurance are more likely, when given a choice, to select insurance coverage with the most comprehensive benefit packages for such services. Attempts by insurers to avoid incurring the costs of covering these "high risk" patients have led to the benefits limits for mental health and substance abuse treatment coverage that characterize current health insurance practice. Some would argue that establishing a basic minimum level of coverage for mental health and substance abuse treatment would eliminate the need for the limits in coverage that currently exist. 16 Walton Francis and editors of Washington Consumers' CHECKBOOK Magazine, Checkbook's Guide to Health Insurance Plans for Federal Employees, 1989; and Patrice Scheidemandel, compiler, The Coverage Catalog, American Psychiatric Association, 1989. Rationale for Including Substance Abuse and Mental Health Coverage in a Benefit Package Several issues need to be discussed in considering the design of a uniform health insurance benefit for substance abuse and mental health treatment services, the first issue being whether to include such coverage at all. The most basic argument for covering such benefits is that these disorders are illnesses for which treatment should be reimbursed on the same basis as any other illness, that providing benefits to reimburse the treatment of substance abuse and mental illness constitutes responsible and humane social policy in response to a significant public need. In addition, requiring the reimbursement of mental health and substance abuse treatment at a level more comparable to that for other illnesses could help to remove or reduce the stigma associated with these disorders and help to improve the availability of and accessibility to such services. Some argue that the treatment of substance abuse and mental health disorders result, in the long term, in reduction in treatment for other health disorders. According to this argument, the net cost of treatment for substance abuse and mental health care would thus be reduced by the extent of savings in general medical care utilization. Reviews of research on the significance of this medical cost offset have found that such reductions did take place. One study reported that 12 of 13 studies in mental health care found reductions of 5 to 85 percent in medical care utilization subsequent to a mental health intervention and reductions of 26 to 69 percent in medical care utilization in 12 studies on alcohol abuse treatment. 17 Others find the research findings on the medical cost offset of substance abuse and mental health treatment unconvincing, because of a variety of methodological limitations in the research. 18 Increasing insurance coverage for substance abuse and mental health treatment would likely result in some cost-shifting from the public to the private sector. Mental health and substance abuse treatment that is not covered by private health insurance has traditionally been supported primarily by government funds. The Federal Government supports such care through Medicare and Medicaid, the Department of Veterans Affairs, and the ADMS block grant. State and local governments support and provide care through such providers as public general and psychi 17 Kenneth R. Jones, and Thomas R. Vischi, "Impact of Alcohol, Drug Abuse, and Mental Health Treatment on Medical Care Utilization: A Review of the Research Literature," Medical Care, (17) (December 1979). 18 David Mechanic, "The Evolution of Mental Health Services and Mental Health Services Research," in The Future of Mental Health Services Research, NIMH, 1989, 5. atric hospitals and other facilities. It can be argued that requiring increased benefit coverage under private health insurance would shift some of these costs to the private sector, reduce the burden of the public sector, and generally broaden the base of financial support for treatment services. Some who oppose a required benefit for substance abuse and mental health services do so because they feel that providers of such services often reside outside the traditional health care system where there is little quality control and little hard evidence that treatment is effective. As many alcohol and drug abuse treatment programs and providers exist apart from the established health care system, this argument goes, there is little assurance that the quality of services delivered meets established levels of care in the traditional health delivery system. In addition, the fact that there is still so little agreement on the efficacy of the various treatments for these disorders makes some observers reluctant to support enhanced insurance coverage. Others may oppose including the benefit in a package because of the nature of these disorders, feeling that the majority of the insured population do not abuse these substances and should not be forced to pay for the self-inflicted problems of those who choose to abuse alcohol and drugs. According to this argument, most members of the population use alcohol and prescription drugs moderately and responsibly, and should not be forced to pay for the consequences of alcohol or drug abuse by a minority of persons who choose to act irresponsibly. Since alcoholism and drug addiction are self-inflicted problems, the proper response to such abuse is for the person involved to moderate such behavior and stop the abuse. Insurance premiums collected from the majority of insurance subscribers should not have to be used to pay for the self-inflicted problems of a few. 19 Design of a Benefit Package If a decision is made to include substance abuse and mental health coverage in a benefit package, further issues arise on what such a benefit package should include. Should mental health and substance abuse treatment services be covered on the same basis as other illnesses, or should there be limits on coverage for such care? If it is determined that substance abuse and mental health care should not be covered on the same basis as other care, where should limits be 19 "Private Health Insurance Coverage for Alcoholism and Drug Dependency Services: State Legislation that Mandates Benefits or Requires Insurers to Offer Such Benefits for Purchase," NASADAD Alcohol and Drug Abuse Report (January/February 1986). made-in types of providers, in numbers of days or visits of care covered, or in dollars of care covered? What kinds of cost-sharing measures, if any, should be included? What kinds of cost containment measures should be included? There is evidence that the existence of increased third-party coverage for health services may lead to an increase in unnecessary and excessive use of such services, a phenomenon known as moral hazard. Some people, particularly those with less severe mental health disorders, will apparently use outpatient mental health services if covered by health insurance that they would not have chosen to use if they had to pay the full cost for the services out of pocket. This appears to be the case for outpatient mental health care more than for other outpatient health care services. Some argue that improving mental health and substance abuse treatment coverage and the increased use of services that may follow because of moral hazard will increase costs substantially. As a result, insurers would have to increase premiums to pay for the care to such an extent that premiums would approach the cost of care. Research seems to indicate that use of ambulatory mental health care is responsive to patient cost sharing. Low copayments with no utilization limits would likely produce significantly higher use of outpatient services and high benefit costs; a plan with a high level of patient cost sharing combined with a catastrophic ceiling produces lower use of services and lower costs.20 On the other hand, research seems to indicate that inpatient mental health care is not responsive to cost sharing, but does respond to prospective payment, length of stay limits, and fixed budget reimbursement.21 A 1986 NIMH report suggesting modifications to Medicare's prospective payment system to cover inpatient treatment for mental illness may have value in designing inpatient coverage under a mandated benefit. This report recommended classifying patients by diagnosis, treatment, and age-"Three groups of mental illness-psychoses, organic disorders, all others-could each be subdivided by intensity of treatment. It is possible to differentiate intensive treatment from routine treatment. (Intensive treatment is typical of specialty psychiatric facilities, while routine treatment is typical of the psychiatric treatment provided in so-called scatter beds, that is, general hospital beds not located in a psychiatric unit. Further, because child and adolescent patients who receive intensive treatment appear to stay much 20 Willard G. Manning, Jr., et al., "How Cost Sharing Affects the Use of Ambulatory Mental Health Services,” Journal of the American Medical Association, (256) (October 10, 1986), 1933. 21 Thomas G. McGuire, “Financing and Reimbursement for Mental Health Services," In The Future of Mental Health Services Research, NIMH, 1989. 91 longer than other patients, age could be a defining characteristic.” 22 22) Mental health and substance abuse for young people may be a matter of particular concern in designing a benefit package. Research appears to indicate that teenagers account for a larger percentage of the costs and utilization of psychiatric and substance abuse health insurance benefits than other age groups. A recent research report notes that much of the care provided adolescents is provided in hospitals, the most expensive of all treatment settings, and suggests that, in some cases, outpatient or residential programs, especially those that involve the entire family, are much less expensive and are equally or more effective for treating adolescent mental health and/or substance abuse problems. 23 In considering coverage for treatment of mental illness and substance abuse, it would be helpful to be able to take into account what is known about the most effective ways of providing such care in designing the most efficient and cost-effective benefit package. It may be, however, that there is too little definitive information available about the relative effectiveness of various treatment modalities and settings to be useful in designing a benefit package. The mental health treatment research review cited above appears to show, for instance, that a combination of short- rather than long-term inpatient care combined with intensive aftercare services in the community, or alternative treatment in the community that avoids hospitalization completely, can be both effective and cost-effective in treating severe mental illnesses. 22 Antoinette Gattozzi, Prospective Payment of Mental Health Care, State Health Reports: Mental Health, Alcoholism, and Drug Abuse (April 1986). 23 "Psychiatric/Substance Abuse Benefits Costs, Utilization Are Highest For Adolescents," Spencer's Research Reports on Employee Benefits (January 1990). There is some feeling in the substance abuse treatment community that the 28-day inpatient treatment model that appears to dominate the field, particularly for alcoholism treatment, may not always be the most effective or cost-effective method of treatment. The alcoholism treatment research cited above seems to favor a short course of inpatient care combined with longer term outpatient care as a more cost-effective form of treatment than longer term inpatient care. Drug abuse treatment research has found that the longer the course of treatment the greater the chances of success. It is difficult to imagine an employer favoring the coverage, under an employee insurance package, of a course of treatment of a year or more in a therapeutic community, not only because of the cost of such treatment, but because of the additional operating costs and inefficiencies of losing an employee for a year or more. In addition, it must be noted that long-term drug abuse treatment research has concentrated on treatment for heroin addiction, and little is known as yet about its applicability to treatment for cocaine abuse. Those without health insurance coverage for mental health care or treatment for substance abuse have traditionally had to depend on publicly-financed care in State and locally-financed facilities, inpatient and outpatient, with their flaws and limitations. With the recidivism and repeated need for treatment that is common among some alcohol and drug abusers, as well as the long-term, often life-long, care necessary for many chronically mentally ill persons, the design of any mandated benefit package would probably need to retain some limits in the courses of treatment for substance abuse that would be covered before a patient were forced to fall back on the public sector for care. The Honorable John D. Rockefeller IV Chairman, The Pepper Commission United States Bipartisan Commission on Comprehensive Health Care Dear Mr. Chairman: In response to your request of August 11, 1989, we have examined the issues that would need to be addressed in ensuring the quality of health care under any plan to expand health care coverage for the uninsured. We have assumed that the current system of multiple public and private purchasers of health care will remain in place for at least the immediate future. In addition, we have examined the adequacy of the knowledge base for structuring such quality assurance activities. However, because we believe that most of the quality assurance issues that would need to be addressed are generic, much of this report does not distinguish between quality assurance for the uninsured and for the general population. This briefing report presents the results of our work as discussed with your staff on January 23, 1990. We begin by noting that quality is multidimensional and that we have focused our attention on the appropriateness of care and the technical and clinical aspects of quality. We also note that health care system design has important implications for quality, and we briefly describe the various levels at which quality assurance activities are currently conducted. We conclude that there is a considerable body of knowledge about, and experience with, the organization and conduct of quality assessment and assurance activities and • This report has been furnished by the U.S. General Accounting Office at the request of the Chairman of the U.S. Bipartisan Commission on Comprehensive Health Care. It is reprinted from a GAO report of the same title, GAO/PEMD-90-14BR (Washington, D.C.: February 21, 1990). a growing interest in improving and expanding these activities among many of the participants, including the medical community, consumers, employers, and purchasers of care. In keeping with this growing interest, we suggest that a comprehensive, national strategy for assessing and assuring the quality of health care is needed. We see at least four elements as essential to a comprehensive national strategy: (1) national practice guidelines and standards of care; (2) enhanced data to support quality assurance activities; (3) improved approaches to quality assessment and assurance at the local level; and (4) a national focus for developing, implementing, and monitoring a national system. The reasons we see for needing a comprehensive national approach and a brief discussion of each of its elements are contained in section 2 of this report. Our conclusions are based primarily on the studies of health care quality assessment and assurance in a number of settings spanning the public and private sectors that we have conducted over the past few years. We have also incorporated concepts and information on quality assurance contained in published sources, including the Institute of Medicine's report entitled Controlling Costs and Changing Patient Care? and the Office of Technology Assessment's report entitled The Quality of Medical Care: Information for Consumers. Finally, we convened a meeting of experts in November 1989 for the explicit purpose of exploring these issues and have had them review a draft of this report. (See appendix I.) We have not conducted a comprehensive review and analysis of existing quality assurance programs. Any references in this report to specific quality assurance programs are examples used to illustrate particular points and do not necessarily represent the "best" programs available. Our work was performed in accordance with generally accepted government auditing standards. We have incorporated the comments of our experts but have not requested comments from any federal agency, since none is evaluated in this work. Unless you publicly announce the contents of this report earlier, we plan no further distribution until 30 days from its date. We will then make copies available to others upon request. If you have any questions or would like additional information, please call me at (202) 2751854 or Mr. Robert York, Acting Director of Program Evaluation in Human Services Areas, at (202) 275-5885. Other major contributors to this report are listed in appendix II. Sincerely yours, Eleanor Chelimsky Assistant Comptroller General care, coordinating services, and making appropriate referrals. • Health care providers may emphasize the decisionmaking process that underlies diagnosis and treatment, the clinical content of care, and the technical skill with which it is rendered. • Purchasers may place greater weight on questions of cost-effectiveness, including the need for individual diagnostic and therapeutic services, the appropriateness of the setting in which care is delivered, and the frequency, timing, and duration of services. SECTION 1-QUALITY, QUALITY In this section, we begin with an overview of the concept of health care quality and how we use it in this report. We draw a distinction between quality assessment and quality assurance, which is important for our discussion of the need for a national, comprehensive quality assurance strategy in section 2. We note some instances in which the design and operation of the health care system itself can influence quality quite independently of any formal mechanism for reviewing the quality of care. Finally, we briefly describe the different levels in the health care system at which quality issues may be addressed. All these views of quality are legitimate and important. However, our primary focus is on the appropriateness of medical services and their clinical and technical quality. This implies a concern for such issues as whether necessary care was provided, whether the outcome was acceptable, whether unnecessary services were provided, and whether the location of care (that is, hospital, nursing home, home, ambulatory setting, and so on) was consistent with the patient's needs. There are important reasons for this focus. First, and perhaps most important, providing appropriate medical care that is effective is the common denominator of the preferences of all three groups. Second, providing improved access to inappropriate care or poor-quality care is not likely to result in improved health outcomes. Third, currently available strategies for assessing and assuring quality are targeted especially to the appropriateness of care and to technical aspects of quality. As a result, focusing attention in these areas offers the greatest potential for near-term improvements in quality. Quality Is Multidimensional The quality of care is a multidimensional concept that defies simple definition. Quality encompasses many aspects of care and means different things to different people. Patients, health care providers, and purchasers may have different notions about what constitutes high-quality care. • To patients, "getting better" (that is, the outcome of care) is probably the primary concern. In addition, having access to care that is affordable, conveniently available, and provided in a manner that respects their concerns and preferences is important. The responsiveness of the delivery system may also be important-for example, meeting patients' individual needs for emergency Quality Assessment Should Be Distinguished From Quality Assurance It is important to distinguish between quality assessment and quality assurance. Quality assessment involves the use of measures of quality, based on either explicit or implicit criteria, to assess the structure, process, and outcome of care and to monitor levels of quality over time. Quality assurance goes beyond the simple assessment of quality to include its improvement. This requires identifying and confirming problems in the quality of medical care, planning interventions to lessen or eliminate the problems, monitoring the effectiveness of the interventions, and instituting additional changes and monitoring where warranted. |