Background

Health care purchasers are caught in a dilemma created over the past 50 years and for which there is no easy resolution. Following World War II, the U.S. economy was strong and its industry dominated the world marketplace. Jobs were plentiful, and employers competed for skilled workers. The American ethic of a benevolent employer was firmly reinforced by years of unions' struggles with management and by a healthy economy under which employers could afford to offer generous health benefits.

For years, the health insurance contract offered ever-increasing benefits, freedom of choice, and first-dollar coverage (few copayments or deductibles). Employers trusted their employees and providers. Although consumers and providers struggled for many years to develop more adequate mental health and substance abuse benefits, most people were happy with the health care system. Furthermore, the U.S. Congress initiated the Community Mental Health Centers Act, Medicare, Medicaid, Hill-Burton, and other programs (see Chapter 3), which contributed greatly to the growth of the health care industry. With these investments, the public and private sectors created health care access and resources that were unparalleled in world history.

Fueled by scientific prowess and expanding financial commitments, the health care system appeared to have no limits in its potential capacities to provide health care. However, unlimited growth could not continue. With the rising costs of health care services threatening the financial stability of their budgets, private and public payers increasingly turned to methods that make health care accountable and affordable and that prevent cutbacks in previously reimbursed health benefits. The widespread initiation of utilization management, health maintenance organizations (HMOs), and other managed care methods during the past quarter century has emphasized cost accountability (IOM, 1989a).

These programs have cumulatively evolved into an industry and have become a strong force in the health care system. Consumers and providers who believe that autonomous health resource decisions on the basis of tradition and the health care contract are consequently in conflict with such policies. The tensions over cost controls have increasingly focused concerns about cost-containment efforts on quality issues such as the following:

• qualifications of and consumers' geographic access to a comprehensive range of providers;
• prevention of avoidable illness and provision of timely and focused treatment interventions;
• availability of services, on the basis of urgency of need;
• courtesy, convenience, and comfort of services;
• compassion and kindness of care;
• competence of providers to institute most appropriate evaluations and treatments, which would result in services that would result in the least risk to the patient and with the best health status outcome; and
• administrative efficiencies of health care services that promote quality through effective communications, consumer and provider education, decision support, and quality management, treatment coordination, and other systems.

The interest in quality is reinforced by consumer demand and empowerment, professional ethics, legal and regulatory interpretation of citizens' rights, and attempts by businesses to satisfy and keep customers in a competitive health care marketplace. For public purchasers who are accountable for public funds, it is important to demonstrate that health care has good value and is worth the investment. The next section will give an overview of different methods for assessing quality.

Methods for Quality Assessment

Employers as Purchasers of Behavioral Health Care

Managed behavioral health care organizations have encouraged the documentation of efforts to account for quality of care and services. Xerox, IBM, GTE, and Digital Equipment Corporation have led the way in establishing quality specifications for their managed behavioral health care organization vendors. Through the imposition of contract guarantees, corporate purchasers reward quality and penalize poor service.

Contract-based performance standards have become the base for industry and voluntary accreditation organization standards, notably those developed by AMBHA (1995) and NCQA (1996a, b). Some payers have developed their own explicit requirements for HMOs that provide care under their health benefits plan for such areas as member services and satisfaction, administrative services, organizational structure and philosophy, provider credentialing and performance monitoring, clinical services management, clinical delivery support systems, and confidentiality. Digital Equipment Corporation, for example, has specific requirements for the behavioral health services that it purchases:

• benefit design,
• access,
• triage,
• treatment approach,
• case management,
• alternative treatment settings,
• outcomes measurement,
• quality management, and
• prevention and early intervention.

Table 6.1 compares some of the more widely used behavioral health care standards.

TABLE 6.1

Cross-Comparison of Managed Behavioral Health Care Performance Indicators.

Trends in Quality Standards in the Private Sector

Because so many purchasers' efforts to become involved in prescribing methods and outcomes goals for quality accountability are in the early stages and because the state of population-based measurement systems is not refined, quality management in behavioral health and other clinical services is in the early stages but is evolving rapidly. As with most evolutions, an experimental phase precedes consensus about what constitutes the best approach.

In addition to the standards listed in Table 6.1, numerous employer coalitions, both local and national, are now embarked on efforts to establish performance requirements for managed care. Examples include the Managed Health Care Association, the Employer Consortium, the National HMO Purchasing Coalition, the Minnesota Buyers Healthcare Action Group, and the Pacific Business Group on Health. Many of these coalitions have significant participation by health services consumers and their representatives, such as unions, advocates, organizations, and insurance commission agencies.

The Foundation for Accountability (FACCT), representing a broad coalition of public and private purchasers and others, has begun to develop and test tools that will allow documentation of population-specific functioning, quality of life, satisfaction with services, and risk reduction for a number of medical conditions commonly seen in health plans, such as diabetes, asthma, breast cancer, coronary artery disease, and low back pain (FACCT, 1995). Mood and anxiety disorders represent other conditions whose prevalence and direct and indirect costs and for which there are problems in the quality of evaluation and treatment are of great concern.

FACCT evolved during 1995 because of the interests of several major private and public purchasers, as well as consumer groups, for public-measurement systems that account for outcomes related to quality such as patient satisfaction, health-related quality of life, and functional status. Their wish to expedite the development of outcomes and methodologies and systems was spurred by Paul Ellwood's long-standing promotion of the national goal of a patient-centered and integrated outcomes management system. To date, FACCT has released measurement methods for a number of conditions (e.g., diabetes, asthma, and breast cancer) and is planning the development of similar tools and pilot programs for behavioral health conditions such as depression. The application and evolution of FACCT methods will be influenced by the amount of funding that is available and how meaningful the collected information will be to consumers and purchasers.

Many purchasers are now prodding their contracted managed care organizations and a few are requiring their contracted managed care organizations to collect and publicly report their Health Plan Employer Data and Information Set (HEDIS) results. This and other public report cards constitute a major trend in health care and are being actively supported by consumers who want meaningful data on which they can make personal health care and health plan selection decisions. Managed care organizations are concerned about the risk adjustment problems with some measures, the cost of collecting data, the high-stakes business risks that can follow questionable performance, and the plethora of reporting requirements that are being imposed under multiple reporting systems. The evolution of other potentially large systems, such as the Joint Commission on Accreditation of Healthcare Organizations (JCAHO), NCQA, Utilization Review Accreditation Commission (URAC), and Council on Accreditation of Services for Families and Children (COA), adds to their concerns about their abilities to simultaneously meet the market's demands for improved accountability and lower premiums.

From these early efforts to establish quality standards and tools that can be used to measure quality, several views are emerging:

• Standards and measures for quality-related components of structure (e.g., state licensure and national accreditation), process (e.g., provider adherence to clinical policies), and outcomes (e.g., level of functioning and patient satisfaction with clinical care) are relevant to conclusions about quality.
• Routine and consistent measurement of specific health conditions and illnesses should be conducted for individuals in a health plan and for the population.
• Risk adjustments, based on individual and population variables, are critical in reaching conclusions about the process and outcomes of health care.
• Health status (physical functioning, role capacities, and objective and subjective well-being) should be consistently evaluated in determining the effectiveness of health care interventions.
• Generic population and disease-specific health measures are relevant to the management of a population's health.
• Plans need to evaluate systematically individual and population health risk behaviors in developing targeted interventions that could reduce avoidable health costs and increase the likelihood of positive health status over time.
• Accreditation, licensing, quality auditing, performance monitoring, and other accountability mechanisms have limited impacts when they are instituted in a piecemeal or uncoordinated fashion. For example, external oversight processes may adequately monitor the overall quality of care, but oversight tends only to identify problems rather than to help solve them, especially when the solutions may involve changes in the internal procedures of an organization.
• Quality of care requires cooperative commitments to quality-related goals by payers, practitioners, consumers, regulators, and managed care organizations, as well as a common and practical system for measuring and analyzing quality-related information.

Purchasers share with responsible managed care organizations and consumers a unifying goal of creating a more responsive health care delivery system, that is, one that is both more efficient and more effective. Over time it is probable that a best practices system will emerge that monitors, measures, and reports on the relevant information needed to determine effectiveness in sensitive, reliable, specific, and valid terms. The process of developing best practices will be facilitated if purchasers and managed care organizations include a variety of stakeholders in the discussions, including practitioners, administrators, researchers, accreditation organizations, public agencies, and the general public.

The quest for best practices and affordable systems is one of the current megatrends in health care, spawning a new industry that may provide the tools that stakeholders in the U.S. health care system need to make quality-based decisions. During the next few years of systems experimentation and consensus development, quality-related accountability will continue to develop in a variety of ways and will require leadership (Ellwood, 1988). It is now unclear by what means payers, providers, consumers, managed care organizations, and other stakeholders will come together to develop consensus about the systems that promote gains in personal health and the public good. Leadership will be needed to guide each step in this development and consensus-building process.

Healthy People 2000

In 1979, the U.S. Public Health Service (PHS) initiated a process of setting objectives and measurable targets for health promotion and disease prevention. Now known as Healthy People 2000, this U.S. Public Health Service process adapted the private-sector “management by objectives” approach and set objectives for improvements in health status, risk reduction, public and professional awareness of prevention, health services, and protective measures. There are now 300 objectives in 22 priority areas addressing health promotion, health protection, preventive services, and data systems (DHHS, 1995).

The development of Healthy People 2000 was directed by the Office of Disease Prevention and Health Promotion, under the leadership of J. Michael McGinnis, in response to a Congressional mandate. The national objectives were developed by 22 expert working groups, a consortium of more than 300 national membership organizations, 56 state and territorial health departments, the IOM, and public review and comment involving more than 10,000 individuals (DHHS, 1990, 1992).

Dissemination of these objectives has been widespread in the public health community through coordination with the American Public Health Association, the Association of State and Territorial Health Officials, the National Association of County and City Health Officials, and other groups. Regular reports describing the national progress toward meeting the objectives are issued, and there is general agreement that the objectives encourage the systematic measurement of needs, the setting of targets, the monitoring of the progress that has been made, and the evaluation of outcomes in public health (IOM, 1989a).

Substance Abuse and Mental Health Services Administration

Health Care Financing Administration

HCFA, which administers the Medicare and Medicaid programs, is the largest single purchaser of managed care in the United States and currently provides direct or financial support to 15.5 million people (Valdez, 1996). As of June 30, 1996, about 11.5 million Medicaid beneficiaries were enrolled in managed care programs, representing about a 140 percent increase in managed care enrollment since 1993. Currently, 10 percent of the Medicare population is enrolled in about 278 managed health care plans across the country, representing about a 67 percent increase since 1993 (Valdez, 1996).

In the area of quality assurance (referring to plan structure and processes), in 1991 HCFA began its Quality Assurance Reform Initiative, which was designed to monitor and improve the quality of managed care for Medicaid recipients. The initiative has developed a guide that includes specific criteria for the design of internal quality assurance programs by managed care plans. Over the long term, HCFA plans to move toward a single set of quality assurance standards for both Medicaid and Medicare beneficiaries within managed care environments (Valdez, 1996).

In the area of performance measures, HCFA worked with NCQA in the development of the Medicaid version of HEDIS and has included Medicaid HEDIS in its Quality Improvement Primer, which has been developed for state Medicaid agencies. In addition, HCFA has worked with NCQA to develop HEDIS 3.0 and by 1997 will require health plans serving the Medicare population to use some of the HEDIS measures. HCFA also is working with FACCT to develop outcome measures for plan performance.

Agency for Health Care Policy and Research

The Agency for Health Care Policy and Research (AHCPR), a U.S. Department of Health and Human Services agency, was established in 1989 with a Congressional mandate to generate and disseminate information that would be useful to consumers, practitioners, and other audiences. The majority of AHCPR's activities are aimed at improving the quality of health care. Accordingly, the agency works with several organizations, including the Foundation for Accountability, the Joint Commission on Accreditation of Healthcare Organizations, the National Committee for Quality Assurance, and the American Medical Association, to help to provide a science base for quality measurement, and to assist in translating research findings to quality measures.

Through its Center for Quality Measurement and Improvement, AHCPR conducts and supports research on the measurement and improvement of the quality of health care, including consumer surveys and satisfaction with health care services and systems. The agency has produced and disseminated clinical practice guidelines in a variety of formats to meet the needs of health care practitioners, the scientific community, educators, and consumers. A clinical practice guideline on the detection, diagnosis, and treatment of depression was released in 1993.

AHCPR sponsors a Computerized Needs-Oriented Quality Measurement Evaluation System (CONQUEST), a prototype system for collecting and evaluating clinical performance measures. The system includes two linked data bases, one on conditions and one on measures, to help clinicians, providers, managed care organizations, and purchasers find clinical performance measures that match their needs. Information is included on approximately 1,200 measures developed by public- and private-sector organizations. Among these are measures for the following conditions: affective disorder, alcohol abuse, bipolar disorder, depression, drug abuse, dysthymic disorder, panic disorder, suicidal ideation, and suicide (mortality).

AHCPR has supported research on the implementation of guidelines in a variety of settings, including HMOs and group practice, and examining a variety of strategies, including incentives and individualized feedback. Other areas of AHCPR-supported research include factors that affect costs, premiums, and choice of health plans; clinical and effectiveness research in HMOs; and managed care in rural areas.

Accreditation Organizations

The committee reviewed accreditation materials from five organizations that accredit behavioral health plans, programs, and services: CARF, COA, JCAHO, NCQA, and URAC. Representatives of these organizations were invited to make presentations at the committee's two public workshops. This section briefly describes each of the organizations, which are further compared in Table 6.2.

TABLE 6.2

Cross-Comparison of Selected Accreditation.

Changing Environment of Accreditation

There has been a proliferation and growth of accreditation organizations to match the structural changes in the industry. As described above, new accreditation organizations form to review any structure devised in managed care. Some organizations are unique, whereas others overlap in their accreditation domains but have a slightly different focus. For example, JCAHO and NCQA both accredit HMOs. JCAHO's accreditation process focuses on a staff model delivery system, whereas NCQA's process is focused on the HMO structure. Also, NCQA standards tend to focus at the highest level of an organization, whereas JCAHO, CARF, and COA are geared more toward particular programs or facilities that may be a division of a larger organization or may be free-standing or independent.

Currently, major purchasers of care may require accreditation for a health care delivery system to be eligible for contracts. In addition, many state insurance boards and employer groups have mandated that HMOs have NCQA or URAC accreditation to operate in their state or to be offered to employees, respectively. However, because of the complexities of health care structures, mandatory accreditation can impose a tremendous burden. Accreditation requirements often overlap in national managed care companies or health care delivery systems that perform multiple functions (e.g., a staff model HMO that is also a provider network). Many times, organizations must obtain more than one type of accreditation to satisfy employers, states, and other stakeholders. A behavioral health care carve-out company, for example, may operate in a state that requires URAC accreditation, whereas a multi-state employer group operating within that state may require NCQA accreditation.

The costs of achieving accreditation are also burdensome. The actual cost of the accreditation survey is only part of the burden. The personnel costs and time involved in preparing for accreditation also can also be extensive, and may be prohibitive for smaller organizations. For community-based organizations, the cost of COA accreditation is done on a sliding scale, with a graduated scale based on total agency revenue (COA, 1996a, b).

Cost certainly makes accreditation prohibitive for many small organizations; it also makes the issue of multiple accreditations unrealistic, despite the demands of states and employer groups. Thus, questions are raised about the utility and validity of accreditation. The accreditation industry is faced with pressure to focus its standards on the relevant issues, collaborate with similar organizations, and consolidate the multitude of accreditation standards to reduce overlap and redundancy.

The Accreditation Process

The accreditation process entails generating standards and then comparing the actual delivery of care with the standards. There are at least seven distinct steps:

1.

Measures of performance, also known as parameters, are identified and recommended as standards.

2.

A process of review leads to acceptance of the standards.

3.

The standard is generally tested internally (“alpha” tested) and then tested on a external site (“beta” tested).

4.

After testing, the standards are incorporated into a review process.

5.

Organizations desiring to be accredited apply to be surveyed.

6.

A site review is performed by peer surveyors who examine the inner workings of the organizations against the standards.

7.

Finally, a process of scoring is developed to determine the organization's degree of compliance with each standard and whether the aggregated results reached the threshold for granting accreditation.

These steps are described in the following section.

A standard, according to Donabedian (1982), is a professionally developed expression of the range of acceptable variation from the norm. A standard has also been defined as the desirable and achievable (rather than the observed) performance or value with regard to a given parameter (Slee, 1974).

A parameter is an objective, definable, and measurable characteristic of the process or the outcome of care (e.g., access to behavioral health care within 5 days of a request in a nonurgent situation). Each parameter has a scale of possible values. For example, a geographic access parameter might require outpatient mental health services to be available within 30 minutes of a consumer's home or workplace. Variables would include, for example, traffic patterns in a busy urban setting where traveling 5 miles could take 1 hour during rush hour. Another variation might be in a rural setting, where there is a scarcity of consumers and services and travel time may be longer because of distance.

The development of the current accreditation standards is based on professional consensus. The extent and diversity of opinions into the consensus process vary from agency to agency, as well as from standard to standard. Some agencies use a wide range of experts and elicit public participation, whereas others may use a closed panel of experts and a board review-editing procedure to develop a standard. The scope and relevancy of the standards by this process are dependent on the input and consensus from all the affected parties. The process of standardizing different views from participants in the development of a standard is not clearly outlined in the public information provided by CARF, COA, JCAHO, NCQA, or URAC.

Unless the accreditation process incorporates principles of quality in establishing standards and the survey process, there is a danger of inconsistency, variance, and unreliability. There are many opportunities in the accreditation process for variance in measures, interpretations, and dispositions, leading to disparate outcomes. The process of accreditation is heavily dependent on the strength of the standard as described, the surveyor's interpretation of the standard, and the applicability of the standard to a real situation.

Accreditation standards written with admirable intentions may not lead to consistent interpretation and/or applicability to the real world. For example, COA has included a standard to define the scope of an agency's mission. It states that the primary purpose of an agency is to provide services to meet the needs of the community for protection, maintenance, strengthening, or enhancement of individual and family life and social and psychological functioning (COA, 1996a). This standard demonstrates responsible intentions but is subject to much variation in interpretation by reviewers and variability in what is considered to be the supporting evidence. During the training of surveyors, it would be important to outline the different variables in this standard. Reviewers should be familiar with the variations to the standard so they are able to assess during a review whether the nuances of the agency comply within the boundaries of the standard.

Therefore, the accreditation label is only as good as the process of accreditation, from the development of the standards through the process of scoring. It is important in the accreditation process that:

1.

Standards are developed through a rigorous process of extensive peer consensus, a review of scientific evidence when applicable, and reevaluations of normative data to determine the true range of acceptable variations.

2.

Standards are objective, measurable parameters specific enough to minimize variations in interpretations by reviewers and the public.

3.

Standards are reviewed for their relevance and importance to the goal of accreditation and the integrative needs of the public.

4.

The validity and reliability of a standard must be known and reflective in the scoring, such that those standards with much variability are given less value than those for which there is stronger consensus.

5.

The implementation process is updated frequently, and there is a clear and recurrent process for establishing inter-rater reliability among reviewers.

6.

The final accreditation dispositions are compared with (trended against) acceptable parameters, that is, informed public perception, as a strong indicator of the competency of the accreditation process.