Evidence of Innovative Uses of Performance Measures Among Purchasers.

Purchasers are interested in quality of care and performance information for a number of reasons including helping purchasers to make value-based purchasing decisions, holding health plans and providers accountable, and monitoring the care received by enrollees. Interviews with purchasers were conducted in several geographic areas. Purchasers included in the study were large employers, business coalitions, and governments. Objectives of the study were to determine what performance measurement initiatives large purchasers have established, to explore how large purchasers use the results of their performance measurement initiatives, and to examine how these purchasers interact and share information in their respective markets.

Background

Managed care enrollment continues to increase (Thompson, Draper, and Hurley, 1999; Marquis and Long, 1999). Many purchasers have found that offering different types of health care plans from traditional health maintenance organizations (HMOs) to fee-for-service (FFS) plans affords them the opportunity to provide a wide variety of health plan choices to meet their employees'/beneficiaries' needs. Other purchasers have selected managed care organizations (MCOs) to offer employees/beneficiaries more comprehensive coverage they might otherwise have not been able to afford under a traditional FFS model of health care delivery. MCOs have offered purchasers the flexibility of comprehensive benefits, usually at lower costs.

In fact, interest in managed care significantly increased in the late 1970s and early 1980s as purchasers were faced with double digit percentage increases in health care costs. Many purchasers began to explore options to traditional FFS health plans in an effort to reduce costs while maintaining the same level of coverage. Managed care seemed to be the most effective way to reduce costs since managed care principles resulted in controlling over-utilization while offering comprehensive benefits.

However, opponents of managed care believed that capitated payment arrangements, often associated with managed care, gave providers the financial incentive to withhold or to limit medical care. Research to-date has not supported this hypothesis (Miller and Luft, 1994; 1997). Enrollees in MCOs tend to have fewer hospitalizations and shorter lengths of stay as well as higher rates of preventive services when compared with their FFS counterparts. Additionally, these studies found quality of care in MCOs at least equivalent to FFS plans.

Backlash against MCOs and anecdotal evidence strengthen individuals' fears of managed care. Purchasers have been sensitive to employees'/beneficiaries' concerns as well as their own interests in holding MCOs accountable for the health care delivered. Moreover, purchasers often made decisions on which health plans to offer with very little information. Purchasers are beginning to demand health plans that offer value by balancing cost and quality. In order to assess the value of health plans, purchasers are using other information to supplement cost and coverage information, such as performance indicators (Lanser, 1999; Miller and Lowe, 1998). Moreover, while increases in health care costs have recently slowed, many feel that this slow growth is ending and future increases in costs will once again put significant pressure on purchasers to seek value in health care (Berger and Dauten, 1999; Galvin, 1998).

The objectives of the study were to determine what performance measurement initiatives large purchasers have established, to explore how large purchasers use the results of their performance measurement initiatives, and to examine how these purchasers interact and share information in their respective markets. While most purchasers agree that performance measurement provides useful information about health plans, how these purchasers use the information varies (Beauregard and Winston, 1997; Castles, Milstein, and Damberg, 1999; Gabel, Hunt, and Hurst, 1998; Lipson and De Sa, 1996; Lo Sasso et al., 1999; Maxwell et al., 1998; Merrick et al., 1999; Meyer et al., 1998; Schauffler, Brown, and Milstein, 1999). Use of performance indicators ranges from requiring the information from health plans but not using the information for decisionmaking purposes to establishing performance targets health plans must meet. While many purchasers are using performance measurement information, many purchasers are not (Hibbard et al., 1997). This study examined innovative uses of performance indicators among various types of purchasers.

The findings presented in this article represent only some of the performance measurement initiatives established by the purchasers interviewed. Innovative performance measurement activities and uses of the results among the different sites are presented to illustrate examples of how purchasers are responding to the various market conditions and demands to measure quality of health care.

Methods

Findings presented are the result of case studies conducted in late 1999 and early 2000. Sixteen purchasers at select sites were interviewed. Purchasers were identified based on their involvement in performance measurement and did not necessarily have to purchase health care to be included in the study.

The first step in identifying study participants was to select the geographic regions or sites for the study. Sites were selected through the consideration of the following criteria: (1) managed care market penetration, (2) level of performance measurement activities, (3) presence of a coalition, (4) presence of large employers, and (5) geographic region. Sites were evaluated according to these criteria and were selected to obtain diversity in these criteria across the sites. The following sites were selected: California, Michigan, New York, Pennsylvania, Washington, and the Washington, DC metropolitan area.

Purchasers within each of the selected sites were then identified. The study team attempted to include various types of purchasers within the market including relevant government agencies, State Medicaid agencies, coalitions, large employers, and other purchaser organizations active in performance measurement initiatives. Representation from various types of purchasers was critical to obtaining an understanding of performance measurement in response to the market and collaborations between purchasers. Due to the limited study timeframe, only one of two purchasers participated in some sites. However, information gained during initial contacts with purchasers to evaluate study participation was used to evaluate the market.

The interviews focused on the following six areas: (1) overview of the organization, (2) selection of performance indicators, (3) collection and calculation of performance indicators, (4) uses of performance measurement results, (5) barriers to performance measurement, and (6) future plans for performance measurement.

Results

California

California represents a mature managed care market. Managed care penetration is extremely high, and most purchasers have been offering MCOs to employees for more than 10 years. Most employers also participate with the Pacific Business Group on Health (PBGH), a large and active coalition. Member organizations of PBGH account for approximately 95 percent of MCO enrollment across the State.

A large employer was interviewed for this site. This employer covers approximately 100,000 lives, including active employees, their dependents, and retirees, and has been active in performance measurement since the early 1990s. Although active in performance measurement in numerous regions across the country, most of the performance measurement activities undertaken in the California area are coordinated through PBGH. Due to time contraints, PBGH declined to participate; however, supplemental information regarding PBGH initiatives was obtained.

The employer participates in performance measurement activities with its plans throughout the country. For example, the employer places performance standards with all of their self-insured plans throughout the country. Standards range from basic to extensive depending on the market. Currently, standards address mostly administrative processes such as customer service, claims processing, member satisfaction, and provider networks. Administrative fees are at risk for plans that do not meet specified standards.

Several efforts in California illustrate the progressiveness of some of the employer's performance measurement initiatives. Efforts are underway to standardize performance reporting across MCOs in California. This standard reporting improves the ability of the employer to make accurate comparisons across MCOs. In addition, the employer is addressing the link between the providers participating with the MCO and performance through another initiative designed to identify the best and highest performing providers in California and to collect targeted performance indicators for these providers within each MCO. This initiative provides MCOs with the incentive to examine their provider networks and to seek higher performing providers.

Purchasers in California tend to be extremely active in performance measurement. Most employers have been monitoring MCO performance for many years and have established targets for MCOs to achieve. MCOs face financial consequences of not meeting these thresholds for performance. Much of the activity in this area is coordinated through the coalition, which is funded by purchasers. Working with the coalition, these purchasers are able to stay abreast of the current level of knowledge and available resources in performance measurement. Purchasers working together through the coalition have benefited from this coordination and are often on the leading edge of performance measurement initiatives.

Michigan

The Detroit metropolitan area is dominated by the Big Three auto manufacturers. These manufacturers dedicate significant resources to health care research and performance measurement and have been active in this area for many years. Each of the manufacturers has a different philosophy although coordination between them occurs when possible.

Managed care market penetration is moderate in this area. Given the longstanding history with traditional FFS plans for employees of these manufacturers, managed care has been increasing at a slower rate than other markets. Employees, especially older employees, still tend to be resistant to managed care. However, enrollment in MCOs continues to rise and many MCOs operate in this region.

One of the oldest coalitions, the Greater Detroit Area Health Council (GDAHC), operates from Detroit. Although actively participating in initiatives with the Big Three, the coalition has a larger presence in coordinating purchasing and performance measurement activity among other purchasers in the region. Moreover, GDAHC works with purchasers in other areas across the country and is one of the driving organizations of the National Business Coalition on Health (NBCH).

For most Medicaid beneficiaries in Michigan, managed care enrollment is mandatory. Approximately one-half of the Michigan Medicaid beneficiaries reside in Southeast Michigan, which includes the Detroit metropolitan area.

Several types of purchasers were interviewed for this site including a large automobile manufacturer, GDAHC, Medical Services Administration, and the Office of the State Employer. The automobile manufacturer covers 620,000 lives and spent over $1.5 billion on health care in the United States alone. The employer offers HMO, preferred provider organization (PPO), and traditional FFS plans to employees with more than one-half of employees choosing HMOs or PPOs. Given the tendency of this population to choose indemnity coverage, the manufacturer offered financial incentives to enroll in MCOs by making these options less costly for employees. The organization's overall performance measurement goals are to improve quality, make information available to employees for more educated decisionmaking, and eliminate inappropriate health care costs.

GDAHC is one of the oldest coalitions in the country. Within GDAHC, three smaller purchasing coalitions exist that together account for over 450,000 covered lives. Members include both public and private sector organizations. GDAHC initiatives focus on one of four strategic goals: (1) improving the health of the community, (2) restructuring health care delivery, (3) promoting and supporting value-purchasing alliances, and (4) improving data/information for decisionmaking.

The Medical Services Administration (MSA) of the Office of Community Health administers the Medicaid program and is highly active in performance measurement. Approximately 730,000 of the more than one million Medicaid beneficiaries are enrolled in managed care. The majority of Medicaid beneficiaries are required to enroll in MCOs. MSA is committed to monitoring, evaluating, and ensuring the quality of health care in these MCOs.

The Office of the State Employer in the Department of Management and Budget in the State of Michigan is responsible for the health insurance of approximately 60,000 active employees and 35,000 retirees, located in all 83 Michigan counties. Active employees have a choice of traditional indemnity coverage or 1 of 14 MCOs. Retirees can choose from indemnity coverage and two HMOs.

All of the purchasers in this area participate on committees discussing the planning and development of performance measurement activities. The Big Three automobile manufacturers and the United Automobile Workers (UAWs) union are drivers of performance measurement in this region. The major performance measurement in this area is the Consolidated Automobile/UAWs Reporting System (CARS) project. The steering committee consists of representatives from the Big Three automobile manufacturers, the UAWsers, MSA, the Office of the State Employer, GDAHC, the Office of Personnel Management (OPM), RAND, the Foundation for Accountability, the National Committee for Quality Assurance, and HCFA The CARS project was established to ensure consistency of health plan performance reporting across organizations. Although methods and philosophies differ, each participating organization shared the vision of disseminating health plan quality information to employees for more informed decisionmaking. They wanted to ensure that the information being disseminated by each organization was consistent for the health plans. Therefore, they collaborated for the CARS project. The CARS project integrates the Health Plan Employer Data and Information Set (HEDIS®), the Consumer Assessment of Health Plan Study (CAHPS®), and health plan accreditation results. Individual performance indicators are used to produce a summary score in several domains most closely related to the FAACT (Foundation for Accountability) framework including Accreditation Status, Access and Service, Doctor Communication, Getting Better/Living With Illness, and Staying Healthy. Each participant uses the results from this project to produce its own specific report cards for employees by choosing the information and the method of presentation.

Aside from the CARS project, performance measurement activities in this area ranged from only including administrative measures to using performance measurement results to determine enrollment. One purchaser was currently renegotiating all existing health plan contracts. Due to this massive rebid process, the purchaser did not have the resources necessary to fully investigate including many performance measurement requirements in the contracts. New contracts would include performance goals for administrative processes such as issuing identification cards and customer services standards; however, the purchaser would like to include performance measures for health care quality in contracts in the future.

In addition to the collection of HEDIS® and CAHPS® data, the purchasers in this area are involved in the collection and use of other innovative performance measurement initiatives. Another purchaser, as part of the Southeast Michigan Employer and Purchaser Consortium, has produced a hospital profile report. This report contains comparative performance information on hospitals in the areas of medical care, surgical care, and childbirth care. Specific indicators include patients' self-reported experiences and quality/outcomes of care as measured by utilization, length of stay, mortality, complications, and cost of care. The information is intended to be used by employees not only for choosing a health plan by providing information about participating providers but also for choosing a provider for medical care once a health plan decision has been made.

The coalition has developed a health plan survey or request for information. This extensive survey incorporates performance information such as HEDIS® and CAHPS® but also includes information on quality initiatives such as disease management and physician profiling, data collection activities, mental health and pharmacy services, access, and member services. The request for information is in the process of becoming the standard method of collecting information from health plans led by the NBCH and its member coalitions. Additionally, GDAHC has a number of products available online to assist purchasers and consumers in making health plan decisions.

The MSA collects a variety of information in addition to HEDIS® and CAHPS® including performance indicators collected from beneficiaries' medical records by an External Quality Review Organization (EQRO) in six clinical focus areas, information on data quality and collection capabilities through onsite reviews, and information on complaints. Combinations of this quality information and other data are used in an algorithm to score plans in clinical and administrative areas. MSA uses this score to determine how to assign members who have not selected an MCO, also known as auto-assignment. Consequently, higher performing plans are rewarded with enrollment through the auto-assignment process.

Purchasers in this area tend to be progressive in their performance measurement activities. Capitalizing on their common goal of accurate performance measurement, purchasers collaborate, when possible, in order to increase efficiencies, to standardize collection and calculation methods to ensure consistency in the market, and to reduce the burden on the health care system, and to benefit from one anothers' experiences. These relationships have been beneficial to the purchasers in this market.

New York

The New York City metropolitan area has many large employers in addition to numerous MCOs offered in the area. Managed care penetration is high with many MCOs having a lengthy history of enrollment. The coalition in this area is not as active in coordinating purchasers for purchasing or performance measurement initiatives.

A large employer was interviewed for this site. Domestically, the employer covers about 100,000 lives for its salaried employees, as well as another 10,000 retirees, and about 3,000 union workers. (The retirees and union workers receive health benefits through different programs that were not discussed.) In 1995, the employer implemented a managed care strategy and began to provide its employees more choices and incentives to enroll in managed care. While the contracts with health care vendors (health plans) have included performance standards since this time, these standards have evolved to focus primarily on quality issues towards the late 1990s.

The employer has a centralized department that is responsible for designing benefits, selecting health care vendors, and maintaining a relationship with each vendor through assessment of their performance. Standard performance measures are collected from their health care vendors, such as HEDIS® results, as well as a request for information that includes performance indicators not included in standard measurement sets. Examples of such areas included in the request for information are provider networks, operations processes such as claims and enrollment, and disease management. Employer-specific reports are collected when possible. Typically, only the large, national health care vendors can provide this information. The employer also conducts its own satisfaction survey among its employees.

Report cards summarizing performance are distributed to each vendor annually. The report card results for health care vendors providing employer-specific results are benchmarked against standards established by the employer, which are often higher than industry standards. Health care vendors that cannot report employer-specific results are benchmarked against available industry standards such as those contained in NCQA, HEDIS®, Quality Compass, or established by HCFA. Each of the health care vendors receiving report cards has performance targets for each of the areas. Poor performance by the national health care vendors can result in 25-40 percent of administrative fees being withheld. Although the employer's goal is not to assess penalties, withholds will be enforced in order to improve performance.

Although external organizations representing purchasers are not as active in this market as in California, the employers in this area tend to have sophisticated performance measurement initiatives. As a result, the health care vendors in the area have become accustomed to the reporting requirements and are able to meet the demands of the employers. Many of these employers are large and are better able to dedicate resources to these initiatives for aggressive performance measurement of health care vendors.

Pennsylvania

The Pittsburgh area has seen a slower growth of managed care than other areas of the country. Similar to Detroit, this area has had a much greater resistance to managed care. Moreover, this market has been dominated historically by traditional indemnity coverage. When employers began to explore managed care options in the 1990s, the largest health insurer in the area offered employers a fixed price over multiple years thereby underbidding the other existing plans in the area and continuing its dominance of the market. Most employers in this area have typically offered a choice of the types of health plans with very few offering a choice of health plans within the same type until recently.

Several types of purchasers were interviewed: a government agency, two coalitions, a regional organization examining health care quality, and a large employer. Located in Harrisburg, the Pennsylvania Health Care Cost Containment Council (PHC4) is an independent State government agency created in 1986 with the objective of controlling rising health care costs by stimulating competition in the health care market. The primary strategy of PHC4 is to provide comparative cost and quality information to purchasers and to consumers to enable more informed decisionmaking, as well as to providers to identify opportunities for improvement.

One of the primary responsibilities of PHC4 is to collect, analyze, and disseminate data and information about the cost and quality of health care in Pennsylvania. All hospitals and ambulatory surgery centers in the State are required to submit service level, risk-adjusted data to PHC4. PHC4 uses the data to produce public reports about health care in Pennsylvania. Over the past 6 years, PHC4 has produced more than 80 public reports and more than 300 customized reports targeted for hospitals, policymakers, researchers, physicians, insurers, and other purchasers. These reports are usually based on a condition or procedure, such as coronary artery bypass grafts, and indicate processes and outcomes at the provider, facility, and, most recently, the MCO level. PHC4 is also developing MCO report cards that will integrate outcomes measures, process measures, patient satisfaction information, HEDIS®-type information, and financial data. For their initial effort, PHC4 is currently working on a short-term diabetes project with MCOs—the working hypothesis is that an effective diabetes management program would decrease hospitalization and/or have treatments occur in a less costly setting.

The Pittsburgh Business Group on Health is a non-profit coalition of 36 large and mid-size employers. No member employers are affiliated with health care. The coalition represents approximately 400,000 covered lives including active employees, dependents, and retirees. Employers must have 250 employees locally or 1,000 employees nationwide to participate in the coalition and include manufacturing companies, banks, government, and academic institutions with some small trade associations. The coalition was the second market to publish performance results in HealthPages, which provides health information specific to a company's health plans and choices. In addition, the coalition in 1998 and 1999 made available to employers for purchase HMO and point-of-service report cards, which contain information primarily obtained from the HEDIS effectiveness of care domain.

The Three Rivers/Heinz Purchasing Coalition is a part of the Three Rivers Area Labor Management Committee. Among its many activities, this coalition currently organizes public sector employers for health care purchasing. After working with each separately, the coalition brought the Allegheny Intermediate Unit, Allegheny County, and the city of Pittsburgh together to form a purchasing coalition. The Port Authority and League of Municipalities joined later. Through an initial request for proposal process, the coalition evaluated the bidding plans and used the information during the contracting process. The coalition uses the GDAHC request for information process to evaluate MCOs for contracting purposes as well as to identify quality improvement activities specific for each MCO. The coalition planned to implement MCO-specific performance guarantees beginning in June 2000.

The Pittsburgh Regional Healthcare Initiative (PRHI) was convened in December 1997, under the name the Working Together Consortium Healthcare Initiative, and is comprised of more than 60 local leaders in business, health care, and insurers working to improve health care in Southwestern Pennsylvania. The primary strategy of PRHI is to hold providers accountable for the outcomes of care they produce and to provide them with the information necessary to create incentives to drive systematic quality improvement. A key component is the focus on quality improvement, rather than punishment for not meeting expectations, through the creation of partnerships between purchasers and providers.

PRHI uses data collected and analyzed by PHC4 to produce reports on provider performance including process and outcome measures. The five pilot areas in which these reports are being produced are: (1) cesarean section (report completed), (2) hip and knee replacement (report completed), (3) circulatory disorders and ischemic heart disease, (4) diabetes as a secondary condition, and (5) depression. PHRI has implemented a three-step process for the use of these quality reports. The first step is to disseminate the reports to providers to address current performance. The next step is to remeasure the following year to determine what providers have done to improve performance. The third step is to remeasure a third year to identify providers who are not responding. PRHI is encouraging purchasers to consider restructuring payment and reimbursement to health plans, providers, and institutions to pay a more favorable rate for high quality and a lower rate to those who are not responding to improve poor quality.

Another area of focus for PRHI is clinical performance improvements to increase quality and financial performance. The target areas selected for this study are medication errors and hospital-acquired infections. These target areas were identified prior to the release of the Institute of Medicine's highly publicized “To Err is Human: Building a Safer Health System” (Kohn, Corrigan, and Donaldson, 2000).

Finally, a large manufacturer was interviewed for this site. The employer has 33,000 active employees, predominantly in the United States. In general, the employer examines HEDIS® and HEDIS®-like performance measures when making contracting decisions in addition to local reports on quality disseminated by regional organizations. The employer relies on an external consultant to evaluate MCOs for contracting, and the MCOs must meet minimum thresholds of performance in order to be considered for contracting by the employer.

Performance measurement initiatives in this area are fairly recent and are continuing to evolve. Purchasers have a wide variety of performance information available to them such as HEDIS® and other process performance measures in addition to outcomes measures produced for specific conditions and procedures using the PHC4 data. This site is unique in terms of the availability of risk-adjusted inpatient and ambulatory surgery center data being collected and analyzed. In general, purchasers feel this information does not meet their needs for evaluating overall health care by MCOs and for holding MCOs accountable; however, they continue to actively participate in the performance measurement activities across organizations as these initiatives evolve and become more relevant to MCOs.

Washington

The Seattle metropolitan area has numerous large employers active in performance measurement. Managed care penetration in this area is moderate. There is not a coalition in this area representing purchasers, and purchasers tend to develop performance measurement initiatives independently.

A large employer and the State Medicaid agency were interviewed for this site. The Medicaid agency also discussed its sister agency, the Health Care Authority (HCA), which administers health care benefits to State employees/retirees and low-income adults. The Washington State Medical Assistance Administration (MAA) administers Medicaid benefits to more than 700,000 beneficiaries. Approximately 400,000 of those beneficiaries are enrolled in MCOs.

MAA collects HEDIS® results from MCOs, conducts studies with the EQRO, and sponsors a statewide survey using the CAHPS® and FAACT survey instruments. Selected results are disseminated to stakeholders including county committees comprised of provider representatives, advocacy group members, and health department staff as well as MCOs. MCOs must develop and implement an MAA-approved corrective action plan for performance improvements based on the results. MAA also uses selected study results and some HEDIS® measures to determine auto-assignments for new enrollees.

MAA and the Health Care Authority joined together in 1998 for a joint purchasing process. MCOs responded to a single request for proposal for both agencies. The request for proposal included performance indicators examining internal and external quality. The agencies will be expanding the performance standards to be included in new contracts.

Worldwide, the large employer and its subsidiaries employ 225,000 active employees and cover approximately 1 million total lives, which include active employees, retirees, and their dependents. The following performance indicators are collected: HEDIS®, employer-specific member satisfaction survey results (most MCOs use CAHPS®), and administrative measures such as customer service and claims processing responses. Minimum performance standards are established, and MCOs are monitored against those standards on an annual basis. The employer is also beginning to include performance guarantees in contracts. This year, the guarantees will focus primarily on administrative processes; clinical and quality performance guarantees are expected to be added the following year.

While purchasers in this area are active in performance measurement, the level of sophistication is still evolving. Without the presence of a coalition or other external organization representing purchasers, purchasers in this area tend to conduct their performance measurement activities independently. Collaborative efforts between purchasers are rare in this area. Purchasers in this area follow industry trends, and their performance measurement initiatives will continue to advance.

Washington, DC

The Washington, DC area is dominated by Federal Government workers. Many MCOs operate in this area with service areas extending into Maryland, Virginia, and Delaware. Additionally, many national organizations addressing health care quality are located in this area.

The OPM, Washington Business Group on Health (WGBH), and the NBCH were interviewed. OPM manages the Federal Employees Health Benefits (FEHB) Program, the Federal Employees' Group Life Insurance Program, and the Civil Service and Federal employee retirement systems. In 1999, FEHB served more than 9 million Federal employees, retirees, and their family members. OPM's primary goal is to provide employees and retirees with quality information for more informed decisionmaking. During each open enrollment period, OPM issues a comprehensive guide with performance information, consisting mainly of consumer satisfaction measures, to employees and retirees. Unlike other purchasers, this guide contains performance results for both FFS plans and MCOs. In addition, OPM sponsors an interactive tool on its Web site to assist employees and retirees in choosing a health plan. This tool weights various health plan characteristics, such as choice of a provider and out-of-pocket costs, and produces a list of suggested health plans for the users based on their preferences for these characteristics.

WBGH is a national, non-profit organization dedicated to the analysis of health policy issues for its larger employer members. WBGH's 150 members include Fortune 500 and large public sector employers and provide health care coverage to more than 30 million employees, retirees, and their dependents in the United States. WBGH concentrates on policy level issues as compared with issues in implementing performance measurement initiatives. Through annual conferences, forums, and member-specific projects, WBGH focuses on health-related policy issues affecting its members. Examples of efforts include presenting best practices for disseminating quality information to employees, improving communication between businesses and community agencies, and implementing mental health parity regulations into benefits packages.

The NBCH is a national organization that brings regional business coalitions together. In 1999, seven business coalitions collaborated to use a common request for information, developed by GDAHC. These members of the NBCH include: GDAHC, PBGH, the Midwest Business Group on Health, the Central Florida Healthcare Coalition, the Buyers Health Care Action Group, the Colorado Business Group on Health, and the Health Policy Corporation of Iowa. One goal of this initiative is to develop a data warehouse of the request for information results that employers can use to examine regional differences in MCO performance.

These organizations and others in the area provide direction and develop standards for MCOs and purchasers in performance measurement. Moreover, these organizations often unite local organizations and efforts being conducted regionally.

Barriers

Purchasers in each of the various sites identified many of the same barriers to performance measurement. HEDIS® and CAHPS® have become industry standards for performance measurement. Many purchasers collect this information to evaluate MCOs, to disseminate performance information to beneficiaries and employees, and to monitor performance against standards. However, many purchasers felt that these performance measurement systems do not meet their needs for several reasons. First, these measurement systems do not provide comprehensive information about all aspects of an enrollee's health care. While these measures provide useful information, some aspects of care are left unmeasured. Second, these measurement systems generally measure structure and processes rather than outcomes. Purchasers feel that measuring outcomes is essential to understanding the true quality of health care. Third, many MCOs cannot produce employer-specific results using these measurement systems. Purchasers are interested in using these measurement systems to monitor MCO performance and to improve the overall health of their employees and beneficiaries. Fourth, these measurement systems produce results at the health plan level. Many purchasers feel that the providers rendering care determine the quality of health care rather than the health plans that simply administer benefits. Therefore, they believe that performance measurement should focus on providers rather than health plans. Finally, HEDIS® and CAHPS® are intended for MCOs since MCOs assume the health care responsibility for a given population of enrollees. However, many purchasers also offer indemnity and PPO products and must evaluate these plans along with MCOs. These purchasers have expressed a strong interest in performance indicators applicable to various types of health care delivery systems.

Another barrier identified by most of the purchasers is the level of resources required for performance measurement activities. Many purchasers faced time, staffing and financial constraints, and lacked the resources necessary to implement all of their desired performance measurement initiatives. For example, many of the performance measurement activities involve medical record reviews since the necessary information is not available through administrative data sources such as claims and encounter data. Medical record reviews are both costly and time consuming in addition to being a disruption to providers who must provide the medical records.

Many purchasers relied on MCOs to provide information and data. The information provided often did not meet purchasers' needs due to lack of sophistication of the MCOs' information systems, data completeness issues due to providers not submitting encounter data to the MCO, and unavailability of data in existing information systems. Similarly, plans and purchasers have difficulties linking data from different data sources. For example, purchasers who collect provider level data often have problems linking the data to the appropriate health plan. In addition, many purchasers had concerns over the validity of the data and information received from health plans. Although many purchasers conduct reasonability analyses and require audited data (e.g., certified HEDIS® compliance audit of HEDIS® results), lack of knowledge about existing audit programs as well as the lack of resources to validate information concerns purchasers.

Finally, most of the large employers interviewed are national companies with employees in virtually every State. While employers participate in the local and regional efforts of coalitions and other external organizations involved in performance measurement, these national employers often make health care decisions on a national basis and cannot integrate local and regional performance measurement results. These employers desire performance measurements initiatives that are standardized on a national level.

Future Directions

In general, purchasers are continuing to refine their performance measurement initiatives. Examples of future plans for purchasers include integrating additional data sources, integrating performance measurement systems, and expanding the scope of current initiatives to include more measures. Purchasers will continue to address the barriers discussed in the previous section.

Unfortunately, many purchasers lack the expertise or tools to overcome these barriers and rely on the evolution of performance measurement within the industry. Given the benefits that many purchasers gain from collaboration with other purchasers, sharing best practices is essential to the advancement of performance measurement. In addition, researchers must help purchasers to address the barriers they face. Areas in which researchers can focus their efforts are developing standard methodologies for integrating existing measurement sets and developing new measures that can be reported at the health plan and provider level as well as outcomes measures.