H E A L T H T R A C K I N G : M A R K E T W A T C H. Raising The Bar: The Use Of Performance Guarantees By The Pacific Business Group On Health

Transcription

1 Raising The Bar: The Use Of Performance Guarantees By The Pacific Business Group On Health By negotiating performance targets with health plans, California employers have made a measurable impact on health plan accountability. b y H e l e n H a l p i n S c h a uf f l e r, C a t h e r i n e B r o w n, a n d Ar n o l d M i l s t e i n 134 ABSTRACT: In the Pacific Business Group on Health (PBGH) negotiated more than two dozen performance guarantees with thirteen of California s largest health maintenance organizations (HMOs) on behalf the seventeen large employers in its Negotiating Alliance. The negotiations put more than $8 million at risk for meeting performance targets with the goal of improving the performance of all health plans. Nearly $2 million, or 23 percent of the at risk, was refunded to the PBGH by the HMOs for missed targets. The majority of plans met their targets for satisfaction with the health plan and physicians, as well as cesarean section, mammography, Pap smear, and prenatal care rates. However, eight of the thirteen plans missed their targets for childhood immunizations, refunding 86 percent of the at risk. Al t h o u g h e m p l o y e r s have been purchasing health care benefits for decades, the majority have not asked health plans to guarantee the quality of care and service they provide. Increasingly, however, large employers are requiring health maintenance organizations (HMOs) to document their performance in the areas of member service, satisfaction, and quality of care so that purchasers can determine the value that plans provide. 1 Employers also have begun to hold HMOs accountable, with sizable financial penalties for unsatisfactory performance. Employers goal in using performance measurement and guarantees is to improve the performance of all health plans on a set of measures that are important to purchasers and that represent high-quality care. In this paper we report on the efforts of the largest employer health care coalition in the United States, the Pacific Business Group on Health (PBGH), to negotiate more than two dozen performance guarantees in with thirteen of California s largest HMOs. 2 For the benefit year the HMOs s exceeded $400 million, with more than $8 Helen Schauffler is an associate professor and director of the Center for Health and Public Policy Studies at University of California, Berkeley, School of Public Health. Catherine Brown is a health care consultant specializing in health plan performance measurement and guarantees in San Francisco. Arnold Milstein is the managing director of the health care group and benefits at William M. Mercer in San Francisco and medical director of the Pacific Business Group on Health. Downloaded 1999 ThePeople-to-People from HealthAffairs.org Health on February Foundation, 01, Inc.

2 million at risk for meeting negotiated performance guarantees. THE NEGOTIATING PROCESS PBGH membership has grown from ten large employers (more than 2,000 employees) when it was founded in 1989 to thirty-five large employers in PBGH employers cover nearly three million employees, dependents, and retirees in California and spend more than $3 billion on health care each year. A subset of PBGH employers participate in the PBGH Negotiating Alliance, a group that negotiates HMO s, standard plan designs, and annual performance targets. Unless performance is judged to be strong, the targets are always set so that plans are required to improve. 3 In the Negotiating Alliance represented seventeen employers that jointly provided health benefits to nearly half a million employees, dependents, and retirees. 4 During the negotiation period for there was one merger (between TakeCare and FHP) among the HMOs participating in the negotiations. 5 Eight employers added a total of nine HMOs in, and none of the employers dropped HMOs. Two HMOs threatened not to participate in setting performance guarantees in both 1995 and, but they were told that if they did not participate, they would not be offered by PBGH employers. In its agreements with the Negotiating Alliance, each HMO placed 2 percent of its annual paid by PBGH employers at risk across a uniformly defined set of performance measures. HMOs and employers collaborated on selecting and defining measures to ensure that all of the measures were consequential and reportable. All PBGH employers were invited to attend the performance measures negotiation meetings. Those employers who offered a particular plan to their employees wielded significant influence in the negotiations with that plan. COLLECTING PERFORMANCE DATA The PBGH health plan performance measure set emphasized three areas of HMO performance: customer service, member satisfaction, and quality of care. In this paper we report only on the quality and satisfaction performance measures. We do not report customer service measures because there are inconsistencies in the measurements used by the HMOs, so the data are not comparable. We report data from rather than 1995, the first year that plans made performance measure data available to the PBGH, because the PBGH identified numerous discrepancies in the collection methodologies used by the HMOs in For the benefit year the PBGH negotiated approximately twenty performance guarantees with thirteen HMOs (the number of measures per HMO varied from fifteen to twenty-one). Three of the performance guarantees addressed member satisfaction: satisfaction and dissatisfaction with health plan and satisfaction with physician. The data for these measures were collected as part of the PBGH Health Plan Value Check (HPVC) survey. Six of the performance guarantees addressed quality of care: cesarean sections, childhood immunizations, cervical cancer screening, diabetic retinal exams, mammography screening, and prenatal care. The data for the quality measures were uniformly collected or audited under the auspices of the California Cooperative Healthcare Reporting Initiative (CCHRI), founded in 1994 to provide statistically valid, comparable Health Plan Employer Data and Information Set (HEDIS) quality measures on California HMOs. All data and medical records were compiled and reviewed by one contractor using uniform procedures. PREMIUMS AT RISK AND REFUNDED For the thirteen HMOs participating in the PBGH negotiations, 2 percent of approximately $420 million in HMO s was at risk, for a total of $8.39 million (Exhibit 1). The total amount of refunded in was nearly $2 million, about 23 percent of the total at risk. 135 H E A L T H A F F A I R S ~ M a r c h / A p r i l

3 EXHIBIT 1 Total HMO Premium At Risk And Refunded By HMOs Negotiating With The Pacific Business Group On Health Negotiating Alliance All performance measures Average amount per HMO Total HMO HMO at risk $32.3 million $645,541 ($34,571 $2.97 million) HMO refunded $149,854 ($8,469 $742,270) Total $420 million $8.39 million $1.95 million (23.2% of total at risk) SOURCE: Pacific Business Group on Health, NOTE: HMO is health maintenance organization. 136 MEMBER SATISFACTION TARGETS Although the majority of the plans met their targets for performance on member satisfaction measures, plans individual performance varied considerably (Exhibits 2 and 3). n SATISFACTION WITH PLANS. Just over half of the HMOs met their performance targets for member satisfaction with health plans (Exhibit 2). Six HMOs missed their targets, falling short on average by 4.2 percent (Exhibit 3). However, half of these plans appear to have improved their rates (by 5.3 percent, on average) over their 1995 rates. The seven plans that met their targets exceeded them on average by 4.1 percent. n DISSATISFACTION WITH HEALTH PLAN. Seven of the thirteen HMOs also met their performance targets for member dissatisfaction with health plans. Six HMOs missed their targets by an average of 1.8 percent, although half of these appear to have reduced their dissatisfaction rates below their 1995 rates. The seven plans that met their targets decreased dissatisfaction levels on average by 2.6 percent from n SATISFACTION WITH PHYSICIAN. Eight HMOs met their performance targets for member satisfaction with physicians, and five missed their targets (by 1.6 percent, on average). The eight plans that met their targets EXHIBIT 2 HMO Performance Targets, Actual Performance, And Premiums At Risk And Refunded For Member Satisfaction Measures, Pacific Business Group On Health Negotiating Alliance, Satisfaction measure Health plan satisfaction Health plan dissatisfaction Physician satisfaction Mean HMO target 78.0% (68% 88%) 9.4% (5% 15%) 84.4% (80% 93%) Mean HMO performance 78.3% (67% 89%) 8.8% (4% 14%) 85.8% (81% 91%) Number of HMOs missing target at risk refunded (%) 6 $ 484,919 $263,326 (54.3%) 6 314, ,873 (36.5%) 5 373, ,468 (43%) Total a a a $1,173,462 $538,667 (46%) SOURCE: Pacific Business Group on Health, NOTE: HMO is health maintenance organization. a Not applicable.

4 EXHIBIT 3 Performance Of Thirteen California Health Plans On Negotiated Satisfaction And Quality-Of-Care Measures, Health plan Aetna Blue Shield California Care (Blue Cross) CIGNA FHP/TakeCare Foundation Health Net Health Plan of the Redwoods Kaiser North Kaiser South Lifeguard PacifiCare Prudential Satisfaction measures Plan satisfaction 73.0% Plan dissatisfaction 13.0% Physician satisfaction 86.0% Quality-of-care measures Childhood Cesarean immunization section 80.1% a a % a Mammogram 71.5% SOURCES: California Cooperative Healthcare Reporting Initiative; and Pacific Business Group on Health, NOTES: HMO is health maintenance organization. Twelve of the health plans also negotiated performance targets for cholesterol screening, and nine negotiated targets for diabetic retinal exams. a Data not submitted. Pap smear 77.1% exceeded them by 3.4 percent, on average. n PREMIUM REFUNDED FOR SATIS- FACTION MEASURES. The total at risk across the thirteen HMOs for the three satisfaction measures exceeded $1.17 million (Exhibit 2).The satisfaction measure with the greatest refunded for missed targets was health plan satisfaction (54.3 percent). The total amount refunded for all missed targets on satisfaction measures was nearly $539,000, or 46 percent of the at risk. QUALITY-OF-CARE TARGETS The majority of the HMOs exceeded their negotiated targets for most of the quality-ofcare measures (Exhibits 3 and 4). However, they fell considerably short on childhood immunizations, and nearly half missed their targets on mammograms and Pap smears as well. n CHILDHOOD IMMUNIZATIONS. Eight plans missed their targets for childhood immunizations, falling short by 3 12 percent. The five plans that met their targets exceeded them on average by 9.3 percent, with individual plans exceeding it by 2 19 percent. n CESAREAN SECTIONS. Only four plans missed their targets for cesarean section rates, and they were only about 0.7 percent off target. Nine HMOs reduced their rates below the target by an average of 3.1 percent. n MAMMOGRAPHIES. Six HMOs missed their targeted mammography rates, falling short by 3.4 percent, on average. The seven HMOs that met their targets exceeded them by 5.3 percent, on average. n PAP SMEARS. Although seven plans met their targeted rates for Pap smears, overall the HMOs missed their Pap smear targets by 5.2 percent. Plans that missed their targets were off by as much as 10.7 percent. The seven plans that met their targets exceeded them by 4.9 percent, on average. n PRENATAL CARE. Only four plans missed their targets for prenatal care, falling short by an average of 8.9 percent (one of the plans missed its target by 25 percent). Despite falling short of their targets, two of these plans appear to have increased their rates over their 1995 rates. The nine plans that met their 137 H E A L T H A F F A I R S ~ M a r c h / A p r i l

5 EXHIBIT 4 HMO Performance Targets, Actual Performance, And Premiums At Risk And Refunded For Quality-Of-Care Measures, Pacific Business Group On Health Negotiating Alliance, Quality-of-care measure Mean HMO target Mean HMO performance Number of HMOs missing target at risk refunded (%) Prenatal care 78.2% (53% 90%) 82.9% (57.5% 96%) 4 $ 324,580 $116,634 (35.9%) Mammograms 69.5% (55% 79%) Pap tests 74.1% (61% 85%) Childhood immunizations Cesarean sections 69.5% (53% 86%) 19.0% (13% 23.5%) 70.0% (61% 77%) 73.0% (59% 90%) 67.2% (50% 80%) 17.4% (12% 19.4%) ,380 91,926 (29.4%) 298, ,557 (50.7%) 354, ,270 (85.9%) 299,838 81,256 (27.1%) Total a a a $1,590,095 $762,992 (48%) SOURCES: California Cooperative Healthcare Reporting Initiative; and Pacific Business Group on Health, NOTE: HMO is health maintenance organization. a Not applicable. 138 targets exceeded them by an average of 11.8 percent, with individual plans exceeding them by 4 37 percent. n PREMIUM REFUND FOR QUALITY MEASURES. The total at risk across the thirteen HMOs for the five quality measures exceeded $1.59 million. The quality measures for which the greatest was refunded were childhood immunizations (85.9 percent) and Pap smears (50.7 percent). The quality measure for which the lowest was refunded was cesarean sections (27.1 percent). The total amount refunded for the five quality-of-care measures was more than $762,000, or 48 percent of the at risk. IMPLICATIONS OF HMO PERFORMANCE The PBGH experience suggests that (1) HMOs can be held accountable for meeting negotiated performance targets; (2) existing measures, such as HEDIS, can form the basis for HMO performance measurement and guarantees; and (3) HMO accountability can be rooted in economic incentives tied to unmet targets. A majority of the thirteen HMOs participating in the PBGH negotiations met or exceeded their performance targets for satisfaction and quality of care. More than $8 million was at risk for meeting all of the PBGH performance guarantees, and HMOs refunded almost $2 million for missed targets. The PBGH performance guarantees have encouraged HMOs to expand their capacity to provide the data required for performance assessment and guarantees. In 1995 the PBGH Negotiating Alliance did not negotiate performance targets for prenatal care, cholesterol screening, or diabetic retinal exam because none of the plans could report baseline rates for these measures. In addition, only three of the nine plans participating in the 1995 negotiations could report baseline data on childhood immunization rates. However, by all thirteen HMOs were able to negotiate performance guarantees for prenatal care, mammography, Pap smears, childhood immunizations, and cesarean sections. 6

6 NEW FEATURES n FEWER AND MORE COMPLEX MEAS- URES. In each year of negotiations the PBGH has refined the performance measure set it uses with the HMOs. In the first two years (1995 and ) the measure set contained mostly customer-service measures (seventeen of thirty-seven measures). The employers wanted to ensure that each plan reported and evaluated member services issues. For the 1997 and 1998 benefit years the Negotiating Alliance narrowed the focus of the measure set to audited HEDIS quality measures: These measures made up the largest component (more than 40 percent) of the 1997 and 1998 measure sets. In addition, five satisfaction measures were included for all HMOs. n NEW PARTNERING MEASURES. The PBGH has recently begun to move quality measurement beyond HEDIS process measures toward intermediate outcomes through newly created partnering projects. These projects include follow-up after an abnormal mammogram, a continuous quality improvement (CQI) program for diabetes care, and a drug formulary task force. The PBGH designed the partnering measures to motivate California HMOs to act collaboratively to advance the health industry s collection and exchange of electronic data. Each HMO meets these performance measures by participating in multiyear projects; targets are not set with each plan. n TARGET STANDARDIZATION. The PBGH Negotiating Alliance initially set achievable targets with each HMO based on their willingness to improve performance. The expected level of target improvement varied by HMO, as did the targets themselves. More recently, either measures have had consistent targets because employers felt that all plans should be performing at a specific level, or all plans agreed to a consistent percentage increase based on prior performance. For example, all HMOs have agreed to an 80 percent health plan satisfaction target. With the HEDIS quality measures, the PBGH requires the same level of improvement of all HMOs based on prior performance. In general, the employers and HMOs agree that the degree to which a plan s performance must improve should not be driven by the negotiating skills of the HMO and/or employer representatives at the table. However, one HMO (PacifiCare) argues that the benchmark method for setting targets works against large plans and would prefer that the PBGH return to negotiating planspecific continuous improvement targets. PacifiCare argues that it is much harder for a large HMO with a significant population base to improve performance by a specific percentage than it is for a smaller HMO, and that targets should be customized by HMOs based on prior performance as well as plan size. CURRENT TRENDS In 1998 and 1999 more employers have requested performance guarantees, and performance measurement is being applied in new venues, such as Medicare HMO plans and other health insurance systems, including preferred provider organizations (PPOs) and point-of-service (POS) plans. n STANDARDIZED MEASURE SET. California employers are besieging HMOs with requests for performance measures. Consequently, several HMOs have developed a core or standard set of measures that they offer to most employers that request them. For example, by mid-1998 Health Net had customized performance contracts with thirty-two California employers, each with more than 1,000 Health Net members. If an employer with a smaller membership requests performance measures, Health Net provides its standard set. When creating its core set, Health Net solicits input from its clients and members. Kaiser Permanente developed a core set of performance measures in. That year Kaiser received requests to report and/or guarantee nearly 500 different measures. Kaiser has added measures to the PBGH performance measure set, allowing it to meet as many clients needs as possible with one measure set. Each year Kaiser reviews all measures re- 139 H E A L T H A F F A I R S ~ M a r c h / A p r i l

7 140 quested by employers and updates the core set. The 1998 set includes forty measures addressing care and services. In 1998 Kaiser had performance agreements with nearly sixty California employers. Prior to adopting the core, Kaiser had different targets with different employers, making it difficult to communicate goals internally. With the adoption of a core measure set, Kaiser has established an internal system for reporting progress against targets to the more than twenty departments responsible for meeting customer goals. These internal reports include the amount at risk for each measure, so each department knows what is at stake. Although PacifiCare supports standardized measure sets that enable HMOs to compete fairly, it also recognizes the value of customized measures. For example, Pacifi- Care has seen improved outcomes for diabetic patients as a result of its diabetic intervention program, which promotes diabetic eye exams and blood testing. PacifiCare supports adding a planspecific performance measure for its diabetic program and would agree to set aggressive targets. Customization of the measure set would allow HMOs to demonstrate their commitment to quality in a specific area. n PERFORMANCE MEASURES FOR MEDICARE RISK PRODUCTS. Employers are increasingly requesting performance measurement for Medicare risk products for their retired populations. In 1998 the PBGH began to mandate that those HMOs offering Medicare risk products to PBGH employers also agree to a set of customer service measures, quality measures, and satisfaction measures. n PERFORMANCE GUARANTEES FOR PPO PLANS. Although the majority of PBGH employers had POS and/or PPO performance measures in place in 1994, the Negotiating Alliance opted not to jointly negotiate with these plans for several reasons. First, because The PBGH s approach to holding HMOs accountable for improving quality is beginning to spread across the country. the employers were self-funded for these plans, performance data were employer specific. Second, the employers found it difficult to reach agreement on an amount for financial penalties (similar to the 2 percent at risk for HMOs), as each employer s administrative services only (ASO) fee was composed differently. Hence, 10 percent of one employer s ASO fee could equal 60 percent of another s. The concept of a per member per month (PMPM) fee arrangement was discussed, but agreement was never reached. Third, because of the nature of the ASO contracts and the self-funding arrangements, the majority of the performance measures would address member services only, not quality. Most POS and PPO plans have not yet been subject to performance assessment and guarantees, but there are no absolute barriers to doing so. Employers offering fully funded POS and PPO plans can negotiate performance measures that closely resemble the set used by the PBGH for HMOs. For most customer-service performance indicators, plans have POS or PPO data, but the data are not broken down by employer. For quality-of-care and satisfaction measures, the primary challenge is ambiguity about which providers are accountable for ensuring quality over time. For example, if a person with diabetes receives care from two or more primary care physicians, which one is accountable for ensuring glycosylated hemoglobin tests with the correct frequency? As provider quality accountability systems evolve within POS and PPO plans, such ambiguity will require more temporally limited units of measurement, such as whether intravisit practice guidelines were met. In addition, such plans may have difficulty capturing accurate and complete out-of-network utilization data.

8 CHALLENGES TO ADOPTING PERFORMANCE GUARANTEES n CHANGING MEASURE DEFINITIONS. The PBGH has been challenged by inconsistencies in performance measure definitions. In its first year the Negotiating Alliance discovered that in certain areas each HMO measured performance differently. There was no standard. More recent challenges have arisen from changing HEDIS methodologies. For example, when negotiating quality targets in late 1995 for plan year, data collected using the HEDIS 2.5 methodology were used to set the performance targets. When data were collected in early 1997, HEDIS 3.0 was used. However, HEDIS 3.0 contained numerous updated measure definitions that initially rendered them incomparable with the data collected and targets set based on HEDIS 2.5. As a result, the majority of the plans were faced with collecting both HEDIS 2.5 and HEDIS 3.0 data in The PBGH has addressed this problem through the development of translators so that it is possible to track trends over time, even if measures change. In addition, the National Committee for Quality Assurance (NCQA) endorsed a health plan satisfaction survey tool that the PBGH began using in 1997, but the NCQA has made no commitment to maintaining the survey tool, the wording of questions, or the response categories from year to year. If the survey changes, comparing data from year to year will be difficult. n CAPTURING OUT-OF-NETWORK DATA. One feature of the PBGH negotiating strategy that makes it particularly important for statewide policymakers is that the quality-of-care performance targets are negotiated for all members of an HMO, not just PBGH members, and include all of those enrolled commercially through employer groups and as individuals, as well as Medicaid and Medicare enrollees. Although this approach has the potential to improve health care quality statewide, it also may make it increasingly difficult for HMOs that serve a growing number of Medicaid recipients to improve their performance in the short run. In particular, areas where plan performance has not recently improved include childhood immunization and diabetic retinal exam rates. The Medicaid population, as well as others enrolled in managed care, may be receiving these services outside of the HMO s network of providers, such as from county health departments, community clinics, or vision care providers, and data on this utilization may not be routinely reported back to the plans or recorded in patients medical records. The PBGH does not accept utilization outside of a plan s provider network as an excuse for low performance. Rather, the PBGH holds all plans responsible for tracking use of services, both in and outside of the plan. n HMO CONTROL OVER PROVIDER PERFORMANCE. It is not clear how quickly HMOs will be able to improve deliverysystem performance. Particularly in independent practice association (IPA) and network-model HMOs, plan administrators may feel like puppeteers with no strings. They must change providers behavior to improve HMO performance, but most have little direct control over providers for two reasons. 7 First, individual plans lack leverage because they rarely approach even one-fourth of most providers revenues. Second, annual quality improvement targets are unfamiliar to providers who have little experience in managing quality in a quantified framework. Employers have three options: (1) keeping their focus on HMO results and expecting each plan to find its own path to provider compliance; (2) contracting directly with delivery systems; or (3) facilitating collaboration among plans and providers to create incentives for providers across all plans. CONCLUSION The PBGH s approach to holding HMOs accountable for improving quality is beginning to spread across the country. The U.S. Office of Personnel Management, the nation s largest employee benefits organization, recently announced its intent to adopt a close variant of 141 H E A L T H A F F A I R S ~ M a r c h / A p r i l

9 142 the PBGH s program. Other purchasers are pursuing other models of performance measurement and reward. For example, General Electric focuses on a custom set of performance measures and links plan revenues to improvement targets. The Massachusetts Group Insurance Commission requires rebates from plans unable to design and achieve a major quality breakthrough within a subset of their provider networks. In contrast, General Motors (GM) delivers performance rewards to plans via enrollment volume increases, rather than through price variations. GM accomplished this by differential subsidization of each HMO s based on a multicomponent value rating, encompassing quality, service, and cost. Other purchasers, such as Xerox, implement volume-based rewards for quality by refusing to offer HMOs that have not achieved accreditation by the NCQA. Adopting a system of performance guarantees may be more difficult in markets where HMO penetration is low. However, there is no reason that large employers and private and public purchasers anywhere in the United States cannot pursue this approach with the plans they offer. As HMO penetration increases, the approach becomes more feasible. In 1997 twenty-two states had more than 25 percent of their total population in HMOs. 8 In addition, it is possible for large employers and purchasing groups to negotiate performance guarantees with PPOs and POS plans for member services, some satisfaction measures, and quality of care. The approach, however, is not feasible for small employers to pursue on their own, because they lack purchasing power. Performance guarantees are a powerful way to hold health plans financially accountable for improving their performance on selected measures. However, like medicine and politics, performance guarantees are more of an art than a science at this time. As knowledge of cause-and-effect relationships is acquired, it will be important to test for mediating variables, such as the percentage of total plan and provider revenue that is affected by the guarantees and the effect of electronic health care information systems. Much work is needed to build the science base upon which to establish effective policy for improving health plan performance. To advance the practice of performance guarantees, further research should focus on estimating the independent contribution of performance guarantees on and the most effective and efficient means of improving health plan performance. The authors gratefully acknowledge the support of the Pacific Business Group on Health, the assistance of Emma Hoo with data collection, and Patricia Powers for her helpful comments on earlier drafts of the manuscript. NOTES 1. T. Bodenheimer and K. Sullivan, How Large Employers Are Shaping the Health Care Marketplace, New England Journal of Medicine 338, no. 14 (1998): The thirteen HMOs are Aetna, Blue Shield, California Care (Blue Cross), CIGNA, FHP/Take- Care, Foundation, Health Net, Health Plan of the Redwoods, Northern California Kaiser, Southern California Kaiser, Lifeguard, PacifiCare, and Prudential. 3. H.H. Schauffler and T. Rodriguez, Exercising Purchasing Power for Preventive Care, Health Affairs (Spring ): The seventeen companies are Automobile Club of Southern California, Bank of America, Bank of the West, Bechtel Corporation, Chevron Corporation, Federal Reserve Bank of San Francisco, Fireman s Fund Insurance Company, First Interstate Bank (since acquired by Wells Fargo), LSI Logic Corporation, McKesson Corporation, Mervyn s California, Pacific Telesis (since acquired by SBC Communications), Ross Stores, Safeway, Union Bank of California, Varian Associates, and Wells Fargo. 5. Prior to the merger, the PBGH had offered Take- Care but not FHP. 6. The Negotiating Alliance and several of the HMOs also negotiated targets in for cholesterol screening and diabetic retinal exams. We do not report them here because the targets were not negotiated with all thirteen plans. 7. Kaiser Permanente is an exception to this problem, as each center contracts with a Permanente Medical Group whose providers are on salary. 8. InterStudy, The Competitive Edge, Part II: HMO Industry Report 8.1 (St. Paul, Minn.: InterStudy Publications, April 1998).