Providing Insights that Contribute to Better Health Policy
Insurance Coverage & Costs Access to Care Quality & Care Delivery Health Care Markets Employers/Consumers Health Plans Hospitals Physicians Issue Briefs Data Bulletins Research Briefs Policy Analyses Community Reports Journal Articles Other Publications Surveys Site Visits Design and Methods Data Files
High-Performance Health Plan Networks: Early Experiences
Issue Brief No. 111
Health plans have introduced high-performance networks to encourage use of network providerspredominantly physician specialistsdeemed high performing on efficiency and quality measures. Early adopters of these networks are large national employers, and, while other employers are interested, actual adoption has lagged, according to a study by the Center for Studying Health System Change (HSC). Enrollment in products using high-performance networks is limited, and objective evidence on the impact on service use, costs and quality is lacking. Early lessons learned indicate the need for effective communication between plans and providers, use of both efficiency and quality measures, industry standards of provider performance, and employer support.
Network Strategy to Lower Costs and Improve Quality
n response to the managed care backlash in the mid-1990s, health plans broadened hospital and physician networks to respond to customer demand for more choice. This strategy sacrificed plans abilities to contain costs because hospitals were able to demand higher payment rates to continue participating in networks and plans were in a weaker position to exclude physicians perceived as being relatively inefficient or lower quality. Recently, however, health plans have developed high-performance networks that encourage enrollees to choose network physicians who score well on measures of efficiency and quality. Health plans analyze claims data to assess network physicians on the basis of efficiency using costs per episode of care, such as treatment of low back pain, and on measures of quality that can be assessed with claims data, such as hemoglobin A1c blood testing for a diabetic. At the market level, if these networks influence enough enrollees to shift to high-performing providers, physicians losing market share might be motivated to improve efficiency and quality to better compete. HSC researchers observed the development of high-performance networks in early 2005 as part of the fifth round of the Community Tracking Study (CTS) site visits. While there was insufficient activity to report on at that time, an in-depth study on high-performance networks was conducted more recently (see Data Source).
High-Performance Network Specifications Differ
igh-performance networks are a recent addition to the tools that health plans and purchasers are using in an attempt to curb costsoften described in terms of efficienciesand improve quality. The prevalence of these networks is limited, as plans have introduced them only in select markets. Where these networks are offered, the number of enrollees using them is largely dependent on employers that include high-performance networks as part of their sponsored health insurance benefit.
High-performance networks typically are not distinct products, but rather an option for use across different product platforms, most commonly preferred provider organizations (PPOs).
The exact specifications of high-performance networks differ across plans. The most common model uses tiered-provider levels, with corresponding enrollee cost-sharing differentials. The first tier consists of the high-performing providers; the second tier consists of the remainder of in-network providers; and the third tier consists of out-of-network providers. Employers often do not differentiate cost sharing between the first and second tiers, offering these networks only as a source of information to their employees about which providers have better performance.
Plans most often target physiciansgenerally specialistsfor high-performance networks; hospitals usually are not included. Assessments of which physicians to include in these networks are conducted by specialty, although some multispecialty group practices have pressed plans to include all of their specialties. Across plans, the criteria used to select which specialties to include focus on those that:
Most plans include specialties such as cardiology, gastroenterology, orthopedics and obstetrics/gynecology, but beyond that, there is less consistency. For example, CIGNAs high-performance network includes approximately 20 specialties, while Blue Cross of California includes all network physicians, including primary care physicians (PCPs). Some plans, such as Aetna and UnitedHealthcare, initially focused on certain specialties but have since added others.
Most plans exclude PCPs from high-performance networks. Plan executives said they do so to avoid disrupting patients established relationship with their PCPa situation that occurs less frequently with specialists. While Tufts Health Plan in Boston does not currently tier PCPs, it does provide enrollees with performance information on these physicians. Tufts has identified blue ribbon PCPsapproximately 20 percent of the plans PCPs who have earned the highest scores based on the plans assessment of costs and quality.
While most plans do not target hospitals for high-performance networks, hospitals are relevant to judgments about physicians because total claims costs per episode of care, including hospital costs and prescription drugs, are used to assess physician efficiency. A benefits consultant said that the focus on physicians is primarily because, with the exception of complex care such as transplants where patients are more likely to choose the facility, people choose their physician who in turn directs them to a hospital. Another benefits consultant noted that because hospitals often have considerable market leverage, they can resist plan initiatives like high-performance networks that threaten their market share. Evidence of this is the lack of a foothold gained by tiered-hospital networks, which plans introduced several years ago. Tufts Health Plan is an exception. Its high-performance network started as a tiered-hospital network and expanded to include physicians.
The methods plans use to profile physicians for inclusion in high-performance networks usually involve some combination of costs and quality. Nearly all plans measure costs using episodes of care which are assigned to the responsible physician.1 They build these using the total costs of care associated with an enrollee, including costs for inpatient and outpatient facilities and prescription drugs. According to one plan executive, To some extent, we realize the physician isnt responsible for costs incurred at the hospital, but we think its good for physicians to think about the efficiency of the hospital they work with too."
There is significantly more variation in the quality measures plans use to profile physicians. Generally, however, the quality assessment is based on physicians adherence to evidence-based medical guidelines and consensus-based quality standards. These assessments tend to be limited to what can be appraised through claims data. For example, the Boston-based Group Insurance Commission (GIC), which administers health insurance benefits for public employees in Massachusetts, requires all contracted plans to include high-performance networks and to use nearly 60 specified quality measures, including those from the Health Plan Employer Data and Information Set (HEDIS), the Agency for Healthcare Research and Quality, and specialty society best practices. The number of measures is expected to increase this year by nearly a third. In contrast, Blue Cross of California uses approximately 20 HEDIS-type measures to assess quality.
Based on the cost and quality data, plans use an algorithm that encompasses both to further assess physicians, although the exact specifications of the algorithms are usually proprietary. Some plans base their high-performance networks predominantly on costs. For others, a physician has to first meet a quality threshold and then is assessed for efficiency.
Profiling is done at the individual physician level, but the results of the individual profiles are then aggregated at the group levelthe level at which contracting decisions, including high-performance network designations, typically are made. There are, however, complexities associated with profiling at the individual vs. practice level. For example, one plan executive explained, We looked at the group and every doctor in the group, but we didnt select doctors that were high performing in a group and exclude others. When you go to a group practice, you dont always see the same doctor. There are cross-coverage issues. A benefits consultant noted that another complexity is the potential for provider pushback at the practice levela group practice may require that all members of the group be designated as high performing or threaten to end its contractual relationship with the plan. Depending on the importance of the group to the plans provider network, such a scenario may present significant challenges for the plan.
Plans vary considerably in the proportion of network providers that are designated high performers. A benefits consultant said, In terms of where they are setting the bar, carriers are all over the place. There is no common rule of thumb of 70/30 or 50/50. All of the carriers that Im familiar with dont go into this with a specific percentage in mind. Plan executives reported a wide range25 percent to more than 80 percentof physicians included in their high-performance networks as compared with their broader networks.
Most plans reported payment rates for physicians in their high-performance networks are the same as for physicians in their larger networks. Plan executives said that while they pay the same for now, they are considering some type of differentiated payment system in the future. One plan, for example, is looking at modifying its incentive program to better reflect quality outcomes. Another plan is looking at how to tie in pay-for-performance metrics.
Not all incentives for physicians inclusion in plans high-performance networks, however, are directly related to payment. Enrollees may obtain information about high-performing physicians through plan directories, which often highlight designated physicians. As one plan executive discussed, Our expectation from a broader view is that by giving members a choice model and incentives to seek care based on the best information we have, it will reward providers with increased membership or allow them to hang on to the market share they have. To the extent we profile and run these programs, physicians become more cognizant and performance should improve. Another plan executive said, We are focusing on how to improve our data capability, recognizing the top 10 percent of providers and giving the bottom percent specific data to improve. If you do that for a period of time and they are not moving, then we may come back and talk about whether they should be in or out of the network.
Employers decide how the benefit structures of high-performance networks are delineated, including any incentives to encourage employees to enroll. An initial decision that employers make is whether these networks are offered to employees as the only network option or to include access to a broader provider network. Large employers tend to offer high-performing networks as a choice of network options.
Employers then decide about enrollee cost-sharing requirements. Benefits consultants and plans often recommend cost-sharing differentials for high-performance networks because they believe that such differentials are key to steering enrollees to higher-performing providers. One plan, for example, generally recommends a 1.5 times differential for copayments$20 for high-performing physicians vs. $30 for other in-network physicians. Another plan makes specialists in its high-performance network available at the lower PCP copayment level.
However, employers are often reluctant to institute large, if any, cost-sharing differentials between physicians designated as high performers and other network physicians. Some respondents believe that large cost-sharing differentials may be premature until there is a higher comfort level with these networks.
Market response to high-performance networks varied widely depending on how the networks were rolled out. Physicians expressed some concern about these networks in most of the target markets. The most common complaint was the lack of communication by health plans. Physicians reported they felt uninformed about their designations as high-performance networks were launched. Furthermore, physicians reported that plans did not always explain fully how performance was assessed or share data about how their performance compared with other physicians.
Physicians also had issues with the methodologies used to determine high-performance designations. They questioned the data quality and whether sufficient sample sizes were used. They also questioned the applicability of the methodologies used across all physicians. In one recently publicized case, the Washington State Medical Association filed suit against Regence Blue Shield alleging the plan used flawed methods and outdated information to exclude physicians from the plans high-performance network.2
Physicians also complained about the lack of standardization in methodologies. Plans define high-performance networks differently, so some physicians are designated as high performing in one plan but not another. This leads to a general skepticism of these networks and the methodologies used to create them.
Plans market share also plays a role in physicians response. Plans with low market share are concerned that physicians will more easily reject their data because of smaller sample sizes or that physicians may just drop out of their network and face few economic consequences if they are unhappy with their designation.
Although there was resistance to high-performance networks in certain markets, plan executives reported closely monitoring those scenarios and adjusting their strategies to minimize conflict. The experience in St. Louis was cited as one such example. In that market, several large providers rebuffed UnitedHealthcares attempt to introduce a high-performance network when they were excluded from the plans high-performance tier.3 Most plan executives expressed the need for effective communication with physicians to minimize conflict, and they recognized that one way to accomplish this was to engage physicians in the process earlier rather than later. Plans that incorporated physician feedback in developing their methodology and that shared data with physicians ahead of time reported a more positive response to their high-performance networks.
A key stimulus for the development of high-performance networks has come from a limited number of large, national employers. While other employers express interest in high-performance networks, not all are prepared to deal with potential employee backlash. Consequently, actual adoption of these networks has not yet matched interest. Employers that have pushed these products aggressively reported feeling as if they are out there alone and everyone else is window shopping as they wait and observe the experiences of early adopters.
High-performance networks have gained a toehold in markets only if large employers have been aggressive in their interest and implementation. This is the case in Boston, Milwaukee and Seattle. In Boston, for example, GIC required contracted plans to develop and offer high-performance networks. In Milwaukee, the business coalition, which represents a large number of employers, was responsible for bringing Humanas high-performance network into the market. In Seattle, Aetna had sizeable national accounts, including Boeing, Costco, Nordstrom and Starbucks, which were active in launching a high-performance network.
Even in these markets, however, enrollment in high-performance networks has not expanded much beyond the initiating employers.
Benefits consultants and plans commented that the adoption of high-performance networks might be higher if employers saw more substantial savings in premiums. Preliminary estimates of savings currently range between 3 percent and 5 percent. To achieve larger savings, more substantive changes in benefit design, such as larger cost-sharing differentials, would be required, which plans and employers appear unwilling to do at this time.
Other concerns also have hampered employers adoption of high-performance networks. One concern of large, multistate, employers is that they want to offer similar health insurance benefits to employees across locations, which is difficult if plans high-performance networks are defined differently. For example, if some plans high-performance networks include primary care physicians, while others do not, this makes high-performance networks less attractive to national employers as part of an overall strategy.
Employers also are concerned that the adoption of high-performance networks may create geographic access problems for employees. For example, employers with employees dispersed across a wide area are concerned that high-performing providers may not be as evenly dispersed. Because of these concerns, some plan executives and benefits consultants noted that while the networks have been developed for large, national employers, they might gain more traction among local employerslike the membership of the Milwaukee coalition. In the future, high-performance networks might be more appealing to small employers who are more price sensitive and less deterred by such issues as network disruption.
To date, because high-performance networks have been offered mostly by large, self-insured employers, they have generally not been on state regulators radar screens. However, one plan executive reported that regulators have raised concerns about these networks creating access issues, particularly in rural areas. The respondent noted that the traditional access perspectivewhether there are enough providers in an arearuns counter to the goal of high-performance networks. He noted that while there should be reasonable access, the question for regulators is whether plans should lower quality standards just to have high-performing physicians in every zip code. He added, If you want to access care from the best doctor, you may just have to drive to the city.
Early Lessons Learned
espondents uniformly agreed that it was too soon to assess the impact of high-performance networks on service use, costs and quality because there is not yet enough volume in these networks. Respondents also said that it is too early to comment on any shifts in physician market share resulting from high-performance networks, although one health plan executive reported early anecdotal evidence of shifting market share among some physicians as a result of the presence or absence of a high-performing designation. Some respondents hoped that high-performance networks would help re-engineer the care delivery system, driving the system toward greater efficiency and higher quality.
While evidence of the effects of high-performance networks continues to develop, evidence does exist from similar efforts that suggest the potential for savings associated with care delivered by physicians who are more efficient and of higher quality. The experience of UNITE-HERE Labor Management Trust Fund, a Taft-Hartley trust providing health care to 120,000 hotel workers and their families in Las Vegas, provides a case in point. In 2003, the Trust screened all 1,800 physicians in its network on the basis of efficiency and quality and as a result, excluded 50 physicians. Physicians deemed as providing higher quality of care also received a special designation in the Trusts provider directory. These network changes reportedly contributed to a significant reduction in the Trusts medical cost trend, and as a result, the cost savings allowed workers to receive salary increases for the first time in several years.4 Despite the evolving nature of high-performance networks, respondents noted several important lessons learned based on their early experiences.
Communication with providers essential
Respondents stressed the importance of communicating openly and honestly with providers when developing and implementing high-performance networks. For example, one plan executive said, Communicate with the provider community about your intentions and give them a forum to give feedback. Dont do something in a black box when youre not willing to explain why you are doing it. He added, Theres no value in gathering data if were not sharing it with physicians and giving them the opportunity to become more efficient.
Several plan executives noted the importance of working with and educating physicians months in advance of either going into a market or introducing a new product or initiative. As one physician representative commented, If you dont include your soldiers who deliver your care during the development and vetting of the product, then you lose from the beginning. The plans that included the physicians at the beginning did a lot better than those who just came at us.
Costs and quality both important
Nearly all respondents said that the success of high-performance networks is dependent on an assessment of both costs and quality. It is possible to have high-quality care delivered inefficiently. A plan executive discussed that networks are typically depicted by a 2x2 graph with costs on one axis and quality on the other. The desire is to get networks moving progressively into the low-cost/high-quality quadrant.
In these early iterations of high-performance networks, much of the focus has been on the cost-side in large part because of limitations in the integrity and sophistication of the quality data. Respondents all agreed that improved data are needed. A plan executive highlighted the problem, saying, From our perspective, were cautious moving forward on differentiating on the basis of quality because we believe that before you can make those assertions that the datasets have to be as valid as possible. A benefits consultant, however, cautioned that to avoid measuring quality is at the plans peril. He said, The last thing anybody will want to see is to open up the business section of the paper to see ‘XYZ carrier switches to cheap doctors.
Standardized industry measures needed
Respondents discussed the need for the industry to move toward more uniform standards of provider performance measurement. One plan executive, for example, expressed concern that physician quality determinations should not be based exclusively on data one plan holds. He commented further that industry collaboration from both the public and private sectors is needed to make informed decisions about the quality of a practicing physician. He said, The most damaging thing would be for competitors to have competing definitions of quality. In that case, we will lose credibility in the eyes of the consumer.
Movement toward any type of industry standardization, however, is not without complication. A potentially major snag arises around antitrust concerns. A plan executive said that in a litigious environment, his plan is cautious about collaborative activities among plans to compare methods used to define networks that might be construed as running afoul of antitrust laws. He added that while they hear from employers that there seems to be some consistency among plans as to which providers are deemed high performing—at least in larger markets—there is still ample opportunity for inconsistencies to occur. Employer groups, such as the Business Roundtable, have pushed for legislation giving insurers access to Medicare Part B data with physicians identified to facilitate more accurate assessments of physician costs and quality through larger samples of patients.5 Several respondents noted that unless there is greater uniformity in the industry, any performance information provided to physicians and consumers will be confusing and of little use. Physicians will be less likely to respond positively—by improving efficiency and quality—if different health plans rate them differently.
Purchaser support needed
One plan executive said that when developing high-performance networks, the plan tries to bring the leading employers with them to talk to providers, emphasizing that the health care system is based on employer-sponsored insurance. The respondent added that providers understand better when the choices are coming from employers. A benefits consultant said that if employers really want high-performance networks, They have to be front and center with plans demanding this and supporting them. They cant stay behind the scenes. Respondents say, however, that employers are often reticent about getting involved. As one plan executive commented, The most fascinating piece of this is that the employer wants it, the health plan builds it, implements it, and then the health plan gets the heat, and the employers who wanted it duck and cover.
It is too early to determine to what extent, if at all, high-performance networks will lead to lower cost and higher quality health care. Initially, the focus has been on the potential gains that might be achieved by moving some enrollees to higher-performing providers, but systemwide, the gains are limited by the capacity of the designated high-performing providers. The larger potential is based on moving sufficient volume away from lower-performing providers that they are motivated to improve, thereby improving the system overall. For this to be realized, health plans have to provide physicians with detailed and accurate information about where their performance is not measuring up. The potential also depends on the degree to which employers institute benefit structures that motivate employees to shift to high-performing providers, which in turn is dependent on whether employees perceive high-performance networks as something of value or as a reminder of what they disliked about managed care.
To examine high-performance networks, HSC researchers collected data at both the national level and in selected markets that have experience with these networks, including Boston, Milwaukee, Seattle and California. Information was collected between January and June 2005 as part of Round Five of the CTS site visits to help identify target markets. The findings are based on semi-structured interviews conducted between May and September 2006 with approximately 20 respondents. Respondents included representatives of national health plans and regional plans offering high-performance networks in the target markets, as well as representatives of providers, employers, and benefits consultants knowledgeable about these local markets. Interviews were conducted by a two-person interview team either in-person or by phone.
Acknowledement: This research was funded by the California HealthCare Foundation.
ISSUE BRIEFS are published by the
Center for Studying Health System Change.