Medigap preferred provider organizations: issues, implications, and early experience.

The Health Care Financing Administration is sponsoring the Medicare Physician Preferred Provider Organization (PPO) Demonstration to assess the feasibility and desirability of including a PPO option under Medicare. Two sites are currently operational. At one site, Blue Cross and Blue Shield of Arizona is offering a PPO linked with a medigap insurance plan. This "medigap PPO" and its initial experience are described, and a preliminary assessment of the viability and effectiveness of medigap PPOs nationally is provided. Impediments to the development and effectiveness of medigap PPOs are identified and possible government actions discussed.

Introduction

Preferred provider organizations (PPOs) are a major new approach to cost containment in the private health care market. A PPO is created through a set of contractual arrangements between an insurer or other third party and a group of health care providers. The basic objective is to create a network of cost-effective providers and to channel patients to those providers through financial incentives such as reduced cost-sharing. PPOs differ in their approaches to cost containment, but commonly used approaches include selective contracting with low-cost providers, obtaining price discounts from providers, and applying utilization management within the provider network. The main incentive for providers to participate in a PPO is the potential for increased patient volume.

The rapid growth of PPOs in the private sector and the widespread expectation that PPOs can reduce health care costs have prompted interest in potential applications to the Medicare program. To assess the feasibility and desirability of including a PPO option under Medicare, the Health Care Financing Administration (HCFA) is sponsoring a Medicare PPO pilot demonstration. Two PPOs are currently participating in the demonstration: Blue Cross and Blue Shield of Arizona (BCBS/AZ) in Phoenix and CAPP CARE in Orange County, California. BCBS/AZ offers a PPO linked with a Medicare supplemental insurance, or medigap, plan, which it is marketing to individual Medicare beneficiaries. CAPP CARE has implemented a non-enrollment-model PPO, not linked with medigap insurance. CAPP CARE does not enroll beneficiaries, but applies its utilization management procedures whenever beneficiaries obtain care from a network physician.

In this article, we describe the medigap PPO offered by BCBS/AZ, discuss its early experience, and provide a preliminary assessment of the viability and effectiveness of medigap PPOs nationally. Our conclusions are based on information obtained through onsite interviews with BCBS/AZ staff, interviews with other insurance industry representatives, and a review and synthesis of relevant prior research findings.

Arizona model

Senior Preferred, the medigap PPO offered by BCBS/ AZ, gives enrollees the financial protection of medigap insurance, but differs from traditional medigap plans by giving enrollees financial incentives to select providers from within a specific network. To attract enrollees to its medigap PPO, BCBS/AZ charges a lower premium than it charges for its standard medigap plan and provides coverage for additional services such as vision and hearing care.

BCBS/AZ developed a medigap PPO to increase its market share and to be more competitive in the medigap industry. Offering a medigap PPO was a relatively low-cost step for the company, as it had already developed a provider network—the most costly phase of PPO development—and had established utilization review and quality assurance programs through its private sector PPO. Also, it already had experience in dealing with the Medicare population through its standard medigap plan. BCBS/AZ developed and implemented its medigap PPO with no formal support from HCFA, and HCFA is incurring no administrative costs for the PPO's operation. The medigap PPO is regarded as a standard medigap product by Arizona's State insurance department, which has regulatory jurisdiction over medigap products sold in the State.

BCBS/AZ currently offers Senior Preferred in the two most populous Arizona counties, Maricopa and Pima, which include the Phoenix and Tucson metropolitan areas, respectively. The total population of these counties is about 2.6 million, of whom 333,000 are Medicare beneficiaries. The beneficiaries in these counties are wealthier than average and incur higher-than-average Medicare charges. The Arizona market overall is quite experienced with managed care products in the private sector and in recent years has experienced a proliferation of PPOs and an influx of enrollees from indemnity plans into PPOs. About 10 percent of Maricopa County beneficiaries are enrolled in FHP, a Medicare health maintenance organization (HMO).

A major challenge in the Medicare context—where the range of available incentives is limited and established relationships with a physician are often strong—is designing an economically viable medigap PPO product that will attract enrollees and encourage them to use network providers. The main incentive to attract enrollees to BCBS/AZ's medigap PPO is a lower premium—about 31 percent lower than BCBS/AZ's standard medigap plan—plus coverage for such additional services as mental, emotional, and lifestyle counseling and vision and hearing exams. Discounts on eyeglasses and hearing aids are also available to PPO enrollees. The PPO's premium is also 3 to 35 percent lower than the premiums of other competing medigap plans in the area.

The provider network includes both physicians and hospitals. The incentive for enrollees to select a physician from within the network is that network physicians have agreed to accept assignment on all claims for PPO enrollees. If enrollees obtain care outside the network from a physician who does not accept assignment, they are not covered for any charges above the Medicare-approved charge. The incentive to obtain hospital care within the network is that the plan fully covers the Part A deductible only if care is received at a network hospital; the deductible is not covered if care is received at a non-network hospital, except in the case of an accident or medical emergency. BCBS/AZ is able to offer this benefit by getting network hospitals to agree to waive the Part A deductible for PPO enrollees. These incentives do not involve any changes in the basic Medicare benefit structure, so they do not require the involvement of the carrier and intermediary.

The BCBS/AZ medigap PPO tries to generate cost savings through more conservative treatment patterns by its network providers and lower provider costs. Only physicians in BCBS/AZ's private sector PPO network are eligible to join the Senior Preferred network. As one measure to contain costs in its private sector PPO, BCBS/ AZ carefully selects and profiles network physicians. To that end, it maintains a data base on physician activity, closely scrutinizes utilization patterns and quality measures, and establishes financial parameters for each specialty and penalties for cost outliers, all based on data from the private sector PPO. Physicians with claims costs that greatly exceed the norm are investigated and warned, and those who do not modify their behavior are dropped from the network. The physician profiling system does not include claims data on Senior Preferred enrollees. The firm believes that, because all physicians in its Senior Preferred network are also in its private sector PPO network, its profiling activities based on private sector data will enable it to maintain a network of cost-effective providers for its medigap PPO. The incentives for physicians to join the medigap PPO network—the potential for increased patient volume and the direct payment of claims—are such that BCBS/AZ reports a waiting list of physicians eager to join the network in most specialties.

In its private sector PPO, BCBS/AZ performs other utilization review activities, such as prior authorization for hospital admissions, mandatory second opinions for selected surgeries, and random retrospective review. These activities are not part of the medigap PPO. Including prior authorization in the medigap PPO would require an interface between BCBS/AZ and the primary payer (the Medicare program) that does not currently exist. For a medigap PPO to conduct prior authorization, it must have a mechanism to deny or reduce payments when services are provided without prior approval. Responsibility for approving claims under Medicare currently rests with the carriers and intermediaries; medigap insurers are not involved. And the regulations governing medigap insurance in most States (discussed later in this article) would prohibit medigap PPOs from denying or reducing payments on claims approved by carriers or intermediaries. There are two other reasons BCBS/AZ does not employ utilization review in its medigap PPO: Much of the retrospective review the firm does in the private sector would duplicate the procedures employed by the Medicare peer review organization (PRO), and no diagnosis data from Part B claims are available for more sophisticated retrospective reviews.

BCBS/AZ also has a quality assurance program to guard against the possibility that its emphasis on cost containment might lead to lower quality care. A key component of this quality assurance program is the medical office review and evaluation program, which provides, through claims review and onsite visits, examinations of: the content of medical records and claims; general office facilities, safety, and hygiene; and laboratory and X-ray facilities and procedures. The quality assurance program, like physician profiling, is based on the private sector PPO with spillover effects for Medicare. BCBS/AZ's quality assurance and utilization review activities are in addition to the quality and utilization review functions performed by the Medicare carriers, fiscal intermediaries, and PROs.

Early experience

BCBS/AZ began marketing Senior Preferred, its medigap PPO, in late 1988. The PPO was marketed to individual beneficiaries using various strategies, including radio, television, and print advertisements and a direct-mail campaign. (PPOs in the private sector are typically marketed to, or initiated by, employers.) Enrollment in Senior Preferred remained low throughout 1989; there were only 836 enrollees at year's end.

Enrollment jumped to 5,443 by April 1990. BCBS/AZ attributes this influx of enrollees to the price difference between its standard medigap plan (called “Senior Security”) and its medigap PPO. In 1989, the premium for Senior Preferred was about 24 percent lower than that for Senior Security. This difference increased to 31 percent in early 1990 when, together with much of the medigap industry, BCBS/AZ raised the premium for its standard medigap plan by 44 percent because of the repeal of the Medicare Catastrophic Coverage Act (Public Law 100-360) and trends in the cost of claims. The Senior Preferred premium was raised by only 24 percent. Most of the beneficiaries who enrolled in the medigap PPO in early 1990 probably switched from BCBS/AZ's standard medigap plan, because Senior Preferred was not being widely marketed to other beneficiaries at that time.

Because of its limited success attracting enrollees in 1989, BCBS/AZ conducted substantial market research. This research revealed that the beneficiaries most receptive to a medigap PPO are older, lower income, and less educated seniors (once they understood the plan)—the opposite of BCBS/AZ's previous target group. A new marketing campaign was planned but has not been implemented as of this writing.

BCBS/AZ has drawn the physicians for its Senior Preferred network from the 2,600-physician network of its private sector PPO, Preferred Care. In Maricopa County, about 10 percent of the physicians are in the Senior Preferred network and in Pima County about 20 percent are Senior Preferred physicians. Senior Preferred has 15 hospitals in its network, representing one-quarter to one-third of all the hospitals in the two counties.

Future of these organizations

To assess whether medigap PPOs will become a major presence nationally, we explore three major issues:

What incentives are available for beneficiaries to enroll in a medigap PPO and, once enrolled, to use network providers?

What incentives are there for providers to participate in a medigap PPO?

How effective will medigap PPOs be in controlling costs?

Our preliminary conclusions about the feasibility and effectiveness of medigap PPOs are based on a review of the relevant literature and on interviews with insurance industry representatives. The results of our preliminary assessment are summarized in the following sections.

Medigap insurance industry

The private medigap insurance industry has existed nearly as long as has the Medicare program. According to recent estimates, more than 70 percent of Medicare beneficiaries have private health insurance to supplement their regular Medicare coverage (Nelson et al., 1989; Rice, Desmond, and Gabel, 1989). Medigap insurance is supplied by Blue Cross and Blue Shield plans, which have a market share of about 40 percent, and by commercial insurance companies, which have virtually all the rest of the market (Meade, 1990). In 1988, more than 45 Blue Cross and Blue Shield plans and more than 280 commercial insurers offered medigap plans (National Association of Insurance Commissioners, 1990).

The States have regulatory jurisdiction over the insurance industry. However, many States voluntarily adopt Federal guidelines for medigap insurance, which were established under legislation passed in 1980 known as the Baucus amendment. The Federal guidelines are based on standards established by the National Association of Insurance Commissioners (NAIC) and are contained in the NAIC model regulations on medigap insurance. The Baucus amendment was passed after a special task force appointed by the NAIC in 1978 found evidence of marketing abuses in the medigap industry.

The NAIC model regulations specify a minimum set of benefits that medigap plans should offer and include provisions to protect consumers from marketing abuses. Of particular relevance for medigap PPOs, medigap plans are required to cover the full coinsurance amount (generally 20 percent) for Medicare Part B expenses after the patient meets the annual Part B deductible. In addition, medigap insurers must cover either all or none of the Part A deductible for inpatient hospital care and the full Part A coinsurance on Medicare-covered stays.

Beneficiary incentives

To be successful, a medigap PPO must design incentives to encourage enough Medicare beneficiaries to enroll in the PPO and then, once enrolled, to obtain most of their medical care from network providers. To be most effective, a medigap PPO must enroll not only patients of network physicians but also other beneficiaries, and then induce this latter group to switch to a network physician. Thus, network physicians can increase their patient load, which is their primary incentive for participating, and contribute to cost savings through more cost-effective treatment of these new patients.

Enrollment incentives

The two main types of incentives that would encourage beneficiaries to enroll in a medigap PPO are: a lower premium than other medigap plans charge for comparable benefits, and coverage for additional services not offered by comparably priced plans. BCBS/AZ offers both of these incentives in its medigap PPO. A medigap PPO's ability to attract enrollees will also be affected by the size and composition of its provider network and the penalties for using out-of-network providers. In general, a PPO that imposes mild or moderate penalties for out-of-network use will be more attractive to beneficiaries than one that imposes severe penalties. A relatively large physician network offers enrollees more options and means many beneficiaries can join the PPO without switching physicians. (But including proportionately more area physicians in the PPO network may diminish the PPO's ability to control costs, particularly for PPOs that emphasize physician screening and monitoring as a means of cost containment.) Finally, a medigap PPO's ability to attract enrollees will depend on whether network providers have a reputation for delivering high-quality care and are conveniently located.

The potential market for medigap PPOs is Medicare beneficiaries currently enrolled in a traditional medigap plan, those not currently enrolled in a medigap plan, or those enrolled in a Medicare HMO. About 71 percent of all beneficiaries not enrolled in a Medicare HMO are covered by medigap insurance; 20 percent rely exclusively on Medicare for their insurance coverage; and 9 percent are eligible for both Medicare and Medicaid (Nelson et al., 1989). Only 3 percent of Medicare beneficiaries are enrolled in risk-based HMOs.

Among beneficiaries with some type of medigap coverage, a medigap PPO is most likely to appeal to those with lower incomes, because the primary advantage of a medigap PPO over a traditional medigap plan is a lower premium. Using data from a national survey of about 2,000 Medicare beneficiaries conducted by Mathematica Policy Research, Inc. for the Physician Payment Review Commission (PPRC), we estimate that 4.7 million beneficiaries nationally (representing 14 percent of the total Medicare population) have some medigap coverage and have incomes below 150 percent of the poverty level.

Beneficiaries currently covered by a traditional medigap plan will switch to a medigap PPO if they believe that the advantages of the PPO—the lower premium and any additional services covered—outweigh the disadvantages. The enrollment experience of BCBS/ AZ in early 1990 suggests that many beneficiaries will switch from a standard medigap plan to a medigap PPO if the premium differential is large enough. But many beneficiaries might remain in their current medigap plan, despite the premium differential, for several reasons. Beneficiaries may be unfamiliar or uncomfortable with the concept of a network, for example, and may be reluctant to enroll in a medigap plan that tries to influence their choice of provider. Some beneficiaries may have a strong attachment to a physician outside the network and may be reluctant to switch physicians to obtain the financial benefits of the PPO. Beneficiaries may also be reluctant to enroll in a medigap PPO that tries to control utilization through prior review, a concept that is likely to be unfamiliar to many Medicare beneficiaries and may cause concern about potential barriers to needed care.

The second potential market for medigap PPOs is beneficiaries who are not currently covered by a medigap plan or Medicaid and thus rely on Medicare as their sole third-party payer. About 20 percent of all Medicare beneficiaries fall into this category, or 6.7 million beneficiaries nationally. The beneficiaries most likely to rely on Medicare as their sole third-party payer are in the lower income brackets; about one-half of them report that their main reason for not having medicap coverage is the high cost (Nelson et al., 1989). So by offering a lower premium than traditional medigap plans, a medigap PPO might attract part of this population.

The third potential market for medigap PPOs is beneficiaries who are currently enrolled in Medicare HMOs. About 1 million beneficiaries, or 3 percent of the Medicare population, are currently enrolled in HMOs under a risk contract with HCFA. Enrollees of Medicare HMOs are beneficiaries who have already expressed a preference or tolerance for managed care in exchange for the advantages of HMO membership—lower (or no) premiums and more coverage than under other medigap insurance. Some of these enrollees might switch to a medigap PPO, given the opportunity, because the PPO has many of the advantages of an HMO but more flexibility of choice in providers.

Enrolling many Medicare beneficiaries in medigap PPOs nationally is likely to require substantial marketing efforts to educate consumers about the PPO concept generally and about the specific features of a given plan. The task of educating Medicare beneficiaries so they can make informed choices about enrollment in medigap PPOs and can understand the financial penalties for receiving care outside the network should not be underestimated. A number of studies suggest that Medicare beneficiaries have a poor understanding of insurance concepts generally and of the benefits covered under Medicare and their current medigap insurance plan (Cafferata, 1984; McCall, Rice, and Sangl, 1986; Nelson et al., 1989). The latter two studies found that low-income beneficiaries and those without medigap insurance are the least knowledgeable. In other words, those most financially vulnerable to high medical bills—who could benefit most from a low-cost alternative to traditional medigap insurance—are likely to be the most difficult to educate.

Consumer ignorance about Medicare and medigap insurance has important implications for the introduction of a medigap PPO option. First, medigap insurers offering a medigap PPO must develop marketing materials that clearly explain the PPO benefit package, particularly the financial incentives to use network providers. If beneficiaries are not adequately informed about the PPO, some who would benefit from PPO membership may not enroll, and others may enroll without adequately understanding the financial incentives to receive care within the network. Enrollees who incur higher-than-expected out-of-pocket costs because they misunderstand the penalties for out-of-network use could become dissatisfied with the PPO and disenroll.

The NAIC model regulations for medigap insurance include a number of provisions to protect consumers from marketing abuses (U.S. General Accounting Office, 1990). But the NAIC should consider whether introducing medigap PPOs into the market would require additional standards for monitoring medigap PPO marketing practices. Regulatory standards may be necessary, for example, to ensure that beneficiaries are fully informed about providers included in the PPO network and the financial penalties for out-of-network use. Standards may also be required to ensure that enrollees are fully informed of the utilization management procedures the PPO uses, particularly prior review.

Incentives to use network providers

Once beneficiaries are enrolled in the PPO, the PPO's ability to control costs depends on the extent to which enrollees obtain care within the network. In their private lines of business, PPOs try to channel enrollees to network providers by having them pay a higher share of costs for care received outside the network. But medigap PPOs are currently limited in the extent to which they can impose penalties for out-of-network use, because the NAIC model regulations require medigap plans to cover the full 20-percent coinsurance under Medicare Part B. Currently, the best way to channel enrollees to network physicians is to require network physicians to accept assignment on all claims for PPO enrollees and to provide no coverage for balance billing incurred outside the network. (Physicians who accept assignment on a Part B claim agree to accept the Medicare-approved charge as payment in full.)

This is a relatively weak incentive, because most Part B claims are currently accepted on assignment. In the past decade, the proportion of Part B charges accepted on assignment has increased dramatically, from 50.9 percent in 1978 to 80.5 percent in 1988 (Physician Payment Review Commission, 1989). Much of this increase occurred after the implementation in 1984 of the Participating Physician and Supplier Program (PAR), which provides incentives for physicians to agree in advance to accept assignment on all claims. Forty-five percent of all physicians signed participation agreements with Medicare for 1989, and participating physicians currently account for about 60 percent of all Medicare expenditures for physician services (Physician Payment Review Commission, 1990). Thus, enrollees in a medigap PPO are likely to be able to find physicians outside the PPO network who accept assignment.

The ability of a medigap PPO to channel enrollees to network physicians by assuring assignment will be further weakened by the limit on balance billing included in the Medicare physician payment reform legislation (Omnibus Budget Reconciliation Act of 1989, Public Law 99-509). When this provision is fully implemented in 1993, physicians will not be allowed to charge patients more than 15 percent above the Medicare fee. The Medicare fee for nonparticipating physicians will be 5 percent below the full fee schedule amount for participating physicians, however, so this requirement means that nonparticipating physicians will not be allowed to charge patients more than 9.25 percent above the fee for participating physicians (Physician Payment Review Commission, 1990). This legislation will have the positive effect of limiting beneficiaries' exposure to balance bills on unassigned claims, but it will also significantly reduce the penalty for using a physician outside the PPO network who does not accept assignment.

Medigap PPOs could influence enrollees' choice of physician more if the NAIC model regulations were modified to allow them to cover less than the full 20-percent Part B coinsurance when enrollees use non-network physicians. Medigap PPOs could then design benefit plans in which enrollees are fully covered for services provided by network physicians but are required to pay a portion (such as 10 percent) of the Medicare-approved charge as well as balance-billed amounts on claims outside the network.

Channeling patients to network hospitals is a particularly useful approach to controlling costs in a PPO's private lines of business, because PPOs generally negotiate reimbursement arrangements with hospitals based on discounted charges or per diems. But Medicare pays hospitals under the prospective payment system, in which hospitals receive a predetermined payment per discharge that depends on the diagnosis-related group (DRG) to which the patient is assigned. Thus, unlike PPOs serving private sector clients, a medigap PPO does not have an incentive to perform aggressive concurrent review of hospital stays to ensure that patients are treated cost-effectively and discharged promptly. The DRG payment system already provides such incentives.

A potential advantage to a medigap PPO of channeling Medicare enrollees to network hospitals is that the medigap insurer may be able to negotiate arrangements with these hospitals whereby the Part A deductible is completely or partially waived for PPO enrollees. This would reduce the medigap PPO's benefit payments, enabling it to reduce its premium and/or offer additional services to enrollees. BCBS/AZ has negotiated such arrangements with its network hospitals. But the insurance industry representatives we interviewed for this article expressed uncertainty about whether current Medicare regulations permit a hospital to waive the Part A deductible under such an arrangement with a medigap PPO.

Prior research on beneficiary choice

Medicare beneficiaries currently have an incentive to select physicians who accept assignment, so their behavior in choosing physicians under Medicare may yield insight into the medigap PPOs' ability to channel enrollees to network physicians through financial incentives. (Even beneficiaries with medigap insurance have an incentive to select a physician who accepts assignment, because most medigap plans do not cover balance bills.) In fact, from the beneficiaries' perspective, the PAR program has some of the basic features of a PPO: Beneficiaries have an incentive to select a physician from an annual directory that lists all participating physicians, because by doing so, they are sure not to be balance billed. Directories that identify participating physicians are available free of charge from the carriers and are mailed to Social Security offices, hospitals, senior citizens' organizations, and participating physicians' offices. Beneficiaries are informed of the availability of these directories in an annual enclosure with their Social Security checks. And beneficiaries who have unassigned claims receive information about the PAR program and a toll-free telephone number to call for information in the “Explanation of Medicare Benefits” form, which is mailed to all beneficiaries informing them of the disposition of each claim.

Despite these efforts to publicize the PAR program, many beneficiaries are unaware of the program and most do not understand it. The PPRC beneficiary survey found that only 52 percent of respondents had heard of the PAR program, and only 25 percent understood that participating physicians have agreed to accept the Medicare-approved charge as payment in full on all claims (Nelson et al., 1989). Levels of awareness and knowledge were lowest among low-income beneficiaries, the poorly educated, and those without medigap insurance—the groups most financially vulnerable to high medical bills and therefore likely to benefit most from using a participating physician. The survey also found that only 8 percent of respondents had seen a PAR program directory, and only 3 percent had used one to find a physician.

Medigap PPOs would presumably take more aggressive action to inform enrollees about the providers in the PPO network than the Government has taken to inform Medicare beneficiaries of the PAR program. Medigap PPOs would presumably mail a list of network providers to all enollees, for example, as did BCBS/AZ. So the way beneficiaries under the PAR program choose a physician may not be a reliable guide to how a better informed group of PPO enrollees would do so. But the difficulty of educating beneficiaries about the PAR program suggest that medigap PPOs will face a challenge in educating beneficiaries about the PPO concept generally and about the financial incentives in a given plan.

The PPRC beneficiary survey found that very few Medicare beneficiaries switch physicians for financial reasons. Overall, 9 percent of respondents indicated that they had changed physicians in the prior year. The most common reasons given for changing physicians (each cited by 2 percent of the sample) were that the physician had retired, died, or moved and dissatisfaction with the quality of care or the physician's personality. Fewer than 1 percent of respondents reported that they had changed physicians because of cost. This low response may partly reflect the beneficiaries' failure to understand the financial incentive to switch to a participating physician. But many survey respondents expressed a reluctance to switch to a participating physician even after PAR was explained to them.

Respondents in the PPRC beneficiary survey were questioned about their assignment experience, and those not usually treated on assignment were asked about their willingness to switch to a participating physician. Many beneficiaries do not understand the concept of assignment, so information about their assignment experience was obtained by asking respondents whether the provider submitted the claim to Medicare and whether the Medicare check had been sent directly to the provider, both of which occur on assigned claims. Survey respondents who reported having a regular physician who does not always accept assignment were asked whether they would be willing to switch to a physician who would always accept the Medicare-approved charge as payment in full and would always file the Medicare claim (that is, a participating physician). Only 9 percent indicated that they would definitely switch, 21 percent would consider switching, and 50 percent would not switch (Table 1). (The other 20 percent were not sure whether they would switch.) Respondents were somewhat more willing to switch from the most recent specialist seen. Of those who had not been treated on assignment on their most recent visit to a specialist, 16 percent would definitely switch to a participating specialist, 18 percent would consider switching, and 46 percent would not switch.

Table 1

Percent of Medicare beneficiaries willing to switch from a nonparticipating to a participating physician for regular and specialist care: United States, 1989

Beneficiary choice	Regular provider of care		Most recent specialist seen
	Percent	Standard error	Percent	Standard error
Would definitely switch	8.8	1.2	16.2	1.7
Would consider switching	21.4	1.7	17.7	1.8
Would not switch	50.2	2.1	46.3	2.3
Don't know	19.6	1.6	19.8	1.8
Sample size	601	—	517	—

NOTE: Only beneficiaries with a regular source of care are included in the table.

SOURCE: (Nelson et al., 1989.)

Table 2 shows demographic and socioeconomic characteristics of beneficiaries receptive to switching from their regular physician and from the most recent specialist they have seen. For the purposes of this table, beneficiaries who indicated they would definitely switch and those who would consider switching were combined in a single category of “potential switchers.” Highlighted here are a few findings from the detailed analysis available in Nelson et al. (1989). The beneficiaries most willing to switch from their regular physician to a participating physician are those who are disabled, male, or black, or who have been with their regular physician for less than 1 year. Somewhat surprisingly, low-income beneficiaries are no more likely than high-income beneficiaries to indicate a willingness to switch from their regular physician. So even among low-income beneficiaries, the relationship with a regular physician is often strong enough that the individual is unwilling to sever it in response to financial incentives.

Table 2

Percent of Medicare beneficiaries who would definitely switch or would consider switching to a participating physician for regular and specialist care, by demographic characteristics: United States, 1989

Characteristic	For regular care		For specialist care
	Percent	Standard error	Percent	Standard error
Total	30.2	1.9	33.9	2.2
Age in years
Under 65 (disabled)	60.5	8.7	49.5	8.5
65-74	30.1	2.7	35.5	3.2
75-84	29.5	3.6	31.0	4.0
85 or over	17.5	3.8	24.6	4.4
Sex
Male	36.0	3.2	37.4	3.4
Female	26.1	2.4	31.2	2.8
Income
Below the poverty line	31.7	5.3	47.9	5.6
100-150 percent of the poverty line	32.5	4.6	35.7	5.0
150-200 percent of the poverty line	33.4	5.0	28.7	6.2
200-300 percent of the poverty line	35.9	4.6	34.0	4.7
Over 300 percent of the poverty line	30.2	4.3	34.4	4.7
Education
8 years or fewer	30.7	4.0	38.6	4.6
9-11 years	34.5	5.3	33.9	5.3
High school graduate	32.4	3.6	30.0	4.0
Some college	27.4	5.3	33.0	5.8
College graduate	22.8	4.8	37.5	5.8
Race
White, non-Hispanic	29.6	2.1	32.8	2.3
Black, non-Hispanic	57.4	6.8	48.7	6.5
Hispanic	*	—	*	—
Other, non-Hispanic	*	—	*	—
Health status
Excellent	33.8	5.3	27.2	5.4
Good	26.6	3.0	37.4	3.8
Fair	33.0	3.5	34.3	3.8
Poor	32.0	5.3	33.1	5.4
Supplemental coverage
Medicare only	35.4	4.7	35.8	4.6
Medicare and Medicaid (with or without supplemental)	NA	NA	NA	NA
Medicare and private supplemental (no Medicaid)	29.2	2.1	33.6	2.5
Number of years with regular provider of care
Less than 1 year	50.9	7.4	—	—
1-2 years	24.4	5.1	—	—
3-5 years	24.7	4.1	—	—
5-10 years	33.7	4.7	—	—
More than 10 years	28.4	3.1	—	—
Sample size	601	—	517	—

Indicates that there are fewer than 25 overvations in the cell.

NOTE: NA means not applicable.

SOURCE: (Nelson et al., 1989.)

Physician recruitment

The success of medigap PPOs will also depend on their ability to attract providers to their network. This is not expected to be a serious problem. First, the rapid growth of PPOs in recent years indicates that many physicians are willing to participate in this type of arrangement. Forty-five percent of U.S. physicians belong to at least one PPO network and more than 25 percent belong to two or more (Managed Health Care, 1990). Second, the supply of physicians in the United States has increased steadily in the past and is expected to continue doing so, which is likely to make the market for physician services more competitive. From 1970 to 1986, the number of physicians per 100,000 residents grew from 156 to 225 and is projected to reach 264 by the year 2000 (U.S. Department of Health and Human Services, 1988). Third, the trend toward increasing rates of assignment and participation under Medicare suggests that many physicians are willing to forego balance billing to maintain or increase their Medicare patient load. PAR has some of the basic elements of a PPO, because PAR program physicians agree in advance to accept assignment on all Medicare claims. The fact that 45 percent of all U.S. physicians signed PAR agreements in 1989 suggests that many physicians may be receptive to participating in a medigap PPO.

In this increasingly competitive market for physician services, many physicians are likely to view participation in a medigap PPO as a way to maintain or increase patient volume. But it is difficult to predict how reform of physician payments will affect physicians' receptivity to medigap PPOs. On the one hand, Medicare fees will increase significantly for physicians who provide mostly primary care, so those physicians may become more willing to accept assignment, which presumably will be one requirement for PPO participation. On the other hand, higher fees for primary care may diminish their concern about reduced patient volume, so they may be less willing to join a PPO. Physician receptivity to medigap PPOs may also depend on the utilization review procedures used, which many physicians do not like, according to anecdotal evidence.

Cost-containment potential

The medigap PPO is a recent innovation, so there is no evidence yet about its ability to control costs. Even for the population under 65 years of age, there is only limited evidence of the cost effectiveness of PPOs. The only reliable evidence on this issue comes from a study of the PPOs offered to employees of five medium-to-large firms (Hosek et al., 1990). This study found that the PPOs were successful in reducing expenditures on outpatient physician services, but the findings on inpatient hospital care were inconclusive. The PPOs included in the study each had a utilization management program that included prior authorization for hospital admissions and concurrent review of hospital stays, but limited retrospective review or review of ambulatory care. None of the PPOs had a system for profiling physician practice patterns.

Previous studies have found that hospital use and total medical spending are reduced by utilization management in conventional fee-for-service insurance plans (Feldstein, Wickizer, and Wheeler, 1988; Wickizer, Wheeler, and Feldstein, 1989; and Institute of Medicine, 1989) and in the managed care environment of HMOs (Manning et al., 1987; Luft, 1981). PPOs typically seek to channel patients into managed care, so these findings lend support to the expectation that PPOs will reduce costs. But the effectiveness of a given PPO is likely to depend heavily on the benefit design and the extent to which it induces patients to select providers from within the PPO network.

PPOs in the private sector use several approaches to ensure that enrollees, once channeled into the network, are treated in a cost-effective manner. These include:

Price discounts from providers.

Selection of only providers with cost-effective practice styles.

Utilization management and review.

Provider monitoring and feedback.

Case management of high-cost illnesses.

The potential applicability of these cost-containment approaches to medigap PPOs is discussed in the following sections.

Provider discounts

In a survey of PPOs by the American Managed Care and Review Association (AMCRA) in 1989, almost all of the responding PPOs reported having some form of discounted pricing arrangement with the hospitals in their network (American Managed Care and Review Association, 1990). Medigap PPOs are less likely than PPOs in the private sector to obtain discounts from providers because the Medicare program has already implemented policies to control prices—most notably, the prospective payment system for hospitals, the physician fee freeze, and the incentives for physicians to accept assignment (Bachman et al., 1989). The most likely route for medigap PPOs to obtain discounts from providers would be through negotiated arrangements in which network providers waive the deductible or a portion of the coinsurance. Some hospitals in BCBS/AZ's network, for example, have agreed to waive the Part A deductible for enrollees in its medigap PPO. This approach is administratively simple as it does not require the fiscal intermediary's involvement and yields cost savings for the medigap insurer.

Selecting cost-effective providers

A 1986 survey of PPOs sponsored by the Health Insurance Association of America found that PPOs were not using cost effectiveness as a criterion for selecting physicians for their network (De Lissovoy et al., 1987). PPOs were drawing their physicians from existing physician panels, such as those of Blue Shield plans, and from the staffs of network hospitals—without screening for cost-effective practice styles. More recent surveys have not addressed this issue, so it is not known whether PPOs have changed their provider selection practices since 1986. BCBS/AZ regards the selection of cost-effective providers as a critical component of its cost-containment strategy.

Screening physicians on the basis of cost-effective practice styles is likely to be most effective at controlling costs when combined with a strong utilization management program. Merely collecting cost-effective providers in a network, and doing nothing to influence their behavior, has a limited potential for cost containment, because no savings would be achieved for these physicians' current patients. Savings in this case could be achieved only by inducing enrollees to switch from high-cost (non-network) providers to network providers.

Utilization management and review

Virtually all PPOs attempt to control enrollees' use of services through some form of utilization management and review. The utilization management programs of most PPOs concentrate on reducing unnecessary and inappropriate use of inpatient hospital services (American Managed Care and Review Association, 1990). The most common technique is preadmission review, which requires that physicians obtain approval from the insurer (or from the utilization management company acting on the insurer's behalf) before an elective hospital admission. Failure to comply typically results in lower reimbursement from the insurer. Most PPOs also employ concurrent review of hospital stays, monitoring the course of treatment during the hospitalization and assessing the appropriate length of stay. In addition, many PPOs require second opinions for high-cost surgical procedures.

Many PPOs conduct retrospective reviews of inpatient hospital stays. These PPOs review claims for hospital care after discharge to assess the appropriateness of care, and a negative assessment can mean denial of payment. The utilization review programs of most PPOs, however, emphasize prospective reviews, which give the PPO a chance to influence treatment.

PPOs have historically put less emphasis on controlling the use of ambulatory services than that of inpatient hospital services. One-half the PPOs responding to the AMCRA survey reported using prior review for selected ambulatory procedures, but no information is available on the types of procedures reviewed. In recent years, BCBS/ AZ has placed increasing emphasis on controlling ambulatory service use in its private sector PPO and now emphasizes this as strongly as the control of inpatient use.

Medigap PPOs could potentially use the same types of utilization management and review procedures as private sector PPOs do. There is an important difference between the two, however. In private sector PPOs, utilization management and review activities are fully integrated with the claims-approval process. PPOs in the private sector can deny or reduce payments on claims that fail their utilization management criteria. Medigap PPOs have no such authority, however, because the responsibility for approving claims rests with the Medicare program, the primary payer. And the NAIC model regulations would prohibit medigap PPOs from withholding payments for deductible and coinsurance amounts on claims approved by Medicare. Thus, if medigap PPOs are to perform utilization management and review, these activities must be coordinated with the processing and approval of claims by the Medicare carriers and intermediaries.

Provider monitoring and feedback

Physician profiling is BCBS/AZ's primary approach to cost containment. De Lissovoy et al. (1987) report that in 1986 nearly one-half of all PPOs were developing physician profiling systems to include in their utilization control programs, and many others planned to begin developing them. Monitoring and feedback are the main elements of an ideal physician-profiling system.

BCBS/AZ's physician-profiling system involves analyzing claims data to monitor the practice patterns of individual physicians. The objective is to identify outlier physicians and to notify them that their practice patterns vary substantially from other physicians in the same specialty. This notification can serve an educational function and encourage physicians to modify their practice patterns. Physicians whose practice patterns continue to deviate significantly from the norm are expelled from the network.

Case management of high-cost illnesses

The objective of case management is to promote more appropriate and cost-effective care for individuals with serious, high-cost illnesses. Typically, the patient is assigned a case manager who assesses the individual's needs and circumstances and helps plan, coordinate, and arrange for the most appropriate care. Such case management is usually voluntary, with no financial penalties for failure to comply.

Case management programs have developed rapidly, reflecting the growing recognition that a substantial proportion of an insurer's benefit costs in a given year are often attributable to a small proportion of its enrollees. Case management services for high-cost cases are currently offered by most commercial insurers, Blue Cross and Blue Shield plans, utilization management firms, and third-party administrators (Institute of Medicine, 1989). Such management programs could offer significant savings for the Medicare program. A medigap PPO is unlikely to provide case management services without receiving additional compensation from Medicare because, as a secondary insurer, the medigap PPO's financial exposure in catastrophic cases is much less than that of Medicare. But medigap PPOs could provide a useful structure through which Medicare could contract for case management services for catastrophically ill beneficiaries.

Other possible effects

The introduction of medigap PPOs could affect Medicare costs in other ways. By offering a lost-cost alternative to traditional medigap insurance, for example, medigap PPOs may increase the proportion of the Medicare population with supplemental coverage. By reducing the net price of care for newly covered beneficiaries, this expansion in medigap coverage is likely to increase their demand for care. To offset such increases in the demand for care, medigap PPOs must maintain an effective utilization management program.

The growth and development of medigap PPOs could induce competitive responses from other market participants, particularly from non-network providers, other medigap insurers, and Medicare HMOs. These competitive responses could significantly affect Medicare costs in the market areas that the medigap PPOs serve. But predicting those responses is difficult because of the Medicare physician payment reforms in 1992 and other major changes in the health care market.

Non-network physicians might respond to the threat of losing patients to the medigap PPOs in several ways. First, they might increase their willingness to accept assignment, which would reduce out-of-pocket costs for their patients. Second, they might try to increase demand among their current patients—for example, by ordering additional tests or followup office visits. Evidence about whether physicians can induce demand for their services is inconclusive, but many health economists believe they can, at least to some extent. Third, non-network physicians could choose to compete with medigap PPOs by combining with area hospitals and another insurer to create their own medigap PPO.

Independent practice association (IPA)-model HMOs are likely to be well positioned to create their own medigap PPO, because they contract with large networks of physicians who treat both fee-for-service patients and HMO patients. HMO management could view creating a medigap PPO as a way to increase its Medicare market share. But allowing a Medicare HMO and medigap PPO to operate in the same area under the same ownership would raise problems, as the HMO would have an incentive to disenroll its sickest patients and enroll them in the PPO, thus shifting high-cost patients from capitation to fee-for-service.

Industry interest

To assess the current level of interest in medigap PPOs in the insurance industry, and to get industry views on the feasibility and effectiveness of medigap PPOs, we conducted telephone interviews with executives of five commercial insurance companies that are major medigap insurers, executives of the national Blue Cross and Blue Shield Association of American, and executives of four Blue Cross/Blue Shield plans that, besides BCBS/AZ's Senior Preferred, have medigap products with some of the features of a PPO: Blue Cross and Blue Shield of Alabama, Hawaii Medical Service Association, Blue Shield of California, and Blue Cross and Blue Shield of Minnesota.

The Blue Cross and Blue Shield representatives were more positive about the current viability of medigap PPOs than were the representatives of commercial insurance companies, but both identified impediments to the expansion of medigap PPOs. The major commercial medigap insurers are not interested in developing medigap PPOs unless some of their main concerns are addressed. Concerns mentioned by both commercial insurers and Blue Cross and Blue Shield representatives include:

The financial viability of medigap PPOs is uncertain because most of the savings generated by the PPOs' cost-containment procedures will accrue to the Medicare program rather than the medigap insurer. The medigap insurer's costs for Medicare-covered services consist of deductibles and coinsurance, so the reduction in medigap payments may not be enough to offset the costs of developing, marketing, and administering the PPO.

Medigap PPOs are less likely than private sector PPOs to obtain price discounts from providers because the Medicare program has already implemented policies to control prices, especially the prospective payment system for hospitals, the physician fee freeze, and the incentives for physicians to accept assignment.

A medigap PPO is currently limited in how much it can penalize enrollees for obtaining care outside the network, as the NAIC model regulations require medigap insurers to cover the full 20 percent Part B coinsurance.

Medigap PPOs cannot conduct utilization management and review as do private sector PPOs unless these activities are integrated with the claims approval decision, currently the responsibility of Medicare carriers and intermediaries. Such integration is necessary for a PPO to enforce its utilization management and review decisions.

The differences between Part A and Part B claims and the complexity of merging these data to monitor resource use during an entire episode of care are viewed as impediments to effective utilization management.

It may be difficult to educate Medicare beneficiaries about medigap PPOs. Many Medicare beneficiaries are not well informed about their Medicare and medigap benefits, and the PPO concept is difficult to understand.

Having medigap PPOs perform utilization review could expose insurers to a lot of liability litigation.

Steps the government could take to alleviate these concerns are reflected in the recommendations that follow.

Conclusions

The BCBS/AZ model offers several advantages as an approach to introducing a PPO option under Medicare. First, it relies on private sector innovation to develop and implement the PPO, with minimal government involvement. Second, it incorporates the PPO into an existing product (medigap insurance) that most Medicare beneficiaries already purchase. Third, the model does not impose extra administrative burdens on the carriers or intermediaries, because the incentives used to channel enrollees to network providers do not involve any changes in the basic Medicare benefit structure. Whether this model will prove to be a viable and effective approach to cost containment is an issue the authors will analyze in the future.

Our preliminary assessment is that important impediments currently limit the development and effectiveness of medigap PPOs. First, medigap PPOs have limited ability to channel enrollees to network physicians, which is critical to cost containment. Currently, the best way for PPOs to encourage enrollees to remain within the network is to guarantee that network physicians will accept assignment and to provide no coverage for balance bills on claims outside the network. This is a relatively weak incentive, however, given the high assignment rate on Part B claims. Medigap PPOs would be more effective at channeling enrollees to network providers if the NAIC model regulations were modified to permit them to penalize enrollees for obtaining care outside the network, for example, by covering 10 percent rather than 20 percent of the Part B coinsurance.

Medigap PPOs are also limited in their available approaches to controlling costs within the network. BCBS/AZ seeks to control costs in its medigap PPO by restricting its medigap PPO network to physicians who are in its private sector PPO network, where it conducts physician profiling and utilization management and review. These activities are not conducted for the medigap PPO, but the firm expects that by drawing its medigap PPO network from its private sector PPO network, the former will consist of physicians with cost-effective practice styles. This approach does not require any interaction with the Medicare carrier. Medigap PPOs cannot conduct the utilization management and review activities conducted by most private sector PPOs unless these activities are coordinated with the claims approval decisions of the carrier and intermediary. Government action would be required to establish and facilitate that coordination. Even if the necessary coordination were established, it is questionable whether medigap insurers would invest the resources to conduct utilization management and review as aggressively as private sector PPOs, because most of the savings would accrue to the Medicare program. Thus, if Medicare wants to encourage medigap PPOs to conduct utilization management and review, it will probably be necessary for Medicare to cover part of the costs of those activities.

The government should also clarify whether medigap PPOs are allowed to negotiate with hospitals to obtain waivers or reductions of deductibles and coinsurance. Industry representatives we interviewed are unsure whether this is permitted under current Medicare regulations. Such arrangements with hospitals would allow medigap PPOs to reduce their claims costs, thus improving their ability to reduce their premiums or offer more incentives to attract enrollees.

The introduction of medigap PPOs may also require government action to help consumers make informed choices about medigap PPOs and protect consumers from abusive or misleading marketing practices. If enrollees are not fully informed about the financial penalties for using out-of-network providers and the utilization management procedures employed by medigap PPOs—particularly those involving prior review—many enrollees could incur higher-than-expected out-of-pocket costs, become dissatisfied, and disenroll.

This article was prepared in the early stages of a 42-month evaluation of the Medicare PPO Demonstration by Mathematica Policy Research, Inc. Future research to be conducted under this evaluation will yield much more information about the viability and effectiveness of the BCBS/AZ medigap PPO. Subsequent analyses will examine a broad range of issues for BCBS/AZ and other participants in the Medicare PPO demonstration. These analyses include an examination of the beneficiaries' decisions to enroll in the PPO and potential selection bias of enrollees, and an examination of the impact of the PPO on the use and cost of services provided to Medicare beneficiaries.

Technical note

Many of the previously mentioned issues may be addressed in a new Bush administration initiative enacted in Public Law 101-508 (Omnibus Budget Reconciliation Act of 1990), which allows the marketing of private individual and non-employer group medigap policies with a type of managed care component, called “Medicare Select.” These policies generally would only cover full medigap benefits when the service was provided by the plan's managed care network. Full medigap benefits would be paid for emergency and urgent out-of-area care provided by non-network providers. Although full medigap benefits would generally not be paid for out-of-plan services, Medicare Select enrollees would still receive full Medicare benefits for these services.

Insurers wishing to market a Medicare Select plan will have to meet the normal medigap requirements. In addition, the PPO provider network must afford sufficient access for subscribers, and there must be an internal system of quality assurance. Beneficiaries signing up for Medicare Select must be informed of payment restrictions and the availability of, and premiums for, the traditional policy offered by the insurer. Under a utilization review contract option, Medicare Select plans could contract with the Secretary of Health and Human Services to determine the medical necessity of services provided to plan enrollees. This would replace the necessity determinations made by Medicare intermediaries and carriers. Plans exercising this option would be subject to the same quality review as HMOs with Medicare risk contracts.

The authorizing legislation permits Medicare Select plans to be marketed in 15 States selected by the Secretary for 3 years, beginning in 1992. The NAIC anticipates approving a new model State regulation governing Medicare Select plans.