This Article Abstract Reprint (PDF) Submit a response to this article Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services E-mail this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to Citation Manager Reprints & Permissions Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (14) Citing Articles via Google Scholar Google Scholar Articles by Hurley, R. Articles by Casalino, L. Search for Related Content PubMed PubMed Citation Articles by Hurley, R. Articles by Casalino, L. Related Collections Business Of Health Managed Care

Business Of Health

A Longitudinal Perspective On Health Plan–Provider Risk Contracting

Abstract

 
During the past decade many health plans adopted risk-contracting arrangements that transferred substantial financial risk and care management responsibility to physician groups and hospital-sponsored integrated delivery systems. Risk transfer arrangements are now believed to be in steep decline, but there is little empirical evidence on this topic, particularly at the local-market level. Data from the Community Tracking Study were used to examine changes in risk contracting from 1996 to 2000. A decline in reliance on risk contracting is evident in nearly all markets. However, retrenchment in risk contracting has followed different patterns ranging from refinements in the scope of risk transfer to reduced use of risk arrangements to total rejection of risk-sharing arrangements. Modified risk-transfer agreements remain viable in several markets, but continued refinement in the nature and scope of risk sharing will be necessary.

By the end of the decade some observers were contending that a "flight from risk" had occurred.⁴ Loss of enthusiasm was attributed to poor utilization management, inadequate payments by health plans, and diminished savings opportunities as further reductions in service use and cost became more difficult.⁵ In addition, a consumer and physician backlash engendered skepticism about health plans’ methods and motives. Plans became more cautious about the use of incentives and increasingly offered broad-network, open-access (no gatekeeper) products that make it difficult to manage care and transfer risk.

Despite a belief that risk transfer is in steep decline, there is little empirical evidence on this topic, particularly at the local-market level.⁶ We find important differences in the trajectory of risk across markets, given the differing configurations of physicians and health systems, diverse health plans, and varying past experience with risk arrangements. Changes in risk-based arrangements range from refining the scope of services under risk in some markets to reductions in plan membership under risk transfer in others to full rejection of risk in still others.

Background

Top Background Study Methods Study Results Discussion NOTES

 
Although the desirability of risk arrangements has been long debated, there is a general consensus that risk promotes cost-conscious provision of health services, leading health plans to use risk extensively in HMO products.⁷ Health plans may transfer financial risk to provider organizations for primary care services only ("primary care capitation") or include specialist physician services as well ("professional risk"). Some risk arrangements also cover hospital and ancillary services and pharmaceuticals (referred to as "shared risk" or "global risk"). We distinguish between global and shared risk based on whether the contracting organization accepts full risk for inpatient care or the risk is shared between the health plan and the physician groups or physician-hospital organization (PHO).

From a provider perspective, a distinction in risk transfer is sometimes made between "business risk" and "insurance risk."⁸ Business risk is risk for services over which the provider organization can reasonably be expected to exert at least a moderate degree of control, such as physician, ancillary, or hospital services. Insurance risk includes services over which providers have little or no control, as in the case of pharmaceutical costs, new benefit mandates imposed during a contract period, or out-of-network service use by plan members.

Some health plans have promoted or permitted risk contracting because they believe that the "delegated model" is important as a means to change the behavior of physicians and hospitals. It may also enable plans to limit their medical expense exposure by shifting financial risk for care, to reduce their administrative costs of managing care, and to lessen friction with providers who have chafed under micromanaged care. Physicians and hospitals have viewed assumption of risk as a way to reduce external interference in clinical decision making and to reap the gains of cost reductions that changes in their practices may yield. In addition, since capitation payments are not tied to generating procedures and office visits, physician and hospital organizations can use the budget capitation provides to promote preferred practice behavior and to achieve organizational goals.

Efforts by physicians and hospitals to create new organizations came into vogue in the early 1990s and included integrated delivery systems (IDSs), independent practice associations (IPAs), PHOs, and physician practice management (PPM) firms, as well as the extensive purchase of physician practices by hospitals.⁹ Many saw risk contracts as an important complement to other contributing factors, a kind of physiological change to shape and sustain the anatomical changes represented by these new organizations. Despite these broad national developments, the growth of risk arrangements remained highly dependent on local market circumstances, capacity, and interest.¹⁰

Study Methods

Top Background Study Methods Study Results Discussion NOTES

 
The Community Tracking Study (CTS) is a multiyear, multimethod project of the Center for Studying Health System Change (HSC) to analyze patterns of change in a representative sample of sixty markets.¹¹ Twelve metropolitan markets with populations over 200,000 were randomly selected for in-depth study, with intensive site visits conducted in 1996, 1998, and 2000. The site visits include protocol-driven telephone and in-person interviews with fifty to ninety informants in each site representing purchasing, health plan, policy, physician, and hospital sectors. A total of 2,220 interviews have been conducted across the twelve sites during the three rounds. In addition, every two years HSC surveys households (with a follow-back survey to explore type of insurance questions) and physicians in the sixty markets. We use results from the 1996–97 CTS Followback Survey to benchmark risk contracting at that point in time.

Data for this study are drawn primarily from 210 interviews conducted between June 2000 and March 2001 with executives from three to six health plans in each market—a total of forty-one plans. Selected plans include market leaders among regional and national plans, and local Blue Cross/Blue Shield plans. In addition, 256 administrative and clinical leaders in the major health systems and physician organizations were interviewed in the twelve markets. Respondents were asked about contracting arrangements, experience with various configurations of risk transfer, and changes made in network arrangements. All interview notes were incorporated into qualitative data analysis software for coding and analysis.

Study Results

Top Background Study Methods Study Results Discussion NOTES

 
Data from the 1996–97 survey and interview results from Round 1 site visits were used to sort the twelve markets into three broad clusters based on the amount (number of covered lives) and scope (services incorporated under risk payments) of risk contracting (Exhibit 1). In this study we focus on professional and global or shared risk contracts. Orange County in Southern California, where shared risk arrangements have been the dominant form for HMO networks for a number of years, is the only market placed in the extensive-risk cluster. At the other extreme, with limited risk, are Little Rock, Lansing, Greenville, and Northern New Jersey, where risk transfer was typically used by a single plan with a single medical group or IPA.

Our findings examine plan-to-provider risk transfer, and thus our market characterizations are based on how sampled plans employed risk arrangements across the twelve markets. It also is apparent that provider organizations in these markets fall into three broad categories. Some groups were accepting extensive (global or shared) risk across most contracts in 1996; other groups took less risk, by virtue of fewer contracts or more limited scope in their contracts; and still other groups had limited risk experience, meaning few contracts and typically only professional risk. While our focus in this paper is on plan-to-provider contracting, we also note changes that occurred among provider groups.

1996 interviews. Round 1 interviewees in all markets expected both HMO enrollment and risk contracting to grow rapidly in commercial and Medicare lines of business. This was particularly true in metropolitan markets such as Indianapolis and Northern New Jersey, where HMO enrollment had lagged behind national trends or where the Medicare HMO market was seen as a growth opportunity for plans, as in Boston and Cleveland. Expectations of growing reliance on HMO revenues led physicians and health systems to position themselves to "move up the food chain," which meant both seeking risk transfer contracts and expanding care delivery capacity through acquisitions or strategic affiliations and alliances.

Medical groups and IPAs in Orange County were already deeply involved in shared risk arrangements with plans that included pharmacy and other ancillary services. In other markets with less experience with risk arrangements, hospital systems and physician organizations were being created as necessary transitional steps to enter into global and shared risk contracts. A notable example was the formation in Boston of two large integrated systems, Partners and CareGroup. PHOs and IPAs emerged in other markets in this period, with hospitals playing an integral role in the former and physicians usually providing the leadership in the latter. PHOs enabled hospitals and their affiliated physicians to enter into global or shared risk arrangements as IDSs. Freestanding IPAs and medical groups typically limited contract scope to professional or shared risk. Phoenix, Seattle, Cleveland, Indianapolis, and Syracuse saw substantial development of such models. Small, embryonic IPAs also arose in some of the markets in the limited-risk cluster.

The viability of risk-transfer arrangements in the moderate-and limited-risk market clusters hinged on plans’ willingness to enter into contracts with newly organized and unproven entities. Some health plans encouraged their growth, while others were skeptical about these entities’ capacity to perform risk-bearing roles. There was especially strong support for risk-bearing entities in the Medicare HMO product, since risk arrangements—typically global or "percentage of premium" contracts—were seen as key in encouraging physicians to reduce the large inpatient expenses of Medicare and to share in the savings that resulted. Plans with well-established networks with individual physician and hospital contracts already in place, such as Blue Cross Blue Shield plans, were less enthusiastic about risk arrangements.

1998 interviews. The Round 2 interviews found that efforts to promote risk transfer set in motion in Round 1 had paid off in several markets. But many provider organizations found it more difficult than expected to profit from risk contracting, and some suffered heavy losses. Insurance premiums had risen little between Rounds 1 and 2. Correspondingly, capitation payments were flat and lagging behind increases in costs, especially as prescription drug costs began to rise, adversely affecting those organizations that were at risk for these services.

Reductions in service use had been expected to provide savings that would accrue to risk-contracting physicians and hospitals. But in some markets reductions in use had already been made, making further savings difficult. In other markets where such reductions had not yet been made, plans and providers found it difficult to institute effective utilization management. Meanwhile, a consumer backlash challenged managed care techniques, including capitation and utilization management. Health plans began to modify products and practices, shifting emphasis away from HMOs and traditional risk arrangements toward point-of-service (POS) and preferred provider organization (PPO) products.¹²

Orange County witnessed the collapse of major PPM firms that had tried to aggregate market clout by purchasing medical groups and managing IPAs.¹³ Although the failures had many causes, interviewees in several markets saw them as a direct challenge to the soundness of the "California model" of risk transfer. These failures and similar but smaller-scale failures of risk-bearing physician organizations in northern New Jersey and Phoenix provoked regulatory attention in all three states, raising doubts about and, in some cases, barriers to risk contracting.

In markets with moderate or limited risk-transfer experience, caution became the watchword. Plans with interest in risk contracts found fewer organizations willing and able to assume and manage risk, especially risk for services over which providers could not exert much control. In midsize and small markets such as Indianapolis, Lansing, Greenville, Little Rock, and Syracuse, the role of dominant local hospital systems was typically a determinant. Some systems had originally embraced risk as a way to lock in specialty and hospital referral relationships through PHOs that accepted global capitation, such as in Indianapolis. But many health systems lost money on their risk contracts, and, over time, most failed to receive enough patients through risk contracts to effect changes in behavior or to justify major investment in management systems and infrastructure. Enthusiasm faded for initiatives that did not grow revenue and consumed resources. Many systems also found that integration efforts between physicians and hospitals were not progressing, and distribution of risk payments among primary care physicians, specialists, and hospitals became a troublesome management issue.

2000 interviews. Round 3 interviews revealed several notable developments since 1998 with important implications for risk transfer. Plans had begun to obtain sizable premium increases from purchasers, and providers expected to receive corresponding increases in capitation payments. Financial distress, attributed in part to the Balanced Budget Act (BBA) of 1997, also led many physicians and hospitals to demand higher payments from health plans and threaten to refuse to continue to participate in networks unless their demands were met.¹⁴ Already facing serious discontent among consumers and employers, health plans acceded to these demands in order to keep networks intact in many instances.

Hospitals also began to refuse to accept risk-based payments, putting global capitation arrangements in jeopardy and making shared risk arrangements less practical. As one hospital executive stated, "We have a basic belief that we don’t belong in and don’t want to be in the insurance business." Physicians also challenged the scope of their risk contracts because of their discomfort with the amount of insurance risk they had assumed. This pushing back led many plans to agree to exclude from risk contracts pharmaceuticals and benefit and service mandates that providers contended they could not be expected to influence, particularly when mandates were implemented after contracts were already in place.

Change in risk-contracting practices between 1998 and 2000 in the forty-one health plans interviewed in Round 3 falls into several patterns (Exhibit 2). While overall only three fewer plans engaged in risk transfer in 2000 (23) than in 1998 (26), the scope of risk transferred was reduced, and plans became more discriminating in selecting risk-contracting partners. Despite turmoil in Orange County related to PPM failures and the refusal of some hospitals to continue risk arrangements, all major health plans there remain committed to the "delegated model." But the price of sustaining this commitment has been large payment increases to shore up contracting medical groups, IPAs, and hospitals. In addition, in five of the six plans there has been refinement in the scope of contracting, including taking back some pharmacy risk and reducing the impact on risk-bearing entities of state regulations such as mental health parity and mandated immunizations, elements of "insurance risk" that providers consider unacceptable.

The markets in the moderate-risk cluster reveal a mixed picture. Between 1998 and 2000 only two of the twenty-three plans interviewed had dropped all risk-transfer contracts, but other forms of retrenchment were evident. Five plans had reduced the scope of risk delegated, and eight plans reduced the number of members covered by risk contracts by contracting with fewer risk-bearing provider organizations. Seattle typifies this pattern: Plans found fewer organizations willing to accept risk contracts, resulting in fewer plan members under risk arrangements.

Risk arrangements remained in place for some members because plans were willing to undertake major changes to salvage them, including increased payments or carving out of pharmacy and other risk from contracts. In other cases, as in Cleveland and Miami, risk-transfer agreements remain principally in use in the Medicare market, although they could be in future jeopardy because of Medicare+Choice policy changes. In Boston, where risk transfer has been in moderate use, plans have had to relent to pressures for large payment increases for health systems that are still reassessing their posture toward risk. One executive characterized the change as "a more thoughtful approach to risk, as opposed to the huge stampede toward risk five years ago." In Indianapolis and Syracuse a single health plan continues to rely on risk contracts. However, in all of the markets in this category, some physician organizations or health systems have achieved enough success in risk contracting to retain support for these agreements, even as other plans and providers eschew them.

Discussion

Top Background Study Methods Study Results Discussion NOTES

 
This longitudinal perspective on twelve randomly selected metropolitan areas provides valuable insights into the extent of risk contracting and how it has evolved as experience has accumulated and broader market conditions have changed. A number of patterns are apparent, including both sizable rate increases and retrenchment. Retrenchment is occurring in two forms: (1) reduction in the scope of services for which providers will accept risk to something closer to business risk; and (2) reduction in plan members covered under risk because fewer provider organizations are willing to participate in risk contracts.

In Orange County, with established medical groups and integrated systems that embraced risk transfer, risk arrangements remain in place, but their scope has been refined to exclude some services providers believe they cannot control, and payment rates have been increased. For markets where risk transfer has been employed more tentatively, such as Seattle, Indianapolis, and Cleveland, the picture is one of a broader retreat in terms of both narrowing the scope of risk arrangements and reducing the number of HMO members covered under risk-transfer arrangements. Substantial capitation rate increases have also been reported in these markets. In the markets with limited risk-transfer experience, risk rejection is evident, with few if any risk-transfer arrangements likely to survive.

The success of risk transfer in a local market hinges on several factors: (1) plans’ competence in selecting suitable risk-contracting partners; (2) presence of provider organizations willing and able to manage risk; (3) ability of plans and providers to agree on the appropriate scope of services for which risk can be transferred; and (4) adequacy of capitation payments to providers. Lack of growth in risk-transfer arrangements appears to reflect the fact that few new provider organizations have developed an interest in taking on the risk and responsibilities these arrangements entail, or the capacity to do so. For plans and their risk partners that have stepped back from these arrangements, risk transfer is unlikely to reemerge in the near term as an important strategy.

It is clear that plan payment rates have to meet provider expenses over time to sustain risk transfer arrangements, but that has not been the case in many instances for a number of reasons. Some providers have been overly optimistic about their ability to manage care. Others may have been naïve in rate negotiation and actuarial estimation. Still others accepted risk for costs they could not be expected to control, or they encountered unexpectedly large cost increases. In other situations, plans may have used the threat of exclusion from their networks to gain providers’ acceptance of what proved to be inadequate rates, or refused to include realistic updates to reflect changing conditions.

Our evidence cannot address fundamental questions of whether transfer of risk provides a better mechanism for generating and sustaining cost savings by devolving more control to physicians, or how risk contracting, and the provider reorganization that it requires, affects the quality of care.¹⁵ It is possible that savings and financial success achieved in some cases by risk-bearing providers resulted because of more effective care management, especially of inpatient services. Alternatively, some apparent success could have been an artifact of timing relative to the health insurance underwriting cycle. More research is needed to assess the impact of risk transfer on cost savings and quality.

Where risk contracting is well entrenched, there appears to be a reasonably high comfort level between plans and physician organizations with the division of responsibility and the amount of autonomy for patient care that it affords physicians. Despite the tumultuous experience of the past few years in Orange County, few observers there are predicting a radical change in reliance on risk transfer.¹⁶ Seattle, Boston, Cleveland, and Indianapolis also continue to have risk-contracting relationships between health plans and provider organizations, although their futures are less certain. Given that the search for a modicum of peaceful coexistence between health plans and physicians and hospitals is far from over, models that have durability and mutual commitment should not be discounted.

Editor's Notes

 
Bob Hurley is an associate professor in the Department of Health Administration at the Virginia Commonwealth University. Joy Grossman is associate director of the Center for Studying Health System Change in Washington, D.C. Timothy Lake is a health researcher at Mathematica Policy Research, in Cambridge, Massachusetts. Larry Casalino is an assistant professor in the Department of Health Studies, University of Chicago, after having spent twenty years as a family physician in Half Moon Bay, California.

This paper has been prepared by the Center for Studying Health System Change, which is fully funded by the Robert Wood Johnson Foundation.

NOTES

Top Background Study Methods Study Results Discussion NOTES

 

1. S. Shortell, R. Gillies, and D. Anderson, "The New World of Managed Care: Creating Organized Delivery Systems," Health Affairs (Winter 1994): 46–64.
2. B. Shenkin, "The Independent Practice Association in Theory and Practice," Journal of the American Medical Association 273, no. 24 (1995): 1937–1942[Abstract]; and G.J. Bazzoli, R.H. Miller, and L. Burns, "Capitated Contracting Roles and Relationships in Healthcare," Journal of Healthcare Management 45, no. 3 (2000): 170–187.[Medline]
3. J. Robinson and L. Casalino, "The Growth of Medical Groups Paid through Capitation in California," New England Journal of Medicine (21 December 1995): 1684–1687; E.A. Kerr et al., "Managed Care and Capitation in California: How Do Physicians at Financial Risk Control Their Own Utilization?" Annals of Internal Medicine (1 October 1995): 500–504; M. Gold et al., "A National Survey of the Arrangements Managed Care Plans Make with Physicians," New England Journal of Medicine (21 December 1995): 1678–1683; and T. Lake et al., Health Plans’ Selection and Payment of Health Care Providers, 1999: Final Report (Washington: Medicare Payment Advisory Commission, 2000).
4. California Medical Association, The Coming Medical Group Failure Epidemic (San Francisco: CMA, 1999).
5. T. Bodenheimer, "California’s Beleaguered Physician Groups: Will They Survive?" New England Journal of Medicine (6 April 2000): 1064–1068.
6. J. Robinson and L. Casalino, "Reevaluation of Capitation Contracting in New York and California," 17 May 2001, www.healthaffairs.org (22 April 2002).
7. A. Hillman, "Financial Incentives for Physicians in HMOs—Is There a Conflict of Interest?" New England Journal of Medicine (31 December 1987): 1729–1734; and L.R. Burns and D. Thorpe, "Physician-Hospital Organizations: Strategy, Structure, and Conduct," in Integrating the Practice of Medicine: A Decision Maker’s Guide to Organizing and Managing Physician Services, ed. R. Connors (Chicago: American Hospital Association, 1997), chap. 17.
8. A. Hillman, W. Welsh, and M. Pauly, "Contractual Arrangements between HMOs and Primary Care Physicians: Three-Tiered HMOs and Risk Pools," Medical Care (February 1992): 136–148.
9. M. Morrisey et al., "Managed Care and Physician/Hospital Integration," Health Affairs (Winter 1996): 62–71; and J. Alexander, "Organization Approaches to Integrated Health Care Delivery: A Taxonomic Analysis of Physician-Hospital Arrangements," Medical Care Research and Review (March 1996): 71–93.
10. L. Kohn, "Organizing and Managing Care in a Changing Health Care System," Health Services Research 35, no. 1 (2000): 37–52.[Medline]
11. P. Kemper et al., "The Design of the Community Tracking Study: A Longitudinal Study of Health System Change and Its Effects on People," Inquiry (Summer 1996): 195–206.
12. D. Draper et al., "The Changing Face of Managed Care," Health Affairs (Jan/Feb 2002): 11–23.
13. P. Elkind, "Vulgarians at the Gate: How Ego, Greed, and Envy Turned Medpartners from a Hot Stock to a Wall Street Fiasco," Fortune 12, no. 1 (1999): 132–145.
14. B. Strunk, K. Devers, and R. Hurley, Health Plan–Provider Showdowns on the Rise, Issue Brief 40 (Washington: Center for Studying Health System Change, June 2001).
15. R.A. Dudley et al., "The Impact of Financial Incentives on Quality of Health Care," Milbank Quarterly 76, no.4 (1998): 649–686[Medline]; and D. Berwick, "Payment by Capitation and the Quality of Care," New England Journal of Medicine (17 October 1996): 1227–1231.
16. J.C. Robinson, "Physician Organization in California: Crisis and Opportunity," Health Affairs (July/Aug 2001): 81–96.

                      What's this?

This article has been cited by other articles:

				B. E. Landon, S. L. T. Normand, E. Meara, Qi Zhou, S. R. Simon, R. Frank, and B. J. McNeil The Relationship Between Medical Practice Characteristics and Quality of Care for Cardiovascular Disease Med Care Res Rev, April 1, 2008; 65(2): 167 - 186. [Abstract] [PDF]

				L. P. Casalino The Federal Trade Commission, Clinical Integration, and the Organization of Physician Practice Journal of Health Politics Policy and Law, June 1, 2006; 31(3): 569 - 585. [Abstract] [PDF]

				M. A. Hall The Death of Managed Care: A Regulatory Autopsy Journal of Health Politics Policy and Law, June 1, 2005; 30(3): 427 - 452. [Abstract] [PDF]

				G. J. Bazzoli, L. Dynan, L. R. Burns, and C. Yap Two Decades of Organizational Change in Health Care: What Have we Learned? Med Care Res Rev, September 1, 2004; 61(3): 247 - 331. [Abstract] [PDF]

				J. Quadagno Physician Sovereignty and the Purchasers' Revolt Journal of Health Politics Policy and Law, August 1, 2004; 29(4-5): 815 - 834. [PDF]

				G. J. Bazzoli The Corporatization of American Hospitals Journal of Health Politics Policy and Law, August 1, 2004; 29(4-5): 885 - 906. [PDF]

				S. Page How Physicians' Organizations Compete: Protectionism and Efficiency Journal of Health Politics Policy and Law, February 1, 2004; 29(1): 75 - 106. [Abstract] [PDF]

				J. C. Robinson The Politics of Managed Competition: Public Abuse of the Private Interest Journal of Health Politics Policy and Law, April 1, 2003; 28(2-3): 341 - 354. [Abstract] [PDF]