May
17-19, 2001
Conference Summary
The Eighth Princeton Conference examined the changing face of managed health care in America, looking at how managed care and other emerging insurance/delivery structures will evolve in the marketplace. The conference focused on the present state of managed care in the public and private sectors, the forces that will shape the future of managed care, and the type of models and structures that are likely to emerge. The speakers included leading researchers, CEOs of major insurers, Representative Nancy Johnson, chair of the House Ways and Means Subcommittee on Health, and Centers for Medicaid and Medicare Services (CMS, formerly HCFA) administrator Tom Scully. The conference was held May 17-19, 2001.
|
Note:
A portion of the conference proceedings are available in streaming
video and audio, through the HealthCast network of Kaisernetwork.org.
The conference sessions can be accessed by clicking here. |
Summary written by Michael Doonan, Ph.D., Schneider Institute
for Health Policy
(to facilitate web access, the summary has been broken into 3 parts. Part 1
is below; the other parts are on separate pages. Click on a topic below to go
to that portion of the conference summary)
Part I: Current Assessment of Managed
Care
Uwe Reinhardt, Ph.D.
Managed Care at the Crossroads: Where it has been and how it will evolve in the future.
Dr. Reinhardt described the American health care system as both stable and dynamic. The system experienced a constant 4.5 percent increase in per year in constant GDP purchasing power over the past four decades regardless of major system changes. Reinhardt then detailed the rise and stumble of managed care, and offered some thoughts on changes in privately and publicly provided health care coverage and the potential for the Internet.
Until the 1990s there were three basic principles for the American health care system:
The doctor managed health care and insurance companies paid bills without question. No one looked at outcomes. It was assumed that physicians were "doing the best they can."
In the early 1990s health insurance premiums grew at double digit rates and forecasts were that 20 percent of GDP would be spent on health care by 2001 and 28 percent by 2010. There was a wide range in the cost of care per patient that could not be justified by the medical profession. In 1996, total Medicaid payments per Medicare enrollee ranged from $7,783 in Miami and $7,700 in Baton Rouge to $3,923 in Portland, Oregon and $3,700 in Minneapolis. Other data showed a huge amount of inappropriate care. The system response was to question the premises of the golden era and managed care began using mechanisms such as practice guidelines, pre-authorization, and practice profiles.
The new vision was that the doctor-patient plan of action would be overseen by a managed care organization based on scientifically-based clinical guidelines. Outcome data would be fed back to doctors in a system of continuous improvement. Doctors want to do the right thing and with the proper data will move to improve the quality of care. Unfortunately, we are at least 10 years before data will be available on individual physicians in any meaningful form. The theory of managed care and managed competition between plans "American style" was implemented somewhat less than elegantly. Managed care came to town riding roughshod, micro-managing physicians and kicking just-delivered mothers out of their hospital bed.
Managed care showed success in reducing premiums, which dropped from a 17 percent annual increase in 1990 to actual premium reductions of .4 percent in 1996. Forecasts of health care accounting for near 20 percent GDP for 2000 did not surface; the actual number is 13.1 percent. However, despite these savings out-of-pocket costs to workers did not change and thus they did not realize that managed care helped them. Rather, workers experienced the pain of restricted access. Now, double-digit premium increases are here again. The reason is that during boom times a tight labor market took pressure off the system to control cost. In this environment, we cannot make the hard choices and "do managed care right." Open networks are becoming increasingly common. Health plans are turning into bill payers, all providers and hospitals are available in all plans, and the ability to negotiate prices has been severely constrained.
There are three choices before the American people. We can go back to the pre-1990 order, create a publicly regulated utility model, perhaps based on the German model, or have a multi-tiered system based on rationing by income class. Reinhardt suggested that under a multi-tiered system;
This third option might come about by changes in the employment-based system and Medicare reforms. Two versions of the defined benefits model was offered. Under the "teenage model" employers continue to pool risk over all employees and organize the market based on managed competition. Employers would contribute a fixed contribution and employees wishing for a more expensive plan would pay the difference. This is similar to reference pricing.
Under the adult model, employers cash out and employees are on their own in the individual market. It is unlikely that employees will accept this and there are many unanswered questions. Who would organize market for employees? Who would do the risk pooling? Risk adjustment? What about risk that can be linked to life style? An e-based health market may help consumers assess options, but without government regulation it will not be able to pool risk across chronically healthy and chronically sick individuals.
There are several Medicare reform plans under consideration. Under an early version of Breaux-Frist, a farmer’s market of health plans would be available to beneficiaries. Plans would submit bids and data to enable choice. The government would contribute a portion of the premium equal to 88 percent of the weighted national average premium that has been risk adjusted. With the wide range of premiums across the country, high cost states would be at a major disadvantage. The latest version of Breaux-Frist makes adjustments based on regions to account for this. Reinhardt pointed out a number of economic and political obstacles to this program including a senior citizen managed care backlash led by the combination of physicians and their Medicare patients.
Reinhardt ended with a comment on the potential for the Internet in organizing care. He also cautioned, "As radio and the telephone did before, the Internet offers tremendous potential for enhancing the efficiency of the health care system and for holding it accountable to its customers. After the shameless debacle of the dot.com industry in the past decade, one is entitled to greet new e-Care announcements with a dose of skepticism."
[back to Table of Contents]Harold Luft, Ph.D.
Current Assessment of Managed Care
Click here to see the associated slide presentation [in pdf format]
Luft’s goal was to provide an update to his previous work with Miller through a systematic synthesis of the managed care literature. This research concludes that HMOs offer the same quality of care, more prevention and reduced hospital use, compared to fee for service plans. On the other hand, HMOs provide less access to more expensive services and decreased consumer satisfaction.
This work is not a meta analysis, which is not possible with the multitude of methodological differences. The work moves from a set of articles to findings or bottom lines. It examines HMOs and some mixed models, but excludes studies purely on PPOs. Articles were found through a wide search using only studies that look at medical services broadly defined. They did not look at impact of managed care on physicians. It was important that the research had some reasonable comparison or control group.
The authors combined studies if they used the same data. Within an observation, there could be multiple results. Results were categorized in gradations from generally favorable and statistically significant to generally unfavorable and statistically significant. Studies in the middle were inconclusive.
Results showed that the quality of care provided by managed care was evenly balanced between favorable and unfavorable. Twelve studies showed better quality, 12 showed worse quality, and 13 were in the middle. Access to care was seen as unfavorable in four of six studies. Satisfaction with HMOs was viewed as worse in eight of nine studies. Prevention efforts were viewed as better in HMOs in 6 or 8 studies (most of these studies related to cancer screening). There were not a lot of studies on utilization. The studies examined showed no difference hospital admissions, but shorter length of stays for HMO patients. They also showed HMOs using fewer expensive resources, but similar uses of emergency rooms. In local communities, high HMO penetration is associated with higher uncompensated care and lower levels of charity care. On the other hand, it is associated with higher adherence to breast cancer screening and lower premiums. In summary, HMOs provide the same quality, more prevention, and less hospital use. At the same time they provide less access and lower satisfaction levels.
There is some potential bias in the results that might not be adequately dealt with. Although it was an extensive search, some studies could have been missed or mischaracterized. With this in mind the search method and classification were kept as open and accessible as possible. HMOs generally enroll healthier people and this was accounted for by risk adjustment, but it is possible there was not enough risk adjustment. There may be publication bias towards studies that have exciting findings.
The managed care backlash or anger at the system is in part a result of limiting access to more expensive resources. However, not much evidence was shown that quality of care suffered based on any objective measures. People might be satisfied with their own care, but not like managed care. Employers have pushed to managed care for cost not quality reasons. HMOs achieved slower rates of premium growth but created more problems in access for the uninsured.
More studies are necessary on clinical procedures and outcomes. There were twice as many studies on quality, however, given the volume of heterogeneity twice, the volume of studies are needed to understand more fine distinctions. Additional research is needed to understand local variation and the influence managed care and fee-for-service systems have on each other.
[back to Table of Contents]Paul Ginsburg, Ph.D.
Center for Studying Health System Change
Click here to see the associated slide presentation [in pdf format]
Ginsburg provided information on the impact of managed care based on the Community Tracking Study (CTS), conducted by the Center for Studying Health System Change. The study collects data through surveys and site visits of a representative sample of communities. Data is used to analyze changes in the organization, shifts in financing and delivery of health care and the impact of these changes on citizens. This is the third round of data collection. Forty to 90 interviews were conducted in 12 markets randomly selected from a sample of metropolitan statistical areas. Each site included a cross section of local health system leaders, five health plans per site, hospital systems, physician organizations, consumer groups, and state and local policy makers. Interview and documentation from all parties was analyzed and results were the product of triangulation, or confirming a finding through multiple sources of information.
How is managed care changing based on current pressures? The environment has become hostile with consumers demanding more services and fewer restrictions on choice of provider. The tight labor market made employers receptive to employee concerns. After years of sharp discounts, providers have pushed back, leading to higher payments and less red tape. Consumers and providers are both pushing for protection. Other pressures on plans include high pharmaceutical costs. In addition, we are currently in an upward phase of underwriting cycle.
There has been a shift in managed care strategy from a time when profitability was sacrificed to expend market share. Currently, aggressive premium increases are accompanied by efforts to eliminate marginal business lines like Medicare, Medicaid, small group and individual markets. For example, in Seattle premiums increased from 10 to 20 percent over the last two years plans. Further, the number of available Medicare+Choice plans were reduced from eight to two and three plans dropped Medicaid products in 2001. Employers and employees are facing higher premiums, while public sector managed care options are dwindling. Higher premiums will lead to more uninsured.
At the same time there is an increase in open access products. The distinction between HMOs and PPOs is evaporating with increased access to open network products. In Phoenix, United’s open access products account for the majority of their members. PacifiCare is preparing to launch a PPO. The implication of fewer restrictions is a weakening of plan accountability. There is a movement towards standardized utilization management across products. Disease management is not growing, but more is being done now than in the past. Group and staff models are moving to strengthen utilization management, but many are resisting three tiered drug programs. These organizations are trying to get physicians to control pharmaceutical costs. Northern New Jersey show signs of moving more towards consumer and physician friendly managed care from places such as Aetna, United, and Amerihealth. The implications are less administrative control and more difficulty in controlling cost on the one hand, but improved relationships with providers and consumers on the other. Provider relationships are being restructured with less risk contracting. Plans are buying stability with higher payment rates. For example, in Boston, plans are keeping wide networks available, including higher cost teaching hospitals, and foregoing hard negotiation tactics of the past.
The implications of restructuring provider relationships are a loosening of control over provider networks, an erosion of a platform for quality and cost containment, and higher premiums. Employers are going to respond more forcefully and more cost-sharing will result. This will not result in more restrictive networks. Pharmaceutical cost sharing is an example. Public sector programs struggle with plan and provider participation. It is likely that there will be a greater difference in provider networks between public and privately financed plans.
[back to Table of Contents]Questions and Comments:
Tom Rice, UCLA: Asked a question about quality of care in HMOs by disease and condition. He suggested people with chronic illnesses might be doing worse in HMOs.
Luft: The earlier evidence was that people with chronic care needs are not doing as well. We haven’t had as many studies focusing on this subgroup and the data is thin. Maybe there are differences that vary by plan and delivery systems. We need to move to more direct measurement of the quality of care provided by particular plans.
Bradford Gray, New York Academy of Medicine: It used to be that a larger market share and bigger network were seen as enhancing negotiating power. Have plans recently concluded that they do not need to worry about pricing going up?
Ginsburg: The critical thing for negotiating is the ability to move away from providers that won’t agree with you. MCOs have lost this ability. As far as premiums go, if a plan is not making a profit per person, adding more people will not help.
Drew Altman, The Henry J. Kaiser Family Foundation: Except for Medicare, how relevant are comparisons with fee-for-service? Is it not more relevant to examine different kinds of managed care?
Luft: "This is not your fathers fee-for-service." Most all health plans managed with utilization review or other tools. We do need to learn more about what successful plans are doing and how they are doing it differently. More attention needs to be dedicated to the process of care.
Pat Powers, Consultant: Physicians say that they treat all people the same. Are patients not going for preventive services under fee-for-service?
Luft: There are differences in coverage. HMOs cover preventive things with less out-of-pocket costs. Some HMOs are proactive in getting in touch with patients to tell them to come in for particular screening.
Ginsburg: Plans want HEDIS accreditation, and some actually give extra payments to physicians for meeting screening and prevention targets.
Stan Wallack, Schneider Institute for Health Policy: Asked about three-tier pharmacy benefits and noted a resistance from physicians to follow multiple rules from multiple health plans.
Ginsburg: Three-tier systems are geared to drive people to lower cost drugs. Kaiser already does this through physicians.
Luft: Under Kaiser’s classic group model, physicians steer people towards particular drugs to control cost and capture rebates.
Larry Lawrence, Maryland Hospital Association: Public managed care programs are in crisis. For-profit health plans are becoming increasingly unavailable to Medicare and Medicare beneficiaries.
Ginsburg: For a while Medicaid payments were high enough to attract business from for-profits plans, but as rates went down they left the program leaving not-for-profits and Medicaid only plans. Medicaid managed is providing less choice of health plans and has a more restricted number of providers.
A question was asked about disease management approaches and why they are not spreading more rapidly.
Ginsburg: Capitation is more favorable to disease management than fee-for-service. Under FFS, physicians will not get paid for education efforts. HMOs have more incentive because they can capture the savings of reduced hospitalization.
Paul Jellineck, Robert Wood Johnson: Asked about micro and macro economic changes and the impact of double-digit premium increases.
Ginsburg: Attention to cost containment waxes and wanes. When we entered boom time, cost containment efforts took a back seat.
Luft: Some changes will be longstanding. For example, there we will continue to have fewer hospital beds. Better information systems are in some places and there will continue to be attention on utilization review. Physician behavior has changed and become more evidence based.
Ann Gauthier, Academy for Health Services Res. and Health Policy: Asked about the quality findings and reasons for the abnormal distribution. Did finding differ based on particular measures of quality?
Luft: We need to look it over again, but it did not seem as though quality outcome was associated with the particular measure of quality.
Margaret O’Kane, NCQA: Suggested that HEDIS helped plans make remarkable achievements in disease management.
Luft: We need to make providers more accountable for the quality of care. Pushing all risk to providers is not going to do it. They would be under different rules with different payers. In California, quality assessment scores are available by medical group. We need to move to different payment schemes that encourage disease management but are not full capitation.
Ginsburg: The health care system is complex and varied. Some things will work in some parts of the country but not others.
Michael Zinner, Brigham and Women’s Hospital: Noted changing provider health plan relationships and the failure of risk contracts. These arrangements ask physicians to assume risk without the resources to effectively manage it. Most physicians want to do the right thing. Often they do not have the right data. Risk is not adequately accounted for and particular physicians are at a disadvantage.
Click here for summary Part II: Four Forces That Will Shape the Future of Managed Care
[back to Table of Contents]