Why Are Physicians So Upset about Managed Care?
Harold S. Luft
Bob Evans et al. (1989) describe some of the conflicts that occur in the Canadian health care system, such as strikes by nurses and physicians and claims of lengthening queues of patients, as "political theatre" because the real arguments over health care budgets often get played out in other guises. It is appropriate to raise the question of whether the current debates over managed care in the United States reflect a similar phenomenon. That is, what is the substance behind the rhetoric?
Given the rapidity with which the health care system is changing, and the slowness with which data accumulate, it is almost impossible to offer an empirical answer to this question. Nevertheless, the observations of a relatively unbiased observer may be helpful. However, one must recognize that we all perceive the world through particular lenses. My location in California, with its somewhat unique form of managed care, certainly biases the relative balance of information I receive.
There have been numerous studies of the performance of various managed care organizations; most have focused on HMOs, and the data often relate to performance of several or more years ago (see, for example, Miller and Luft 1994 and 1997). While more studies have appeared since those reviews, it is unlikely that they would change the overall assessment that there are numerous findings on both sides of the balance sheet. Yet, the debate over managed care has become more and more charged.
It is useful to first discuss what is meant by "managed care" because this may help explain why perspectives may differ. The second section will examine some of the evidence, such as it is, on what may be happening [End Page 957] with respect to physicians. The third section discusses issues of perception and reality.
What's in a Name?
"Managed care" is a term that has even less specificity than the label "low in fat" in your super creamy dessert. While the origin of the term is apparently lost in history, it is useful to place it in some context. We know that the term health maintenance organization (HMO) was created by Lewis Butler and Paul Ellwood to encompass the prepaid group practice model plans, such as Kaiser and Group Health, as well as the loosely organized individual practice association (IPA) or foundation for medical care models. This umbrella term was needed because prepaid group practice organizations (PGPs) were still unacceptable to organized medicine, and the Nixon administration wanted a vehicle able to accept capitation payments and responsibility for a fixed set of services. Health maintenance organization seemed to be good name for such entities. It conveyed the notion of being able to contain costs by maintaining people's health, even though there was little evidence that this was the cause of lower costs in such plans. More importantly, almost all the evidence on HMO performance related to the PGP model (Luft 1987), but IPAs and, later, network model plans, rarely made that distinction in their claims.
The term managed care began to be used in the late 1980s and appeared to have roots in "marketing" terminology, rather than precise science. In the early 1980s there were HMOs and conventional insurance plans that paid on a fee-for-service basis. Indemnity plans literally indemnified the patient for a set amount after he or she had paid the physician directly. Service benefit plans, such as Blue Shield, had contracts with physicians who agreed to accept the fee schedule as payment in full; the contracts permitted any coinsurance to be collected from the patient. Blue Shield plans were initially established by state or local medical associations, and until recently had strong physician representation on their boards. Not surprisingly, the fees were set on a usual, customary, and reasonable basis so that few fees were markedly reduced. In the late 1970s and early 1980s both types of plans began to implement utilization review, second opinion programs, and preadmission certification in an effort to contain rapidly rising premiums. However, many of these activities could be purchased from third-party administrators, rather than the carriers themselves.
In the early 1980s several legislative measures began to change the [End Page 958] nature of medical care payment. Medicare's prospective payment system for hospitals established the precedent for paying providers a fixed amount, rather than reimbursing them for their fees. Medicare's resource-based relative value system (RBRVS) established fee schedules for physicians, not unlike the old indemnity plans, but added the restraint that participating providers could not bill the patient for additional amounts if their fees were above the schedule. Legislative initiatives in California paved the way for new models of payment arrangements. These initiatives included a law that established a MediCal "czar" who would secretly negiotiate payment rates with hospitals and parallel legislation that allowed private health plans to selectively contract with providers (Bergthold 1984).
The ability to selectively contract with physicians offered an enticing new opportunity for insurers. Previously, they could affect utilization primarily by adjusting the benefit package or copayments and deductibles, but these variables were often set by large employers when they purchased coverage. Utilization review before the service was rendered required expertise not usually available in companies designed for checking eligibility and paying claims. Independent vendors were often hired to perform this function, but some large employers would buy this service directly and seek out small third-party administrators to process the paperwork. In contrast, large insurers had substantial economic advantages in developing contracts with physicians and other health care providers. In exchange for the promise of getting more patients, providers agreed to accept discounts off charges, or fixed fee schedules, and to participate in prior authorization programs. The companies now had a unique (or at least apparently differentiated) product: their specific network of physicians who agreed to lower fees and utilization management administered by the firm.
The rapid early growth of these plans--a familiar phenomenon in most innovations--resulted in an interesting dynamic. The plans, as did the state of California before them, promised increased patient load for those providers signing contracts. Due to the oversupply of providers, the fear of being left out was probably more of a motivation to sign than the promise of more patients. Not surprisingly, most physicians signed up, often without comprehending fully the nature of the contract. Pressured by purchasers to have a broad network of providers in order to minimize patient reaction, plans rejected few tenders. Therefore, a contract delivered few new patients but lower fees for all. Not surprisingly, many physicians felt "taken" by this outcome. [End Page 959]
It would thus appear that the reaction to "managed care" is simply the complaints of physicians reared in a system that paid what they charged who now had to adapt to a system in which they have to negotiate over price. From an economics/policy perspective, such complaints would receive an understanding, but unsympathetic response--most small suppliers of goods and services have little control over the prices they can charge. However, more changed with managed care than just the magnitude and control over payment.
Managed Care from the Physician's View
The rate of growth in medical care costs has slowed markedly in the last few years, much to the relief of payers and policy makers (Levit et al. 1998). However, since medical care expenditures are medical provider incomes, a cynical economist might believe that physician complaints about the horrors of managed care are merely a reflection of falling incomes. Few would be so crude as to complain about wages when they are already in the top 1 percent of the income distribution. Thus, one would expect physicians to focus on how managed care affects quality and patient care, much as the firemen on diesel locomotives argued that loss of their jobs would jeopardize passenger safety.
Table 1 presents the most recently available data on physician practice and incomes, as reported by the American Medical Association. With the exception of a dip in 1994, physician income has continued to rise throughout the 1990s, although at a slower rate in the latter part of the decade, not unlike the incomes of other workers. Weeks worked per year and professional hours per week are remarkably stable, and the number of patient visits per week, while up slightly from 1995, is still substantially below the rates reported for 1990 and 1991.
These data offer a few additional insights. Contrary to the expectation that an increasing proportion of physicians' time is wasted doing mindless paperwork for managed care plans, the proportion of professional activity hours spent in direct patient care has only varied between 0.82 and 0.84. Likewise, the allegations of having no time for patients are contradicted by an average time per visit (including all services) that has risen from 24 to roughly 26 minutes.
There is a quote attributed to H. L. Mencken: "When they say 'It's not about money,' it's about money." But, that doesn't mean it's only about money. The contracts associated with managed care had several other important implications, many of which are overlooked from the general [End Page 960] [Begin Page 962] policy perspective. One is that plans require the physician to bill them, not the patient, although copayments could be collected directly from the patient. From the patient perspective, this is clearly desirable, as anyone who has dealt with a shoebox full of bills, receipts, and insurance claims will attest. In some ways, this is no different than the old Blue Shield system, with its service benefits. However, rather than having to do this for one carrier (there is only one Blue Shield in any one geographic area), a physician might have to deal with literally dozens of managed care plans, each with a different claims form. While direct payment from plan to physician makes sense when there is only one plan, or a uniform process, it definitely increases the physicians' administrative burden when there are a multitude of plans.
Managed care, however, is more than just discounted fees. In fact, many, but not all observers do not count preferred provider organizations (PPOs) as managed care because all they offer is discounted (fixed price) fee-for-service coverage. Instead, managed care typically includes some mechanisms that can plausibly control the number and mix of services rendered. This is often accomplished through some type of prior authorization, either by the plan itself, or by a primary care gatekeeper who has to approve referrals and selected services. It is through such tools that the large insurance carriers sought to bring to their broad networks of providers the volume control so apparent in the group and staff model HMOs with whom they were trying to compete. Group/staff HMOs can establish practice patterns through informal guidelines and implicit triage if their patient load grows faster than the providers' capacity. Broad-based networks have excess capacity and, because they include nearly all the physicians, their practice patterns are the community standard.
Much of the practice of medicine is not based on well-documented evidence, but this is not to say that physicians explicitly recognize the weak foundations on which they sometimes have to base their decisions. In fact, social customs and patient expectations are such that we invest great confidence and certainty in our physicians and expect them to know the correct facts and make correct decisions. Medical professionals are expected to have access to specialized knowledge that is unavailable to the general population. Requiring that the physician seek prior approval before following his or her best judgment strikes at the core of this self-perception. If such prior approval must be granted from a physician who does not know the patient, most likely a primary care physician "ruling" on the judgment of a specialist, the reaction can be anticipated. Frequently, however, the decision is to be made by a nurse or [End Page 962] clerk, even more directly challenging the role of the physician as a professional.
It is clearly infeasible for a nationwide carrier to have patients routinely examined for such approvals, so they have to rely on the 1-800-MOTHER-MAY-I model of telephone approval. Yet they recognize the danger in attempting to actively manage care in this way. In practice, the vast majority of requests is approved, and most denials are made on the grounds of coverage, not appropriateness issues. That is, the plan may deny payment for a service because it is experimental, and the contract excludes such services, or the physician is recommending a drug or device that is not covered under the plan. The plan can then say that it is not deciding on the medical appropriateness of the service, which would arguably require a complete examination of the patient, but it is merely saying that the contract excludes certain services.
It is easy to understand the clinician's frustration with such an approach. The requirement that approvals be requested is demeaning, and it rarely seems to be based on clinical judgment but rather is couched in arcane issues around coverage. More importantly, most physicians have contracts with dozens of managed care plans, and some have many different benefit packages, reflecting the demands of their customers, the large employers. Prior to managed care, this variability was present, but the physicians were able to simply bill the patient and let him or her worry about whether the plan would reimburse to cost. That risk, and the attendant paperwork headaches, has now been shifted to the physician.
Moreover, there are few alternatives to feeling managed. Medicare is by far the largest single payer of physician services in the United States, and it provides very strong incentives for physicians to be "participating providers." From the beneficiary perspective this is very good, because it places a cap on fees and requires the physician to submit claims directly to the fiscal intermediary. Yet, from the physician perspective, it is one more set of constraints, and one from which it is almost impossible to escape.
Feeling Managed without Being Managed Effectively
It appears to be the case that, at least from the perspective of physicians, managed care has led to the worst of all possible worlds and in part this appears to be a reflection of the terminology as well as the methods used. The industry seems to have chosen the term managed care to differentiate [End Page 963] its product from the simple reimbursement approaches used before. After all, a "managed care product" seems truly different and worth buying instead of plain vanilla insurance. However, though a plan may attempt to develop guidelines and encourage, or even require, its physicians to alter their practice patterns, the multiplicity of plans and guidelines means that little can happen outside of integrated delivery systems that can provide clear and consistent messages.
In spite of all these pressures, physician incomes have not fallen and visit times have not shrunk, yet that is not the perception of most clinicians. It may be that the data inadequately capture the reality of what happens in the office. But, to a large degree, the reaction we see may be due to physicians' perceptions of loss of control. Two examples illustrate this. There are many complaints about the pressure on fees from managed care contracts and that it is hardly worth seeing patients any more. Prior to managed care, physicians "wrote off" roughly 15 percent of their billings as uncollectible, usually after months of trying to get paid. From an economic perspective, a 15 percent fee reduction by a managed care plan, with a guarantee of payment, would actually be advantageous because of the increased certainty. Yet, one can easily see how a physician would feel better reducing a fee for someone who can't pay than accepting a lower fee from a faceless insurance company. If the CEO of that firm gets paid enormous amounts, the anger will become rage--Robin Hood could not stand being bested by "Little John."
The surprising constancy of time spent with patients shown in Table 1 may also be explained in terms of perception. When Medicare implemented payment for hospital care based on diagnosis-related groups (DRGs), it published the average length of stay on which those payments were based. Hospitals and physicians then began talking about how patients needed to be discharged when they hit their DRG limit, that is, the published length of stay. Interestingly, an average was converted to a maximum, ignoring all those discharged earlier, and a payment limit was converted into a care limit. It may well be the case that if fees are set more rigidly under managed care, there will be less variability in patient visits, and physicians may feel they are cutting short visits they would like to continue longer. It may also be that in the past, extra long visits would occur and be billed for, but not reimbursed at the full fee, leaving the patient to pay the extra amount, or the physician to eventually write it off long after the clinical encounter.
To make the situation worse, many benefits attributed to the classic HMOs have not been transferred over to the newer forms of managed [End Page 964] care. The ability of the integrated delivery systems, such as Kaiser and Group Health, to monitor and improve their practice patterns, to reconfigure inpatient and outpatient care, and to follow their enrolled population is rarely matched in the contract-based plans. This is largely because they represent only a small fraction of any clinician's practice, and because the clinician has contracts with multiple plans. Some plans seek to narrow their networks and thus increase their ability to present a consistent message to physicians. Unfortunately, they lose attractiveness with payers, who want to be able to assure employees that they do not have to change physicians every time plans are switched in order to constrain premium growth.
Implications and Speculations
While it is not entirely clear what is happening with respect to managed care, the current situation does not appear to be stable. The multiplicity of plans and contracts imposes substantial transaction costs on both physicians and the plans, and more importantly, prevents the development of improvements in clinical practice needed to counterbalance the inexorable cost pressures of new technologies. Otherwise, medical care expenditures will start to rise again, with the inevitable political pressure to constrain at least Medicare and Medicaid costs. Furthermore, physicians will continue to feel managed, and will be able to report numerous examples of problems that impact, or could impact, patient care. They are likely to be more effective in making their case than the railroad firemen, especially given the popularity of insurance companies.
Change is likely, but it is difficult to foresee exactly how it will occur, or in what direction. A complete return to unmanaged fee-for-service is unlikely. Can one imagine Medicare eliminating its fee schedule, or demanding that beneficiaries return to filing claims in order to be reimbursed? Likewise, it is difficult to imagine a return to "anything goes" ordering of tests and procedures. Instead, there may be a movement to less intrusive, but more effective systems.
For example, careful reviews of the evidence, such as those supported by the Agency for Health Care Policy and Research through its evidence-based practice centers, may lead to the creation of widely accepted practice guidelines. Health plans may find that retrospective monitoring is sufficient to assure that the vast majority of physicians are practicing within the guidelines. While physicians in various areas may choose to implement the guidelines somewhat differently, there may be little to be [End Page 965] gained from having multiple guidelines imposed on a single physician--any one of the rules will be better than a cacophony.
If one simplifies the rules, one also eliminates the product differentiation among contracting plans. Prior to managed care we had numerous insurers all selling essentially the same product: reimbursement. Thus, one could conceive of a market for multiple health plans that would be stable, but that may not be where the future lies. Instead, we may find either the elimination of reliance on employer-paid coverage or a shift to employee-driven choice among a set of clearly differentiated health plans. Any choice-based scenario, however, must effectively address the problems of risk selection.
In any event, we should remember that, just as medical technology is constantly changing, our health policy options are also in flux. "Managed care" was not created as a grand policy solution but as a term given to a market creation. Managed care filled a particular niche at a certain point in time, but the way it has played out has created numerous counterpressures, which, in turn, will transform the system yet again.
University of California, San Francisco
Harold S. Luft is Caldwell B. Esselstyn Professor of Health Policy and Health Economics and director of the Institute for Health Policy Studies at the University of California, San Francisco. His research and teaching has covered a wide range of areas, including medical care utilization, health maintenance organizations, hospital market competition, quality and outcomes of hospital care, and risk assessment and risk. He chairs the National Advisory Council of the Agency for Health Care Policy and Research and is on the board of the Association for Health Services Research. In addition to numerous articles in scientific journals, he has authored a number of books, including Health Maintenance Organizations: Dimensions of Performance (1981) and Hospital Volume, Physician Volume, and Patient Outcomes: Assessing the Evidence (1990).
References
Bergthold, L. 1984. Crabs in a Bucket: The Politics of Health Care Reform in California. Journal of Health Politics, Policy and Law 9(2):203-222.
Evans, R. G., M. L. Barer, G. M. Anderson, J. Lomas, R. J. Labelle, C. Fooks, G. L. Stoddart, D. Feeny, A. Gafni, G. Torrance, and W. Tholl. 1989. Controlling Health Expenditure: The Canadian Reality. New England Journal of Medicine 320(9):571-577.
Gonzalez, M. L., and P. Zhang, eds. 1998. Socioeconomic Characteristics of Medical Practice, 1997-1998. Chicago, IL: American Medical Association.
Levit, K., C. Cowan, B. Braden, J. Stiller, A. Sensenig, and H. Lazenby. 1998. National Health Expenditures in 1997: More Slow Growth. Health Affairs 17(6):99-110.
Luft, H. S. 1987. Health Maintenance Organizations: Dimensions of Performance. New York: Wiley-Interscience, 1981. Reprint, with a new preface, New Brunswick, NJ: Transaction Books.
Miller, R. H., and H. S. Luft. 1994. Managed Care Plan Performance since 1980: A Literature Analysis. Journal of the American Medical Association 271(19):1512-1518.
------.1997. Does Managed Care Lead to Better or Worse Quality of Care? Health Affairs 16(5):7-25.
