Article Re; Medical Necessity from New england Journal of Medicine

[Guest guest]

Interesting reading for the group. At least I find it interesting!

Ken Mailly, PT

President

RPA/Mailly Consulting Inc.

Orange/Wayne, NJ

NEJM Home | Search | Log On to Full Text

Table of Contents | Previous Article | Next Article

The New England Journal of Medicine -- January 21, 1999 -- Volume 340, Number 3

Who Should Determine When Health Care Is Medically Necessary?

In the United States there has been a radical shift in the power to determine when health care is medically necessary and therefore covered by insurance. From the 1950s through the late 1970s, physicians' medical opinions largely dictated coverage and were rarely challenged by insurers. Physicians no longer have this extraordinary level of autonomy. Insurers now routinely make treatment decisions by determining what goods and services they will pay for. The line between clinical decisions about necessary medical care and decisions about insurance coverage is particularly blurred in managed-care plans. The power of insurers to determine coverage potentially gives them the power to dictate professional standards of care for all but the wealthiest patients. At present, the power of insurers is constrained by law in two ways. First, insurance contracts typically use clinically derived professional standards of care as the basis for determinations of medical necessity and coverage. The courts frequently reverse insurers' attempts to deviate from these standards, particularly where the insurer has ignored the patient's medical record, the opinion of the treating physician, experts' opinions, and pertinent data from government-sponsored or peer-reviewed studies. (1,2,3,4) Second, in cases in which patients have appealed insurers' decisions under the Employee Retirement Income Security Act (ERISA), the courts have consistently overturned decisions that were arbitrary and capricious and that represented an abuse of the trust between the insurer and the insured. (1,2,4) The courts are particularly sensitive to the fact that insurers have a conflict of interest because they stand to gain financially from denying coverage. (3) ERISA regulates health and welfare benefits for more than 140 million workers and their families; thus, many determinations of medical necessity are now held to a minimal standard of fairness. Even this minimal level of protection is threatened, however. Insurers are learning how to write contracts so that standards of medical necessity are not only separated from standards of good clinical practice but also essentially not subject to review. (5,6) Insurers are also pushing for legislation to change ERISA. Legislation passed by the House (7) and introduced in the Senate (8) would have granted insurers the explicit power to define medical necessity apart from current standards of medical practice by allowing them to classify as medically unnecessary any procedure not specifically found to be necessary by the insurer's own technical-review panel. The Senate proposal would also have granted insurers the power to determine what evidence would be relevant in evaluating a claim for coverage and would have permitted insurers to classify some coverage decisions as exempt from administrative review. An alternative bipartisan bill introduced in the Senate would have required that insurers' standards of medical necessity reflect " generally accepted principles of professional medical practice " ; prohibited insurers from " arbitrarily interfer[ing] with or alter[ing] the decision of the treating physician " ; required insurers to provide an external administrative review of any coverage decision; and required that the external reviewer consider, among other evidence, the patient's health and medical information, the opinion of the treating physician, valid and replicable studies in the peer-reviewed literature, the results of professional consensus conferences, practice guidelines based on government-funded studies, and guidelines prepared by insurers that have been determined to be " free of any conflict of interest. " (9) None of these proposals became law, but similar bills are likely to be introduced in the next session of Congress. Groups such as the Advisory Commission on Consumer Protection and Quality in the Health Care Industry have recommended only procedural protection for patients, such as the right to a timely review of an adverse decision about coverage. (10) Although procedural reforms are important, no amount of procedural protection can help patients if insurers are given broad power to determine what standards will be used to make decisions about coverage. Such " protection " would be an instance of winning a battle but losing the war. In our view, an insurer should be able to set aside the recommendations of a treating physician only in restricted circumstances. Decisions about coverage should continue to be weighed against clinically accepted standards of medical practice. An insurer's decision should be lawful only if the insurer can prove that the decision rests on valid and reliable evidence that is relevant to a patient's individual circumstances. We advocate neither a return to total autonomy for treating physicians in determining insurance coverage nor a system in which insurers decide on coverage according to criteria that are totally independent of professional standards of clinical practice. Rather, we propose maintaining the middle position represented by current law. This middle position requires insurers to act reasonably and weighs the reasonableness of their conduct against professional standards of practice as reflected by valid and reliable evidence.

Background

Deciding whether proposed medical care is covered by insurance involves the interpretation of legal terms found in a health-plan contract. A contract typically defines covered services as those that are " medically reasonable and necessary. " Usually, coverage hinges on the answers to two questions. First, is the proposed service a type that is covered? For example, radiography is covered as a type of diagnostic service. Second, even if the contemplated care is a type generally covered, is its use medically reasonable and necessary in this particular case and thus warranted? To answer the second question, the decision-making process should be individualized and factual. Three general issues are made explicit in the case law governing decision making by private insurers. First, who has the power to decide what is medically necessary? Second, what evidence can and must be used to justify that decision? Third, who bears the burden of proof: must the insurer demonstrate with acceptable evidence that a recommended treatment is unnecessary, or does the patient bear the burden of demonstrating that the treatment in question is necessary? Traditionally, insurers considered care and services medically necessary whenever treating physicians said they were necessary. However, as the cost of health insurance escalated, commercial insurers, Medicare, and Medicaid began to review the medical necessity of physicians' treatment recommendations as the basis for determining which procedures and services would be covered. (11) Even so, for more than 200 years, the courts in malpractice cases have turned to clinical practices for evidence of when and under what circumstances medical care should be considered medically necessary, (12) and insurers therefore have relied on the prevailing clinical standards of care. Recently, however, it has been suggested that the definition of medical necessity should be separated from clinical practices. Eddy proposed that coverage of medical care should be determined largely on the basis of studies showing its benefit in a population. (13) Havighurst championed the inclusion of provisions in health-plan contracts that explicitly delineate covered and uncovered services and that generally separate coverage from what he termed " professional control. " (14) Implicit in such recommendations is the belief that insurers should have the power to make a conclusive and nonreviewable decision that a particular service is unnecessary and therefore excluded from coverage. Thus, for example, in applying a contract provision that excluded high-dose chemotherapy for certain forms of " advanced cancer, " an insurer would be allowed to decide unilaterally that a patient's cancer was sufficiently advanced to trigger the exclusion. In response to the pressures of cost, insurers may add provisions to contracts that exclude coverage of care for specific diseases and that link medical necessity to broad standards based on evidence involving large groups of patients rather than to clinically derived professional standards applied to particular patients. Such provisions would be unfortunate, for three reasons. First, a sole reliance on broad standards derived from generalized evidence is at odds with good medical practice. Second, there are practical limits to designing studies that can answer all clinical questions. Even if such studies could be conducted, it would be impractical to incorporate all the results into contract provisions. Third, much of the existing evidence that might be used to justify wholesale, nonreviewable exclusions of coverage is of insufficient scientific quality to displace clinical judgment.

Problems with the Use of Generalized Evidence of Medical Necessity

The practice of medicine has a core ethical dimension and requires that the physician use his or her knowledge of the particular patient in deciding on the course of treatment along with the patient. Thus, even when there are good data on the effectiveness of an intervention, it is bad medical policy to apply general rules arbitrarily to all cases. A recent case involving a child with severe cerebral palsy illustrates this point. The child's pediatrician and pediatric neurologist prescribed physical therapy to prevent muscular atrophy and possibly allow the child to walk. On the basis of a single article in a medical journal, however, the insurer's medical director decided that intensive physical therapy can never speed the development of such children. A federal appeals court found the decision to be without any rational basis and required that the insurer cover the physical therapy. (2) All standards of care, including those based on rigorous scientific evidence, implicitly or explicitly use outcomes as the benchmark, and the choice of an outcome is inherently value-laden. Take the example of a patient with a broken wrist. For most patients who perform ordinary activities with their hands, simply setting the bone might suffice to restore a reasonable range of motion. For a concert pianist, however, who requires a wider range of motion, more aggressive surgery might be essential. Which outcome should be the basis for the decision about insurance coverage: the functional level necessary for playing the piano or the functional level required for ordinary activities? Even in such a simple case, much less a more complicated one, this question cannot be answered without taking into account the value of the surgery, which in turn depends on the identity, goals, and values of the individual patient. Accounting for individual variation is part and parcel of clinical practice and is largely what medical practice is all about. Insurance provisions regarding medical necessity must be flexible enough to take into account the needs and circumstances of each patient. Decisions about insurance coverage that rely solely on broad standards derived from generalized evidence diminish the ability of clinicians to perform their roles, because they lock in certain outcomes and effectively impose " one size fits all " values. Such decisions also formalize the generalizations based on large-scale evidence and incorporate them into the insurance contract as a hidden limitation of coverage. At the extreme, physicians and patients are left with no discretion. The clinician's knowledge of the individual patient is rendered irrelevant.

Practical Limitations of Generalized Evidence

There are also practical limitations to basing decisions about medical necessity on generalized evidence. Much of medical practice is the result of tradition and collective experience. Many basic medical interventions have not been studied rigorously. Researchers often examine intermediate results or report process measures rather than ultimate outcomes. Furthermore, aside from a handful of procedures that are explicitly identified as covered or not covered, most medical care is not specifically described in a health-plan contract, and for good reason: the number of procedures and the circumstances of their application are limitless. No amount of effort could produce a contract that specified all of them. If coverage depended on generalized evidence concerning large classes of patients, then most care would not be covered. Many evidence-based studies of medical interventions examine their efficacy in a controlled clinical setting rather than in actual practice. (15) Interventions are often studied in isolation, yet physicians need to know the benefits of one over another. Moreover, randomized clinical trials are expensive to conduct, particularly if attempts are made to account for variations in the characteristics of patients, coexisting conditions, and other such factors. Given the enormous number of procedures and the individual circumstances of patients, limiting insurance coverage to a host of separately validated and specifically described procedures is impractical.

The Rigors of Science

In addition, we must consider the actual circumstances in which insurance companies apply evidence of medical necessity. Carefully developed criteria for coverage are of relatively limited consequence if the decision maker is allowed to disregard critical evidence or use flawed, biased, or otherwise unreliable evidence in deciding whether a particular patient's situation meets the criteria. Decision making by insurers about medical care is not well understood. In one study of the evidence considered by insurers in making decisions about coverage, researchers collected data on the process and information used by 231 physicians at health plans representing 66 percent and 72 percent of the U.S. population covered by health maintenance organizations and indemnity plans, respectively. (16) Medical directors were asked to rank the sources of information they used to make decisions. They were offered many choices, including medical journals, opinions of local or national experts, documents of the Food and Drug Administration, practice guidelines, Medicare policies, National Institutes of Health (NIH) consensus conferences, and the practices of other, larger plans. Even the most highly ranked source of information -- medical journals -- was used or considered optimal by less than 60 percent of the respondents. Most sources of information were used or considered optimal by 40 percent or less of the respondents, and information generated by the trade associations representing the health plans and undefined practice guidelines was ranked ahead of information from national experts, government documents, and NIH consensus conferences. In another study, insurers consistently denied coverage for growth hormone therapy for children, even in the case of diagnoses for which there was a 96 percent treatment-approval rate (virtual consensus) among surveyed pediatric endocrinologists. (17) These findings are consistent with the fact that materials such as the practice guidelines prepared by Milliman and on, (18) a well-known actuarial firm, often rely on insurers' own decisions rather than on well-designed scientific research. The need to control costs and generate profits also brings into question the reliability and soundness of decision making by insurers. The sine qua non of scientific research is the production of objective results, and objectivity is ensured through a process of open and vigorous debate among persons who have no financial stake in the outcome. Yet much of the decision making about insurance coverage is based on unpublished, proprietary, and unreviewed data. Furthermore, methods are undisclosed and unexamined unless litigation ensues. An endeavor that operates in this fashion and is subject to the conflict of interest inherent in the insurance industry cannot justifiably be called scientific.

The Need for National Reform

National legislation is needed to ensure continued linkage between decisions about the medical necessity of care for individual patients and clinical practice, for three reasons. First, the states have little or no power in this regard, because ERISA shields many or most decisions about medical necessity from the reach of state laws that govern insurance contracts. (19) Second, although patients may sue under ERISA to reverse an adverse determination of medical necessity, they can recover only the value of the benefits denied, not damages for injuries sustained as a result of the denial. Thus, even though a federal court ultimately reinstated coverage of physical therapy for the child with cerebral palsy, the family could not sue the insurer to recover damages for harm suffered by the child while they awaited the ruling. Third, even ERISA's limited remedy will be precluded, as insurance contracts are rewritten to give insurers the authority to base decisions about medical necessity on generalized, questionable evidence that is separate from clinical practice and that does not take into account the circumstances of particular patients. (5,6) National action is needed, not only to ensure a rapid and fair process for review of coverage decisions, but also to guard against arbitrary exclusion of services from coverage. Congress should enact legislation that maintains the link between medical necessity and professional practice standards. An insurer should be able to set aside the decision of the treating physician only if the insurer can show that the proposed treatment conflicts with clinical standards of care or that there is substantial scientific evidence, regardless of clinical practices, that the proposed care would be unsafe or ineffective or that an alternative course would lead to an equally good outcome. By substantial evidence, we mean a sizable number of studies published in peer-reviewed journals that meet professionally recognized standards of validity and replicability and that are free of conflicts of interest. These are the requirements currently written into most contracts and enforced by the courts, which require insurers to act reasonably and not arbitrarily. (1,2) There is obviously room for interpretation of what is meant by substantial scientific evidence and for debate about the validity and replicability of various methods of evaluation, such as randomized clinical trials, consensus-based standards of appropriate care, and meta-analyses. Similarly, there is room for debate about how generalized evidence might affect an individual patient. Such debate is necessary and healthy. Finally, Congress should ensure that patients have access to a speedy, external administrative review of all coverage decisions, not merely those that insurers decide are subject to review. Congress should also stipulate that external reviewers may consider a broad array of evidence.

Sara Rosenbaum, J.D. Washington University School of Public Health and Health Services Washington, DC 20006 M. ford, J.D. Rutgers University School of Law Camden, NJ 08102-1203 Brad , M.D., M.P.H. Phyllis Borzi, J.D. Washington University Medical Center Washington, DC 20006

Text

References

Table of Contents | Previous Article | Next Article

Copyright © 1999 by the Massachusetts Medical Society. All rights reserved.

[Guest guest]

thx alot...this will probably be applicable for the ethics piece in the

course on psycho-social aspects of illness and disability which i teach

in spring quarter.

ie on, DrPH

Associate Professor

Arizona School of Health Sciences

3210 W. Camelback Rd.

Phoenix, AZ. 85017

FAX

frankie@...

------------------------------------------------------------------------

[Guest guest]

Ken,

The NEJM on medical necessity was eye opening in that I had never heard

about proposals in the Congress and Senate to broaden insurer's ability to

deny coverage and then limit patient's and physician's recourse when they

disagree with the insurer's decision. It's bad enough now that insurers

cannot be sued for the decisions they make, now they are trying to escape

accountability on all fronts. It's incredible they try to do this when

providers are held so accountable for the decisions we make.

Mark Dwyer, MHA, PT

mdwyer1@...

------------------------------------------------------------------------