Psychiatry, Managed Care, and Crooked Thinking

[robin_himself]

I found this article to be very close to my experience with certain docters. Its a long read but you'll find it to be very true. Please note your own experience.



In a context of unrestrained growth of American medicine, it became evident that restraints were needed, or theoretically the entire economy would eventually become “medicalized.” With the failure of President Clinton’s health bill, the United States has experienced a proliferation of “managed care,” often operated as a profit-making venture. Virtually all systems of managed care or managed competition currently operating in the United States present major ethical issues for physicians, many of which are inherent in the system and its profit motive and most of which have been widely discussed in the literature.

For many centuries, medicine was a professional endeavor. In classic Greece and Rome, physicians did not charge patients for treatment, although Roman physicians did charge for treating slaves.1 Remuneration, freely offered, was not rejected. Historically, this ethos has had a major influence on the classic delineation of professions versus trades. The physician’s ideal was service and the advancement of the art, not the acquisition of large financial rewards. Although the modern era has provided many exceptions to this old rule, few physicians have acquired the amount of money paid to the chief executive officers of many of the managed-care firms. With managed care, medicine has become appreciably more entrepreneurial than it was a few years ago, and most of the managed-care companies expect to earn a profit.

The role of physician-reviewers in managed-care systems poses complex ethical and clinical issues for physicians treating patients. Public discussion of the role of physician-reviewers who work for managed-care companies is minimal. These physician employees of managed-care companies, who have never examined the patient, approve or disapprove the treatment decisions of the practicing physicians whom they indirectly supervise. They are often anonymous, and their credentials and work experience are frequently unavailable. Some have been known to misrepresent their credentials and experiences. For example, a physician-reviewer provided mandates regarding my care of an adolescent patient with severe anorexia nervosa. Because many of these mandates were clinically ridiculous, I asked him about his credentials and expertise with patients who had eating disorders. When he told me that he had headed the eating disorders program at a distinguished medical center, I talked to the current head of that program, who said that the reviewer had been a visiting resident for 3 months and had performed without distinction. During a holiday, however, he had been in charge of the eating disorders program, with backup by an attending physician, for 24 hours.

Kassirer2 addressed some of these issues elegantly and concluded, “…what oath, promise, or pledge did we ever make, either as individuals or as a profession, that obligates us to restrict care? We pledged, instead, to provide care.”

In this commentary, I address philosophic issues regarding managed care, with a primary focus on the physician-to-physician review process, in which a physician employed by the managed-care company discusses the case with the treating physician to approve or disapprove payment for treatment. The discipline of psychiatry had managed-care principles and reviews applied to it before most other medical specialties, and I hope that the points raised herein prove applicable to other specialties and helpful to other physicians.


PHILOSOPHIC ISSUES


Neurobiology

The traditions and philosophic underpinnings of business differ from those of medicine. Nevertheless, they have been imposed on medicine, and many physicians, without thoroughly thinking through the issue, are beginning to accept them.

Without a research base, managed-care companies opt for biologic treatment of almost all psychiatric disorders; indeed, those disorders that do not respond to biologic treatment are deemed, ipso facto, to be nonpsychiatric disorders. The twisted thinking of such argumentation is fascinating. Managed-care companies do not acknowledge principles of medicine such as the biopsychosocial model; the distinction among disease processes, syndromes, and illnesses as encountered by patients; and differences among patients. Decisions are presented in pseudoscientific verbiage but are driven by the marketplace. For example, attention-deficit disorder is a common disorder of childhood with pronounced morbidity. An extensive biologic research base has demonstrated a state of hypoarousal in the reticular activating system. Numerous other biologic findings are relevant to this syndrome. Nevertheless, perhaps because it is common, chronic, and expensive to treat, some managed-care companies believe that it does not exist or that it is a “social problem.” The hazard is that, over time, clinicians can easily acquire the same style of skewed thinking.

In managed-care medicine, patients seem to be viewed as widgets on an assembly line. The assumptions are that patients are somehow standard, that their illnesses are identical to the illnesses of all other patients in the same diagnostic group, and that treatments should be the same and precisely standardized. Individual patients are not considered in this type of medicine.

In my opinion, no work in psychiatry during the past half century has had the excitement and promise of the neurobiologic work that is being done. Neurobiologic philosophers and philosophic neurobiologists have made substantive findings and constructed complex models during the past 20 years. Officials of managed-care companies, however, have oversimplified the substantial advances that have been made and have ignored the fact that we are still in the earliest phases of understanding the relationships between brain and mind, which may ultimately require complex metamathematical models, as Mender3 has thoroughly described. Although drugs have been developed that are heuristically useful in the treatment of certain psychiatric syndromes, ridiculous conclusions are sometimes stated by managed-care reviewers. Because an agent that seems to help depressed people inhibits the reuptake of serotonin, as well as many other pharmacologic actions and epiphenomena, managed-care reviewers argue that it is evident that depression must be a disorder of serotonin regulation.

In my experience, a patient’s treatment is rarely approved by a managed-care company unless the patient is receiving medicine.

Recently, I treated a 16-year-old girl who had been doing well in life until 3 weeks before admission, when she had sudden onset of a complete inability to walk. After an initial misdiagnosis of a neuromuscular disease, it was concluded that the patient had a conversion disorder, and she was referred to our facility and hospitalized. She had classic conversion with “la belle indifférence.” She was apparently free of conflict and reported that she had never even had a bad dream. She was treated with psychotherapy, physical therapy, and imagery and relaxation. Two days after admission, she began to experience dysphoria and to express her sadness and anger. To me, this was a hopeful sign. The managed-care reviewer, however, viewed this behavior as a sign that the patient was depressed and insisted that she be treated with a high dose of fluoxetine. I told him that I disagreed completely, that, in fact, her dysphoria about her situation was a sign of progress. I refused to treat her with an antidepressant, and he discontinued payment for her hospitalization. Fortunately, the girl’s parents could afford her treatment and agreed to pay for it. The initial treatment regimen was continued, and the girl was walking normally within 3 days. Thus, this neurobiologic reductionism, based on a poor understanding of both neurobiology and psychiatric phenomena and presumably favored by managed-care companies because of the apparent cost savings involved, can harm efficacious, brief, and definitive treatment.

As Karlsson and Kamppinen4 pointed out, reductionism is not a “bad” process. Scientific advances occur in part because of reductionistic thinking, but the reductionistic thinking of scientists and that of managed-care companies may be very different entities and processes. Human beings are complex, and l’ homme machine is as remote from our experience today as when de la Mettrie5 proposed it centuries ago. In managed-care settings, many psychiatrists seem not to contemplate the assumptions behind their psychopharmacology. Human beings are neither computers nor masses of neurons. With billions of neurons functioning in our brains, a reasonable assumption is that humans are complex.

Outcomes

“Outcome measures” are an unusual blend of medical research, quality improvement approaches, and business strategic outcome applications. “Clinical outcomes” are not always clinically meaningful.

In managed-care approaches, so-called providers—individuals or agencies—are asked to demonstrate that they have superior “outcomes,” usually in relationship to a specific clinical entity that may in fact be an artificial composite of several clinical entities. Good outcomes for providers lead to a better relationship with the managed-care company and ultimately more money or less supervision (or both). Thus, in contrast to the situation in research, a major incentive exists for an outcome to be “good”—indeed, it is expected to be good. Therefore, efforts are made for the studies to demonstrate good results.

Outcome studies often seem slanted toward convincing readers that one facility is better than another. Thus, short-term rather than definitive results are usually analyzed. By design, factors are few. Statistics are often extremely primitive, and the techniques of design of experiments are often inadequately applied. Usually, actual outcome, which is complex and multifaceted, is not studied; instead, a perceived correlate of good outcome is studied. Occasionally, some validity and scientific rigor are demonstrated in the study of this correlate but not always.

For example, an adolescent patient was involved by his managed-care company in an outcome study of hospital treatment of schizophrenia. No effort was made to categorize schizophrenia by type—it was assumed to be a unitary entity. The patient did not actually fulfill criteria for schizophrenia. The perceived correlate of good outcome being studied was outpatient compliance with medication. Of note, compliance with medicine is often a necessary condition of “good” outcome, but it is certainly not a sufficient condition. This particular patient complied well with his medicine—I do not think that he missed a dose. Indeed, he helped bolster the statistics of this particular outcome study and had one of the best “outcomes” in the cohort. Clinically, however, he took not only every dose of his neuroleptic medicine but also every dose of psychotomimetic substances, especially LSD (lysergic acid diethylamide) and phencyclidine, that he could obtain. He was soon completely out of contact with reality and had to be rehospitalized immediately and for a long time frame. Thus, his good “outcome” was an extremely bad outcome.

True research on outcomes necessitates a knowledge of the natural history of syndromes and their many variants, much of which remains obscure. Good outcome studies also necessitate complex design and multivariate analysis.

The outcomes movement in psychiatric managed care is inherently simplistic and made even more simplistic by the implicit assumption that all syndromes and their variants are diseases. In fact, many psychiatric syndromes seem to be final common pathways with differing causes and outcomes. For example, a child can be oppositional and defiant during the course of a decade or more, or he can have development of oppositional and defiant behavior in the context of grief. These hardly represent the same pathologic process—indeed, in child and adolescent psychiatry, “pathology” is sometimes adaptive—but this is assumed in many outcome studies.

If there is to be outcomes research, it should be research—the scientific examination of important questions, with peer-reviewed design and data analysis—and not a series of venal efforts to convince businessmen in managed-care companies that one treatment is better and less expensive than another based on statistical tricks and sleight of hand.

Some colleagues in psychiatry, numerous colleagues from other mental health disciplines, and many residents view outcomes studies as synonymous with research. This perception is unfortunate and augurs poorly for the scientific base of clinical psychiatry.

Logical Errors and Errors of Argument

Many physicians spend a substantial amount of time discussing cases with managed-care reviewers for the purpose of continuing treatment for their patients. This phenomenon is considerably more widespread in psychiatry than in most other medical specialties, but in time, it will be reality for other specialties as well. In my experience, many physician-reviewers use logical and semantic tricks that confuse physicians treating patients.

In his old and useful book, Thouless6 discussed so-called straight and crooked thinking, and Asher7 directly applied Thouless’ concepts to medicine in his hilarious and incisive article entitled “Straight and Crooked Thinking in Medicine.” Physicians continue to observe all the illogical thinking that Asher discussed. Herein I apply some of these principles to situations involving managed-care reviews of patients’ treatment.

Post Hoc Ergo Propter Hoc.—One of the most common logical missteps is the post hoc ergo propter hoc argument. Many managed-care reviewers use this liberally, and they assume a causal relationship between temporal events.

For example, a 15-year-old girl was being treated for severe anorexia nervosa. She was making considerable progress. Nevertheless, the managed-care reviewer demanded that she be treated with the drug sertraline, and if she did not receive this drug, her bills would no longer be paid. No pressing clinical rationale existed for the use of this medicine—the girl was not depressed, and the drug has no demonstrated efficacy in the treatment of anorexia nervosa. Because the patient continued to require the treatment that was being provided and because use of sertraline seemed relatively benign, I acceded to the demand of the company after a thorough and frank discussion with the parents and the patient and with their consent. The patient continued to progress, and the eventual outcome was good, as expected. At the end of treatment, the managed-care reviewer said to me, “Well, now you know that these people respond to sertraline.”

Reification of Syndromes.—Another logical lapse in the managed-care industry is the reification of syndromes, in which a constellation of behaviors is assumed to be a disease process.

All Instead of Some.—A frequently demonstrated logical error is the implication of all when only some is supported. A managed-care reviewer insists that patients taking specific serotonin uptake inhibitors will do better when a tricyclic agent is added. This is sometimes true but certainly not always.

A common, illogical argument by reviewers occurs when the patient has more than one syndrome. Basically, this argument is that, because Y is worse than X, no rationale exists for treating X. Adolescents who have been horrifically abused and have posttraumatic stress disorder often have major depression as well. Reviewers consistently argue that, because major depression in adolescents can be life-threatening, only the depression needs to be treated. They do not acknowledge that each patient’s illness is unique, multifaceted, and complex.

Splitting the Difference.—When the clinician and the managed-care reviewer disagree, some reviewers suggest the adoption of a mean between the two extremes—“splitting the difference.” If the disagreement concerns whether to use 20 or 40 mg of fluoxetine, a trial of 30 mg may be reasonable. If, however, the physician advocates a vigorous effort to treat the affective component of a psychotic illness and the reviewer disagrees, compromising on a homeopathic dose of an antidepressant makes little clinical sense.

Circular Arguments.—Circular (and repetitive) arguments by managed-care reviewers are commonplace, and tautologies and speculations are omnipresent.

Techniques of Argument.—The logical errors I have detailed are sometimes masked or even overwhelmed by the techniques of argument used by some managed-care reviewers. Discussions are often filled with emotionally toned words as well as caustic, ad hominem comments about the expertise of the physician. Under attack for incompetence, the treating physician may become grateful for any crumbs thrown his way. He may accede entirely to the demands of the reviewer simply to demonstrate that he is competent.

Diversionary Tactics.—Irrelevant diversions and objections are a frequent component of the managed-care reviewer’s armamentarium. Threatened on an irrelevant point, the physician may be willing to agree to the reviewer’s proposed protocol.

Suggestion by Prestige.—Managed-care reviewers also use the technique of suggestion by prestige. For example, the previously mentioned reviewer claimed that he had headed a prestigious program. This is a clear-cut instance of the use of prestige by false credentials. Other methods include name-dropping of experts and the facile use of jargon. Imperfect and false analogies are common. Reviewer techniques to anger the clinician often lead to a poor argument by the clinician. One approach to induce this angry response is the attribution of prejudice or other motives to the clinician.

Pseudoscience.—The use of pseudoscientific models by managed-care firms also causes many problems. Most of the companies are moving toward symptom-specific protocols for treatment. This approach is based on an assumption that syndromes are diseases and that their course is relatively invariable. Such protocols do not consider the nature of clinical decision making, which is not a purely scientific process. For example, a seasoned, good clinician, when faced with a case of social phobia, is expected by the managed-care companies to obtain and implement a protocol that overlooks the intangibles of the individual patient. In fact, such a clinician actually reasons something like this: “This patient needs individual psychotherapy, probably of a cognitive type—at least at first. He may benefit from group psychotherapy, but only if the group is supportive and well led and only if he can tolerate it. Family therapy might help, especially if he can realize that his father has a forme fruste of the same thing he has, but we must be careful not to precipitate something worse in the father. Literature shows that antidepressants are helpful. In fact, this patient reminds me a little of a patient who did not respond to fluoxetine, and he also reminds me a lot of a patient in whom maprotiline was effective. So, I will probably try a monoamine oxidase inhibitor with him.” Such an approach is similar to that of the protocol: individual, group, and family psychotherapy plus an anti-depressant. The clinical reasoning, however, is full of caveats and (largely preconscious) inferences that the protocol cannot simulate. This sort of thinking represents clinical wisdom and should not be discarded.

Reliance on Protocols.—Protocols, used to their worst extremes, can endanger patients’ lives. The following three case vignettes are examples.

An 11-year-old girl with malignant, rapidly advancing anorexia nervosa was covered by a managed-care company. She was allowed, by protocol, to see a family practitioner for her syndrome, but the family practitioner was not allowed to monitor electrolyte levels because this type of follow-up was not in the protocol for this (temporal) stage of the disorder and its treatment. Her course progressed rapidly downhill. Her parents brought her to the emergency department after she had become weak and disoriented. Her potassium level was 1.9 mEq/L. She was admitted to the hospital. The managed-care company refused to pay for the hospitalization because the “outpatient protocol has not been completed.” One supposes that, in this instance, “completion” would have meant “death.”

An 18-year-old patient with anorexia nervosa was admitted to the hospital because of major electrolyte abnormalities and a cardiac arrhythmia. She had advanced secondary osteoporosis, with bone age of someone older than 70 years and a severe mixed neuropathy. The managed-care company offered only a 2-day hospitalization because she had not had extensive outpatient treatment.

A 15-year-old boy jumped off a bridge, which spanned a freeway, and certainly would have died had a policeman not caught him as he was falling. The boy was then hospitalized. Because this was the “first episode” of major depression, the managed-care company agreed to pay for a 23-hour period of observation.

Of course, in all these cases, we hospitalized the patients for a considerably longer time frame than that recommended by the managed-care companies, without receiving further compensation.

Traditions

Medicine has rich traditions that date back at least 3,000 years. Managed care has few traditions and relies on the ethos of the marketplace. The following case vignette illustrates this situation.

A 16-year-old patient had taken a massive overdose of acetaminophen. She had severe liver damage but recovered after a few days in the intensive-care unit. She was suicidal but did not want to be in the hospital. Thus, she was transferred to the adolescent psychiatry unit, begging to be dismissed. I discussed the case with the physician-reviewer of the managed-care company, and he agreed that she could stay in the hospital. Later in the day, however, a nurse at the managed-care company telephoned the patient and asked her if she could “contract” not to kill herself. The patient, of course, agreed. The managed-care company then refused to pay for further hospitalization because the girl had “contracted for safety.” Such “contracts” are not useful. These contracts have rapidly become lore and tradition in managed care, but they have little established research base. Furthermore, the nurse who telephoned the patient had no fiduciary relationship with the patient.

Language

Managed care is evolving its own language, a strange and nongrammatical language. Words make a difference.

In addition to a substantial amount of jargon, the managed-care language is replete with “verbing” and “gerunding.” In this unusual language, day hospitalization programs are referred to as day; partial hospitalization is referred to as partial; extended outpatient efforts are called extended; outcome measures are termed outcomes; and psychiatrists, as Horton8 has pointed out, are relegated to performing med. checks and psych. evals. Syntax and grammar are fascinating: “I’m going to med. check him, and then he can probably go to partial.” All this is, of course, at the cutting edge.

Language does matter. The Orwellian logic and semantic confusion described in this article have an effect on the providers of care. Physicians need to rationalize what they do. I recently heard a valued colleague say that “discharge advocacy is the highest ethical obligation.” Hippocrates and Maimonides would be surprised. Another colleague speculated that “the hospital isn’t really a place to treat patients.”

In the linguistics of managed care, there is little room for doubt, ambivalence, or abstraction. Less is more.


CONCLUSION


Psychiatry has been rapidly trivialized into psych. evals and med. checks, and many psychiatrists, intimidated by he comments and techniques of managed-care reviewers, comply totally. As Horton8 wrote, “Most of all, I am saddened when colleagues silently bow to this business pressure to do less in less time, as if it were proven to be better or more ethical treatment.” Patients, poorly served, are losing trust in their physicians and in medicine, as Rogers9 has described.

Romano10 offered some thoughtful comments about this situation in one of the last articles that he wrote.

…it seems to me that we suffer from a deficiency disease. I think there is marked evidence of a deficiency of outrage. I do not think we are sufficiently outraged when we consider how pervasive are venality, corruption and ineptness….Are there not sufficient reasons to be outraged? Perhaps the magnitude of this behavior is such that we have become benumbed and anesthetized or feel powerless to do anything about it.

Perhaps, we are not completely powerless in facing the onslaught of the managed-care ethos. Strategies to challenge this system may come from an awareness of its ethical shabbiness and its logical fallacies. As professionals, we should realize this tawdry structure and behave accordingly. In particular, we owe it to our patients and our profession to point out ethical and logical fallacies as we deal with physician-reviewers.

www.mayoclinicproceedings.com

Reputation Comments on this post:

	Excellent article!
	You do a great job of newshawking. This article was a good find and outlines many similar notions to my own, having wor...