Health Affairs, 22, no. 5 (2003): 8-20 doi: 10.1377/hlthaff.22.5.8 © 2003 by Project HOPE	New Online   Obama's Success & Next Steps   3 Health Policy Options for President Obama   A Hybrid Proposal   Bipartisan Compromise

This Article Abstract Reprint (PDF) Submit a response to this article Alert me when this article is cited Alert me when Comments 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 ISI Web of Science 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 ISI Web of Science (22) Citing Articles via Google Scholar Google Scholar Articles by Mechanic, D. Search for Related Content PubMed PubMed Citation Articles by Mechanic, D. Related Collections Access To Care Insurance Coverage Mental Health/Substance Abuse

Mental Disorders

Is The Prevalence Of Mental Disorders A Good Measure Of The Need For Services?

PROLOGUE: Although President George W. Bush warmly endorsed expanded mental health parity legislation in April 2002, congressional action has been slow in coming. The primary sticking point has been whether the conditions covered by such a law would be broadly or narrowly defined. Congress may be forgiven for having stumbled into an area with no easy answers, the following essays suggest. Advocates for increased access to mental health services have damaged their own case at times by citing inflated estimates of the prevalence of mental disorders—based on expansive definitions—which have both undermined advocates’ credibility and increased fears that parity would lead to uncontrolled spending increases. In fact, states that have passed strong parity laws have not seen rapid spending increases because managed behavioral health care has reduced the use of costly inpatient care more than enough to offset increased use of outpatient services. On the other hand, if lawmakers insist on laboratory evidence to validate every claim for a behavioral health benefit—a standard that is not applied to medical benefits—then adequate coverage will be impossible to achieve.

In his thoughtful examination of the appropriate foundation for coverage decisions, David Mechanic finds "no neat solutions." Prudent use of resources must somehow be balanced with an effort to overcome widespread undertreatment described in the 1999 U.S. surgeon general’s report on mental illness. Patients’ perceived need for services might not provide a reliable standard for coverage, since subjective factors could bias self-estimates either upward or downward. Objective functional impairments are meaningful but difficult to measure. "It is an illusion to believe we can avoid muddling through to some extent," Mechanic concludes. In a Perspective following this paper, Darrel Regier takes a more optimistic view of the progress that has been made in refining the science underlying mental health diagnostics and epidemiology.

Mechanic is the René Dubos Professor of Behavioral Sciences at Rutgers University and directs its Institute for Health, Health Care Policy, and Aging Research. He has held a variety of university and government positions, including extensive involvement in the epidemiological studies referenced in his paper.

Abstract

 
Mental disorders are highly prevalent, but prevalence is different from need for treatment. Some mental disorders are a major source of distress, disability, and social burden, and many people who could benefit from treatment do not receive it. Need is typically self-defined or defined by clinicians who are motivated to bring treatment to those who could benefit. Defining need appropriately requires consideration of the duration and reoccurrence of disorder, associated distress and disability, and the likelihood that treatment will be beneficial. Demand may be promoted inappropriately by clinicians and drug manufacturers who profit from expansion of demand. Future assessments of need must be based on evidence and take into account priorities for care and cost-effectiveness.

Others advocate comprehensive coverage for a limited group of serious mental illnesses such as schizophrenia and bipolar disorder. There is agreement that people with high levels of need should receive care but considerable skepticism that the presence of disorder itself is a reliable indicator of need. The crux of the issue is the relationship between psychiatric disorders as defined by the Diagnostic and Statistical Manual (DSM) of the American Psychiatric Association and a credible concept of need as seen by policymakers and the public. The discussion that follows seeks to illuminate this issue.

Concepts Of Disease And Disorder

Top Concepts Of Disease And... Community Prevalence Studies And... What Do Prevalence Estimates... Why Is Mental Health... Need For Treatment And... Toward A Reasonable Conception... Is Benefit From Treatment... Editor's Notes NOTES

 
As medical science has progressed, its categories have evolved from vague symptom constellations that reflect the suffering of complaining patients to specificity of disease definition as measured by objective observations, laboratory procedures, and imaging devices.¹ Nevertheless, many patients come to doctors with complaints that do not fall within precise parameters, and much of the stuff of everyday medicine involves ambiguity, uncertainty, and contested definitions. The practice of medicine is defined as much by patients’ expressions of distress and dysfunction and patterns of seeking help as by diagnostic assessments and evidence of treatment efficacy.

The uncertainties of diagnostic medicine and treatment have been compounded as increasingly sophisticated instruments for viewing and analyzing the body have become available and as the gap between what patients experience and what can be observed has grown. While, on the one hand, patients often have complaints for which a physical basis cannot be established, on the other hand, new technologies make finding bodily anomalies with no experiential reference and uncertain clinical significance more common. As William Black and Gilbert Welch have observed, "despite clinicians’ best intentions, many patients may have been labeled with diseases they do not really have, and many have been given therapy they do not really need."² The problem of assessing need in psychiatry is a subcase of such difficulties throughout medical care.

Description versus diagnosis. Mental disorders continue to be among the most contested disorders because of the lack of laboratory measures that substantiate patients’ complaints or disturbing behavior. Clinicians have developed a large number of descriptive categories that attempt to characterize such problems in specific terms, but such diagnoses are more conveniences to assist communication and further inquiry than they are representative of confirmed diagnostic theories.

A confirmed diagnostic theory identifies a related cluster of signs and symptoms representing an underlying disorder; it then provides information on the expected course or natural history, causes, and needed treatment. Diagnoses are hypotheses, and to the extent that they are correct and scientific, they guide the clinician toward appropriate actions. Many medical diagnostic theories remain uncertain and are only partially confirmed, but diagnostic theories in psychiatry are more likely to be partially confirmed or unconfirmed with limited understanding of causes, uncertain understanding of course, and fewer specific treatments. Treatment, thus, is often based on clinical experience and experimentation.³

Burden of mental disorders. Nevertheless, mental disorders are a source of immense personal and societal burden; those labeled "schizophrenia" are among the most painful and disorienting in all of medicine. The major depressive disorders are more common but also a source of much misery and more burdensome and disabling than most common medical illnesses are.⁴ Alcohol and drug abuse conditions, perhaps among the most contested of major mental health categories, are a source of much trouble, however we characterize them.

Impact of psychodynamic theory. Theories are influential in how society characterizes mental disorders. For much of the twentieth century, psychoanalytic thinking had great influence on American psychiatry, and most disorders were viewed as arising from the biosocial development of the child.⁵ Accordingly, it was believed that most of us had dysfunctional emotions and behavior that could be corrected through assessing unconscious motivation and attaining greater understanding. More serious disorders were seen as exaggerations of these same processes.

The psychodynamic view of mental disorders was nonspecific, and efforts at careful diagnosis received low priority. A small number of psychiatrists brought a medical view to diagnosis and sought to develop a more precise system of psychiatric nosology, but they had little influence prior to the development of the third edition of the DSM (DSM-III).

Psychiatric epidemiology. Psychiatric epidemiology, which is the source of estimates of psychiatric disorder in community populations, developed out of the pragmatic need during World War II to screen prospective soldiers for psychiatric vulnerability and incapacity to avoid psychoneurotic breakdown in the armed services.⁶ Following World War II, important large-scale community surveys of psychiatric illness were initiated, but these were still largely fashioned within a developmental conception of mental disorder that viewed mental distress as a continuum from mild to severe in contrast to a set of discrete medical disorders different from ordinary distress.

The Epidemiological Catchment Area (ECA) studies, initiated by the National Institute of Mental Health (NIMH) following the presidential mental health commission’s report in 1978, was the first effort to bring a medical diagnostic conception to large-scale community surveys.⁷ This approach translated criteria for disorders in the DSM-III into question sequences that could lead to estimates of the prevalence of specific mental disorders.⁸ The purpose of the project was to obtain prevalence rates of specific mental disorders and to examine their relationships to a range of nonbiological variables such as demographic factors, family history, and life events. This major effort was then followed by the National Comorbidity Study (NCS) of a national representative sample of people ages 15–54.⁹ The NCS continues and is the source of most of the prevalence estimates now used for the most common mental disorders.

Community Prevalence Studies And Their Import

Top Concepts Of Disease And... Community Prevalence Studies And... What Do Prevalence Estimates... Why Is Mental Health... Need For Treatment And... Toward A Reasonable Conception... Is Benefit From Treatment... Editor's Notes NOTES

 
Many smaller epidemiological studies were undertaken prior to the ECA and NCS, with reported prevalence rates of mental illness varying from 1 percent to more than 50 percent of the population, depending on the restrictiveness of the definition.¹⁰ The ECA studies investigated a subset of the most common and important diagnoses in the DSM-III. Based on surveys in five study sites, researchers estimated that 32 percent of the population had one of the disorders studied at some point in their lives, and 20 percent had an active disorder. The NCS, which was based on a more restricted age group but studied a comparable set of conditions, estimated that almost half of the population had one of the disorders studied at some point in their lives and that 30 percent had such a disorder in the prior twelve months.

Advocates for the mentally ill were receptive to these overall high rates, and high rates for specific disorders as well, because they believed that this information helped to make the public case for the importance of mental illness and the need for public and legislative support. Moreover, the ECA and NCS studies showed that relatively few persons with these disorders were receiving any form of treatment and that there seemed to be great unmet need. No one, to the best of my knowledge, seriously argued that all or even most of those with a measured disorder required treatment, but advocates used these estimates to make the case for expanded treatment.

The high estimates of mental illness have received much public attention and commentary, but they are not plausible representations of need. These high estimates lack credibility with many policymakers, and to the extent that they are believed, they may convey the impression that the treatment of mental illness is a bottomless pit, not a cost-effective way of investing new medical resources. We will never really know whether the dissemination of high estimates was successful in bringing attention and support to mental health concerns, but we can examine the relevance of such estimates for ascertaining need and consider alternative approaches to making such judgments in a credible way.

What Do Prevalence Estimates Tell Us About Need?

Top Concepts Of Disease And... Community Prevalence Studies And... What Do Prevalence Estimates... Why Is Mental Health... Need For Treatment And... Toward A Reasonable Conception... Is Benefit From Treatment... Editor's Notes NOTES

 
Perceived need. Need is a slippery concept and can be viewed from at least three perspectives. It can be conceived as a judgment by people in pain and distress that they require help (perceived need). Many of the people assessed in surveys as having a mental disorder do not define themselves that way, or if they do, they do not want any formal assistance or intervention.¹¹ Perceived need also can include some judgments made by significant others and by public authorities that the person requires intervention and treatment regardless of their beliefs or wishes. Again, these issues are not unique to psychiatry. Much effort in medicine is devoted to convincing people to seek treatment for hypertension, high cholesterol, and other indicators of risk.

Many people said to have a mental disorder on the basis of epidemiological surveys report that they want to handle their distress themselves. Such preferences could be influenced by ignorance of available treatment or fear of stigma or shame. Educational campaigns might be carried out to reduce stigma and encourage people to be more accepting of interventions known to be useful, but until people change their conceptions, they cannot be said to have a perceived need. In contrast, many people who do not have a demonstrable DSM disorder but who perceive a need nevertheless seek medical or psychiatric care.¹² Such people are abundant in our health care system because they often place high value on medications and psychotherapy as a way of managing difficult life problems. Some observers derogate such people by calling them "the worried well," but the absence of measured disorder does not prove the absence of need.

Disability. A second concept of need, typically used by epidemiological analysts, is defined in terms of experienced decrements in capacity to perform usual social roles and activities—that is, disability.¹³ Disability can exist in the absence of a defined mental disorder. However, within this perspective the need for mental health services can be seen most typically as the presence of a mental disorder accompanied by disability believed to be associated with that disorder.

Demonstrated value of intervention. A third concept of need stems from new knowledge and technology and is defined by the scientific demonstration of the value of intervention. New drugs, technologies, or therapies are developed and demonstrated to reduce distress, disability, and even death. People might not perceive a need for care and might regard their distress as part of everyday living, but efforts can be made to encourage care because such an intervention has been demonstrated to be helpful. An exemplary case is the treatment of hypertension, which is typically asymptomatic but harmful to health. Few doubt the value of encouraging people with elevated blood pressure to seek treatment. There are similar instances in psychiatry. For example, there is evidence that antidepressants can be helpful to people with major depressive disorders who might not perceive a need for care or seek it. Thus, public health campaigns are initiated to encourage more people to seek treatment for depression, and efforts are made to improve the recognition and treatment of depression by primary care physicians.¹⁴

A major complication is that pharmaceutical companies, device manufacturers, and clinicians have strong economic motives to convince people that they have needs they might not recognize, which can be alleviated by new products and interventions. The profit motive encourages exaggeration of such needs and the benefits achieved by treatment. New technologies and drugs, and their promotion, often stimulate new needs for problems the public accepted as a matter of course. In this way, commercial and professional interests shape conceptions of need.

Nonintervention. The fact that a person has a disorder does not mean that treatment is necessary or desirable. Many disorders are mild, nondisabling, and self-limited and need no professional intervention. Whether interventions are needed could depend also on the sources of informal supports available to people and the strength of their coping resources. Excessive dependence on treatment also can diminish development of people’s own coping resources. Epidemiological studies find that most people have some somatic symptoms most of the time, but most are mild and nondisabling.¹⁵ If we reported medical disorders the way we report mental disorders, the headlines would have to declare that almost everyone is sick much of the time. Such headlines would not be incorrect but trivial. Dandruff is a disease, but it need not necessarily be treated in most instances. There is no disagreement that viral respiratory infections are diseases, but most require no intervention. Most people most of the time have symptoms they can present to a doctor that will lead to a diagnosis.

Changing the mental disorder criteria. As criticisms have grown about the expansiveness of the DSM-IV and the epidemiological estimates based on the DSM criteria, efforts have been made to revise prevalence estimates to make them more credible. One recent example is an effort by William Narrow and his colleagues to reduce ECA and NCS estimates by adding "telling a professional about the problem" and "using medication" to the list of criteria for indications of dysfunction.¹⁶ By using such criteria, Narrow and colleagues reduced prevalence rates of disorder by 17 percent in the ECA studies and by 32 percent in the NCS, which resulted in an estimate of any disorder among adults ages 18–54 of 18.5 percent.¹⁷

Similar exclusion criteria have been used in the National Health Interview Survey in assessing acute symptoms to reduce the reported prevalence of common, mild, self-limited symptoms, which are believed to be ubiquitous, unreliably reported, and of little medical significance.¹⁸ Such criteria, however, hopelessly confuse the prevalence of disorder with the social processes of illness behavior and help-seeking. The resulting estimates may seem more credible and more statistically reliable, but they are constructed on a misguided theoretical basis.

Jerome Wakefield and Robert Spitzer have critiqued Narrow and colleagues’ re-estimates based on Wakefield’s theoretical concept of mental disorders as harmful dysfunctions.¹⁹ Mental health professionals commonly regard harmful behavior itself as a disorder, although much that is harmful or undesirable is not necessarily "sick." But there is a strong inclination among the public and some clinicians to label behavior we dislike or do not understand as "sick." By Wakefield’s criteria, evidence of dysfunction is required.²⁰ There has to be a plausible case that some underlying process is not performing as it should, consistent with human evolutionary history. Thus, a person who becomes depressed following the breakup of a valued love relationship might be functioning normally, while a person with comparable depressive symptoms without any adverse event to precipitate them can be said to have a mood dysfunction. Although the DSM makes exception for bereavement, it does not make exception for loss of a valued lover, loss of a job, or many other devastating losses. While the dysfunction criterion is theoretically useful, the difficulty with it is uncertainty as to when processes are inconsistent with evolutionary development and the absence of objective criteria to demonstrate such dysfunction. In extreme cases, dysfunction may be obvious, but the practical challenge is in defining ambiguous cases and not the extreme.

Why Is Mental Health Treated Differently?

Top Concepts Of Disease And... Community Prevalence Studies And... What Do Prevalence Estimates... Why Is Mental Health... Need For Treatment And... Toward A Reasonable Conception... Is Benefit From Treatment... Editor's Notes NOTES

 
Although some claims of disorder can be validated by independent laboratory and other investigations, many medical conditions and all major psychiatric disorders cannot be validated in this way.²¹ Nevertheless, judgments are needed for disability determinations, legal decisions, and insurance purposes. The preferred approach to establishing need is to make a functional assessment of the person based on the medical record and careful observation by an informed but unbiased professional. This is impractical in routine, everyday practice, and illness and need are most often self-defined by patients who are troubled in some way.

Medical treatment. The content of clinical practice is determined by patients’ seeking help and the complaints they bring. Medical practice is permissive, in that doctors treat a wide range of problems irrespective of whether they have confirmed disease theories or even whether a specific diagnosis is possible. In American medicine, despite managed care restrictions, specialty referral is relatively easy to obtain, and patients with medical complaints have little difficulty getting treatment if they have insurance or can pay. It is unusual to ask whether the patient is really sick or sick enough to justify the visit; the complaints and desire for help typically are seen as sufficient justification.

There are, of course, contested areas of care where patient and doctor might have perspectives that clash. These are typically areas where physicians can find little evidence for continuing complaints on repeated investigation or attribute the complaints to psychiatric factors, but patients insist on an organic diagnosis. This commonly is the case in disputes about chronic fatigue syndrome, chronic myalgia, chemical sensitivity disorder, Gulf War syndrome, chronic Lyme disease, and the like. Some patients with such conditions have organized, have networks of supporters, and constitute a social movement to change medical conceptions and practices. Such instances are important in understanding how conceptions of disease are socially constructed but statistically are relatively uncommon.

Mental health’s unique history. The treatment of mental disorders in the United States, in contrast, has had a unique history that distinguishes it from the treatment of other disorders. Prior to the mid-twentieth century, most moderate mental disorders received no treatment other than the care people received from their general physician and from family, friends, and clergy. The care of people with serious mental illness was a state responsibility, provided in custodial mental hospitals for people who could not safely be cared for by themselves or their families. Those who were affluent typically were more likely to receive psychotherapy from office-based psychiatrists or care in small private psychiatric institutions. Most, other than the most seriously and persistently ill, received no formal care.

By the 1950s American psychiatry was primarily oriented to more affluent private patients in office-based practice who were culturally attracted to therapy; the treatment was psychodynamic in orientation and rejected a medical model of mental disease.²² Few psychiatrists worked in U.S. public mental hospitals, where most people with serious mental illness were cared for. The segregation of such people in isolated custodial institutions probably contributed to the stigma associated with disorders of mental functioning. In addition, psychiatrists were separate from physicians in other medical specialties and commonly were held in low regard by their physician colleagues.

This traditional pattern began to change with deinstitutionalization of the mentally ill, the growth of health insurance for psychiatric inpatient care, the expanded training of psychologists and social workers, and the development of new antipsychotic and antidepressant medications.²³ Expanded insurance coverage allowed the development of specialized psychiatric and chemical dependency units in general hospitals, and by the 1980s psychiatry had tilted toward a medical model. But treatment standards remained unclear, mental illness continued to be greatly stigmatized, and employers and insurers limited their potential cost liabilities by restricting inpatient days, imposing high coinsurance, and having expenditure ceilings that were more restrictive than for other disorders.

Although insurance for mental illness was largely for inpatient care, outpatient coverage increased as well. Norms for the duration of psychotherapy were vague, and some patients were attracted to psychotherapy as a self-actualization experience. Psychodynamic therapies were highly attractive to some groups more for cultural than for medical reasons.

Insurers and payers protected against the fear that insurance would induce unnecessary demand by imposing high copayments and limits on visits and expenditures. Nevertheless, in private corporations where insurance was reasonably comprehensive, the use and cost of mental health services became a cause of growing concern. Employers turned to managed behavioral health care organizations (MBHOs) to manage mental health benefits, and such management now covers the mental health care of much of the population.

Managed behavioral health care was successful in reducing mental health spending in the private sector, largely through sizable reductions in inpatient days.²⁴ Increasingly, the public sector also has turned to MBHOs to manage services for the Medicaid population. There is much controversy about how well these organizations have performed and the extent to which care is appropriate and sufficiently intensive for the most seriously mentally ill. There are many indications of neglect of the most vulnerable and severely mentally ill people, many of whom are uninsured or are in Medicaid programs in states that provide shallow coverage. Repeated claims are made that our jails and prisons have become our new mental hospitals.

Need For Treatment And The Parity Debate

Top Concepts Of Disease And... Community Prevalence Studies And... What Do Prevalence Estimates... Why Is Mental Health... Need For Treatment And... Toward A Reasonable Conception... Is Benefit From Treatment... Editor's Notes NOTES

 
The concept of need is central in the parity debate. Although there is evidence that parity under a managed mental health program requires only very modest additional expenditure, opponents claim that it will provoke demand, increase costs, and encourage employers to eliminate insurance coverage.²⁵ One approach advocated to limit moral hazard is to offer parity only to those who are believed to be most in need—that is, people who have diagnoses typically defined as constituting disabling mental illnesses such as schizophrenia and bipolar conditions. Some advocates seek to differentiate so-called diseases of the brain from problems of everyday living.

Many of the conditions defined as diseases of the brain are highly disabling, but the distinction itself between disorders that are or are not brain diseases stretches credulity. Any condition that meets the criterion of harmful dysfunction is a disorder of the organism with both biological and social dimensions, and this applies across a wide spectrum of conditions in the DSM. The brain is implicated in all affect and behavior, and distinctions between brain disorders and other disorders—barring observable pathologies of the brain—have no scientific merit, although they might be pragmatically convenient. Such distinctions, however, are also unjust because people with disorders that do not fit this criterion could have as much pain and disability as those whose disorders fit the criterion and could benefit as much or more from intervention.

Toward A Reasonable Conception Of Need

Top Concepts Of Disease And... Community Prevalence Studies And... What Do Prevalence Estimates... Why Is Mental Health... Need For Treatment And... Toward A Reasonable Conception... Is Benefit From Treatment... Editor's Notes NOTES

 
The foregoing discussion suggests that prevalence of mental disorder alone is a poor measure of need. Similarly, selecting a small subset of the most persistent and disabling diagnoses excludes people who appropriately require care. Clinicians and researchers have understood this for some time, and they have used a variety of ways to estimate need more realistically by considering diagnostic category, persistence and recurrence of symptoms, comorbidity, and various measures of impairment or poor functioning. Analysts are often restricted in aggregate assessments by the limited range and quality of available measures that allow them to make more accurate estimates.

Researchers have introduced concepts such as impairment, duration, recurrence, and treatment in different ways. In some instances, these items are used in efforts to make prevalence estimates more plausible, as in the estimates by Narrow and colleagues noted earlier.²⁶ In other cases, disability measures are introduced to approximate which disorders need treatment; in still others, these criteria are used in combination with diagnoses to designate those in the population who can be said appropriately to have serious mental illness or serious and persistent mental disorders.²⁷ All of these efforts are helpful but not fully satisfactory.

Need for longitudinal data. A major deficiency of measures to assess need is the lack of adequate longitudinal data on the persistence of symptoms, conditions, impairments, and disabilities and how patterns change over time. Such data are indispensable in understanding the burdens of psychopathology and the benefits of treatment. Fortunately, second-wave data from the NCS and from the Healthcare for Communities (HCC) survey will allow us to begin to address such issues. Most studies in psychiatry are too limited in time period covered to allow accurate assessments of costs and benefits.

Need for care management. Managed behavioral health care as currently administered is poorly regarded by many mental health professionals and patient advocates, but if we are to have a responsive mental health care system that is equitable and affordable, the benefit will require management. Such management should be built on individual assessment of need, guided by evidence of treatment efficacy and by a knowledge base informed by understanding of the benefits of treatment in terms of impairments, disabilities, and social role capabilities. Although we remain at a distance from convincing cost-effectiveness studies across a range of disorders, such considerations must remain an important objective.²⁸

Is Benefit From Treatment A Reasonable Criterion For Defining Need?

Top Concepts Of Disease And... Community Prevalence Studies And... What Do Prevalence Estimates... Why Is Mental Health... Need For Treatment And... Toward A Reasonable Conception... Is Benefit From Treatment... Editor's Notes NOTES

 
The question remains: Should insurance cover mental health services of proven effectiveness if patients fail to meet the formal criteria of specified disorders? This is an increasingly important issue, as medical science and technologies bring new possibilities for improving performance, extending healthy life, and enhancing mood and quality of life. When only modest spending amounts are at stake, clinicians will find justifiable diagnoses. But as expenditure possibilities mount, the question of what should be covered or not and under what circumstances will become more pressing. This is not entirely new territory; we have already had serious debates concerning contraception services, in vitro fertilization, Viagra, and growth hormones, to name a few.²⁹ But such instances are the tip of the iceberg where payers, consumers, and conveyors of new drugs and technologies clash.

What about treatment for life events? These debates are particularly relevant to mental health. Should services be available to a person depressed over a major life event such as divorce and unemployment, even if the depressive symptoms are normal responses and expected to abate without intervention, if there is evidence that distress and disability can be reduced? What if such treatment prevents future distress and life problems? And if not, why is this different from the use of medications to reduce cholesterol, lower blood pressure, prevent migraine headaches, or whatever?

Fairer coverage needed. U.S. psychiatry over the past fifty years has been a deviant instance of the important medical principle that first priority should be given to people with the most severe illnesses and greatest needs. Studies have repeatedly found that psychiatric care, particularly at the outpatient level, is disproportionately used by people who are better insured and better educated. Comparisons with health insurance in Canada suggest that mental health care there is distributed in closer correspondence to need than in the United States.³⁰ This misallocation requires correction through fairer insurance coverage and appropriate administration of a parity benefit.

Stigma remains. Stigma remains as a strong inhibition on demand for mental health services and will continue to do so. Although more people find mental health services acceptable than in the past, only a modest proportion of those said to have mental illnesses seek or want care.³¹ Bringing people into care who can clearly benefit is a worthy goal.

Clinicians as decisionmakers. There are no neat solutions to the dilemmas that I have described. Ultimately, judgment is required to make determinations as to which patients have the greatest needs and the intensity of care they require. The most appropriate decisionmakers are clinicians with psychiatric expertise, guided by current knowledge of the evidence and best practice and appreciation of the costs and benefits of alternative treatment approaches. These clinicians should be in a position to exercise "expert" judgment, appropriately supervised, and not biased by self-interest to recommend or deny treatment.

Expert management. The expert management of the mental health benefit may need correction. Clinicians, utilization reviewers, and case managers who make need assessments must be properly trained and qualified and appropriately supervised. They must be current on evidence of treatment effectiveness and best practice. But much uncertainty still characterizes mental health management, and professionals with varying training and theoretical dispositions often disagree. Thus, it is inevitable that decisions will be contested. An accepted structure for reviewing and challenging decisions in a practical and meaningful way is essential. Such a process can help to fine tune the management process through iterative review and correction. Norman Daniels and James Sabin have offered a good starting point in their model of procedural justice that encourages careful deliberation, explanation for decisions, opportunities for questioning decisions, and correction of inevitable errors of judgment.³² It is an illusion to believe that we can avoid muddling through to some extent. The hope is that we can do so thoughtfully.

Editor's Notes

Top Concepts Of Disease And... Community Prevalence Studies And... What Do Prevalence Estimates... Why Is Mental Health... Need For Treatment And... Toward A Reasonable Conception... Is Benefit From Treatment... Editor's Notes NOTES

 
The work underlying this review was supported in part by grants from the Robert Wood Johnson Foundation and the National Institute of Mental Health. The views expressed are the author’s and imply no endorsement by these funders.

NOTES

Top Concepts Of Disease And... Community Prevalence Studies And... What Do Prevalence Estimates... Why Is Mental Health... Need For Treatment And... Toward A Reasonable Conception... Is Benefit From Treatment... Editor's Notes NOTES

 

1. R.A.Aronowitz, "When Do Symptoms Become a Disease?" Annals of Internal Medicine 134, no. 9, Part 2 (2001): 803–808.[Abstract/Free Full Text]
2. W.C.Black and H.G. Welch, "Advances in Diagnostic Imaging and Overestimations of Disease Prevalence and the Benefits of Therapy," New England Journal of Medicine 328, no. 17 (1993): 1237–1243.[Free Full Text]
3. It is essential to recognize that confirmation of diagnostic theories varies a great deal throughout medicine. A current diagnosis of pernicious anemia, for example, provides definitive guidance on the cause, course, and needed treatment. Many diseases, such as multiple sclerosis, Boeck’s sarcoidosis, and cancer are less well understood, and diagnosis provides only limited information. For further discussion, see D. Mechanic, Medical Sociology, 2d ed. (New York: Free Press, 1978), chap. 3.
4. K.B.Wells et al., "The Functioning and Well-Being of Depressed Patients: Results from the Medical Outcomes Study," Journal of the American Medical Association 262, no. 7 (1989): 914–919.[Abstract]
5. A.V.Horwitz, Creating Mental Illness (Chicago: University of Chicago Press, 2002).
6. D.Mechanic, Mental Health and Social Policy: The Emergence of Managed Care, 4th ed. (Boston: Allyn and Bacon, 1999), 47–55.
7. President’s Commission on Mental Health, Report to the President from the President’s Commissionon Mental Health (Washington: U.S. Government Printing Office, 1978).
8. L.N.Robins and D.A. Regier, eds., Psychiatric Disorders in America: The Epidemiological Catchment Area Study (New York: Free Press, 1991).
9. R.C.Kessler et al., "Lifetime and Twelve-Month Prevalence of DSM-III-R Psychiatric Disorders in the United States: Results from the National Comorbidity Survey," Archives of General Psychiatry 51, no. 1 (1994): 8–19.[Abstract]
10. B.P.Dohrenwend et al., Mental Illness in the United States:Epidemiologic Estimates (New York: Praeger, 1980); and President’s Commission on Mental Health, Report of the Task Panel on the Nature and Scope of the Problems, Vol. 2, Appendix (Washington: U.S. GPO, 1978).
11. D.Mechanic, "Removing Barriers to Care among Persons with Psychiatric Symptoms," Health Affairs (May/June 2002): 137–147.
12. In the NCS, 45.2 percent of people who sought help did not have a discernable mental disorder during the previous twelve months. I thank Rich Shank for running this calculation for me.
13. SamShapiro and his colleagues, participants in the ECA, made an early effort to measure need by combining diagnosis with disability measures. See S. Shapiro et al., "Measuring Need for Mental Health Services in a General Population," Medical Care 23, no. 9 (1985): 1033–1043.[ISI][Medline]
14. K.B.Wells et al., "Impact of Disseminating Quality Improvement Programs for Depression in Managed Primary Care: A Randomized Controlled Trial," Journal of the American Medical Association 283, no. 2 (2000): 212–220;[Abstract/Free Full Text]K.B.Wells et al., Caring for Depression (Cambridge: Harvard University Press, 1996); and M.Schoenbaum et al., "The Effects of Primary Care Depression Treatment on Patients’ Clinical Status and Employment," Health Services Research 37, no. 5 (2002): 1145–1158.[CrossRef][ISI][Medline]
15. L.A.Green et al., "The Ecology of Medical Care Revisited," New England Journal of Medicine 344, no. 26 (2001): 2021–2025.[Free Full Text]
16. W.E.Narrow et al., "Revised Prevalence Estimates of Mental Disorders in the United States: Using a Clinical Significance Criterion to Reconcile Two Surveys’ Estimates," Archives of General Psychiatry 59, no. 2 (2002): 115–123.[Abstract/Free Full Text]
17. Ibid.
18. D.Mechanic and M. Newton, "Some Problems in the Analysis of Morbidity Data," Journal of Chronic Diseases (June 1965): 569–580.
19. J.C.Wakefield and R.L. Spitzer, "Lowered Estimates—but of What?" Archives of General Psychiatry 59, no. 2 (2002): 129–130.[Free Full Text]
20. J.C.Wakefield, "Diagnosing DSM-IV: Are We Making Diagnostic Progress?" Contemporary Psychology 41, no. 7 (1996): 646–652.
21. In the history of psychiatry, once disturbances of behavior have been linked to specific lesions, such as paresis due to syphilis, they have typically been taken over by other specialties.
22. G.N.Grob, The Mad among Us: A History of the Care of America’s Mentally Ill (New York: Free Press, 1994).
23. Mechanic, Mental Health and Social Policy.
24. Ibid.
25. R.Sturm, "How Expensive Is Unlimited Mental Health Care Coverage under Managed Care?" Journal of the American Medical Association 278, no. 18 (1997): 1533–1537.[Abstract]
26. Narrow et al., "Revised Prevalence Estimates of Mental Disorders."
27. Shapiro et al., "Measuring Need for Mental Health Services"; R.C. Kessler et al., "The Twelve-Month Prevalence and Correlates of Serious Mental Illness (SMI)," in Center for Mental Health Services, MentalHealth, United States, 1996, ed. R.W. Manderscheid and M.A. Sonnenschein, DHHS Pub. no. (SMA)96-3098 (Washington: U.S. GPO, 1996), 59–70;National Mental Health Advisory Council, "Health Care Reform for Americans with Severe Mental Illnesses: Report of the National Advisory Mental Health Council," American Journal of Psychiatry 150, no. 10 (1993): 1447–1465[Abstract/Free Full Text]; and R.C.Tessler and H.H. Goldman, The Chronically Mentally Ill: Assessing the Community Support Programs (Cambridge, Mass.: Ballinger, 1982).
28. D.Mechanic, "Managed Care and the Imperative for a New Professional Ethic," Health Affairs (Sep/Oct 2000): 100–111.
29. See, for example, R.Klein and H. Sturm, "Viagra: A Success Story for Rationing?" Health Affairs (Nov/Dec 2002): 177–187.
30. R.C.Kessler et al., "Differences in the Use of Psychiatric Outpatient Services between the United States and Ontario," New England Journal of Medicine 336, no. 8 (1997): 551–557.[Abstract/Free Full Text]
31. Mechanic, "Removing Barriers to Care."
32. N.Daniels and J. Sabin, "Limits to Health Care: Fair Procedures, Democratic Deliberation, and the Legitimacy Problem for Insurers," Philosophy and Public Affairs 26, no. 4 (1997): 303–350.

                      What's this?