Limitations of the Diagnostic and Statistical Manual of Mental Disorders — also known as the DSM5
The Diagnostic and Statistical Manual of Mental Disorders (DSM-5; American Psychiatric Association, 2013) may be among the most controversial and polarizing books in the world. Informed by the medical model, the DSM-5 is the official diagnostic manual of mental disorders authorized by the American Psychiatric Association (APA, 2013). It has engendered debate in the public arena as well as professional circles. At the same time, it has been a best-seller on The New York Times, Amazon, and other book lists. Often referred to as the “bible of psychiatry,” it is required reading for mental health clinicians who seek insurance reimbursement and for students in graduate courses for the mental health professions. Yet the DSM-5 is at odds with many core values that therapists hold about clients and therapy, including an emphasis on the role of the therapeutic alliance and a wellness perspective.
Mental health treatment has become increasingly integrated with medicine. Surveys have found that a majority of primary care physician visits are related to psychological issues. Drug companies run advertising campaigns marketed directly to consumers using DSM-5 diagnoses. And prescription drug addiction has become an epidemic. A comprehensive research program also found that patients with schizophrenia who received low dosages of medication with individual talk therapy made greater progress over a two-year period than patients with schizophrenia who only got medication (Kane et al., 2015). Despite this landmark finding, we can expect the trend toward medicalization in our field will persist. Meanwhile, much of what talk therapy offers has been be lost to the field’s emphasis on the medical model.
Many therapists have also noticed that while their caseloads and paperwork have increased, their reimbursements have decreased, and they have lost control to insurance companies of making some of the key decisions in their clinical practices. Tuition for a graduate school education in a mental health discipline has also grown at a rate significantly higher than inflation often putting clinicians in huge debt. Nevertheless, the field continues to churn out therapists at a higher rate than ever while at the same we hear reports that there are critical shortages of mental health services. Insurance coverage for mental health care remains limited and there continues to be an ongoing struggle for parity among and between mental health providers.
My impetus for writing this article was a personal and professional journey that included developing a solution-focused approach to counseling (Guterman, 2013). Like many therapists, I was initially struck by the idealism of being a professional helper. In the 1980s, after working in various mental health settings including psychiatric hospitals, community mental health centers, and private practice, I became disillusioned by the field’s emphasis on diagnosis and psychopathology. In the 1990s, I shifted from a rational emotive behavior therapy (REBT) approach to a solution-focused model because I had sought alternative ways of conceptualizing clients, problems, and change.
Signs of Struggle for the DSM
The medical model holds that through scientific knowledge a physician can know the true cause of a disease, formulate an accurate diagnosis, and prescribe the appropriate treatment. A closer examination of the DSM-5, however, reveals that its diagnoses are not accurate representations of mental disorders and they are not necessarily effective in determining what treatment approaches are best for particular disorders. I am not negating that there is a phenomenon of mental illness, that people experience distress, or that there are deviations from norms, although the latter are based largely on cultural norms that we have created in the first place. Clinicians often overlook that the DSM-5, itself, provides a disclaimer that its manual should not be used as the sole basis for treatment planning and, instead, recommends a comprehensive evaluation be conducted for this purpose (APA, 2013, p. 19).
The DSM-5 is a nosological system which pertains to the science of classifying diseases. Classifying or diagnosing mental disorders go back millennium. Due to the constraints of language, one cannot not name or classify. Some kind of name or classification has always been needed to communicate about the problems that clients present for treatment. The question is for what purpose and how. For now, the DSM-5 is the prevailing diagnostic system of mental disorders and despite its limitations, serves as a common language for mental health professionals who work from diverse theoretical orientations and in different clinical settings. Imagine how a psychiatrist might react if, in a team meeting, I was asked for my DSM-5 diagnosis of a client and I replied, “I am sorry Dr. Jones, but I don’t find utility in using diagnostic metaphors.”
Although the DSM-5 often conflicts with therapists’ values, it is important to learn the language of diagnosis to communicate with colleagues. At the same time, the DSM-5 lacks scientific validity and reliability and does not adequately inform treatment. Nevertheless, insurance companies and funding agencies will not reimburse clinicians unless there is a DSM-5 diagnosis and treatment is organized around such diagnoses. This is a catch-22 for many clinicians.
The DSM-5 lacks validity and reliability
The roots of psychiatry can be traced to the premodern era which was informed largely by superstitions and corresponded to treatments which today are considered barbarous and inhumane. For example, trephining involved drilling a hole in the head to let evil spirits out. Exorcism, which is still practiced today, presupposes a person is possessed by demons. The religious ritual of exorcism is aimed at evicting such entities. During the civil war, when black slaves ran away to freedom, physician Samuel A. Cartwright “discovered” the diagnosis drapetomania. Treatment for drapetomania involved whipping slaves as a preventative measure (Bankole, 1998). Even many of Cartwright’s contemporaries mocked what has come to be seen as a quintessential example of pseudoscientific racism (Bankole). Superstitions about medicine are ubiquitous even in the modern era if we consider that when the HIV virus emerged in the early 1980s, some thought — and still do — that people afflicted with this diagnosis have been punished by God. The belief that clients with mental health problems are possessed by evil spirits persists throughout the world today.
Inhumane “treatments” based on bad science were widespread in the U.S. during the 20th century, including involuntary sterilization (Largent, 2008) and lobotomy (El-Hai, 2005). In the U.S., conversion therapy is a mental health treatment that is still practiced. Aimed at changing a client’s sexual orientation from homosexual to heterosexual, conversion therapy is based on the anti-scientific view that sexual orientation is a choice. Legislation prohibits conversion therapy for minors in some U.S. states and has been rejected by professional mental health associations, yet it continues to be offered by clinicians where permitted by law. Critics have suggested that these treatments are better described as forms of psychiatric coercion and control (Glasser, 2004; Szasz, 2010).
Since psychiatric drugs were developed by the middle of the 20th century, they have become the most common form of psychiatric treatment and are also considered by some to be based on dubious scientific findings. I am not anti-psychiatry nor am I against psychiatric drugs. But I recognize that like any treatment, there are both risks and benefits to psychiatric drugs. I also understand that the DSM-5 diagnoses do not reflect object brain diseases which, in turn, inform the prescribing of drugs that can ameliorate such brain diseases. Pharmacology is to a large extent a trial and error process.
A medical diagnosis is considered valid when it is an accurate understanding of a client’s condition or disorder. However, the diagnoses set forth in the DSM-5 are not objective medical conditions similar to how diabetes and heart disease are independently verifiable through objective tests. Instead, they amount to symptoms and behaviors reported by clients and/or others which, in turn, are observed by the clinician. What is and is not observed by a clinician depends on many factors, especially the clinician’s tendency to look for, find, and interpret information so that it confirms their preconceptions, or confirmation bias about so-called psychopathology.
For example, the DSM-5 diagnosis of Adjustment Disorder includes the criterion that symptoms or behaviors are out of proportion to the intensity of a stressor relative to external and cultural factors (APA, 2013, p. 286). But how, exactly, does a clinician make this determination? Unlike objective tests (e.g., lab values) which can be independently observed and measured, the decision-making process when formulating DSM-5 diagnoses are largely subjective. It could be argued, then, that any given DSM-5 diagnosis says at much, if not more, about the clinician than the client being diagnosed.
A careful review of the DSM-5 suggests it does not hold that its diagnoses are valid (APA, 2013, p. 5). The DSM-5 has stated that the APA recognizes past science did not produce valid diagnoses for mental disorders (APA, p. 5). Although the DSM-5 acknowledges “real” and “durable” progress has been made in the past two decades as a result of scientific advances (APA, p. 5), it also states that a diagnosis does not set forth implications about the etiology or causes of a mental disorder (APA, p. 25). However, clinicians may still think of the DSM-5 as a valid diagnostic system for various reasons, including because it corresponds to the medical model which is based on a scientific view. Consider that the DSM-5 was developed by the American Psychiatric Association and that the physicians who make up this organization hold authoritative knowledge which, although a poor basis to solely justify truth claims, creates a medical mystique for many clinicians and the public in general. Confusions and seeming contradictions arise because the DSM-5 is a descriptive nosology (excluding substance-induced and medically caused conditions), yet it claims to be based on the medical model.
Field trials have also showed that the DSM-5 has poor reliability (Frances, 2013; Vanheule et al., 2014). Reliability refers to the extent that the same or compatible results will be derived in different clinical experiments. For the DSM-5 this means that the same diagnosis will be formulated by different clinicians based on the same clinical presentations. Lack of reliability for the DSM is not new.
Williams et al. (1992) conducted an extensive reliability study of the DSM-III-R with pairs of trained clinicians interviewing 600 clients to determine if they agreed on a diagnosis (classification, not subtype). Inter-rater reliability ranged from only .68 to .72 for Axis I disorders. It has been also been noted that in field tests for the DSM-5, the APA has arbitrarily changed its definitions of what it considers acceptable reliability. (Frances, 2013).
Based on the lack of scientific validity and reliability for the DSM-5, the National Institute of Mental Health (NIMH) largely abandoned the DSM for research purposes in 2013 because it held it does not lead to useful research. Instead, it reported new Research Domain Criteria for studying mental disorders and is now studying how the brain and its trillions of synaptic connections work. It follows that until the objective causes of brain disorders are discovered through brain science, which is unlikely to happen in the foreseeable future, many, if not most, of the DSM-5 diagnoses should be banished.
The DSM-5 does not inform treatment
In recent years evidence-based practice has emerged as a significant trend in psychiatry and the field of mental health. Evidence-based practice refers to those treatments that have been shown through randomized clinical trials to be effective over placebo or no treatment groups. This trend reflects an effort to develop rigorous research methodologies aimed at evaluating best practices and establishing improved accountability. However, a significant limitation of evidence-based practice is that it tends to match diagnoses to interventions in a decontextualized manner (Duncan, 2014). Effectiveness in therapy depends less on the type of treatment selected and more on the resources and strengths of the client and the therapeutic alliance (Duncan, Miller, Wampold, & Hubble, 2009). Research has found that client factors and therapeutic alliance factors account for 40% and 30% of improvement in therapy, respectively — the highest percentage among common change factors. Model factors, on the other hand, only account for 15% of improvement.
There are at least two reasons the field may be obsessed with developing new therapy models despite the finding that the models used by therapists play the smallest role among common factors in bringing about change. The first reason may be that the field of mental health continues to search for the ultimate therapy model — the panacea, if you will — for all ills. It has been over a century since the first therapy models were developed and the field has still not come close to discovering a cure for mental illness. Yet new approaches are continually being developed in an effort to find The Holy Grail. In recent years, mindfulness has been lauded as the new panacea despite recognitions of its limitations. Granted, neuroscience may fundamentally transform mental health in the future in ways that are unimaginable today (Martin, Guterman, & Kopp. 2012). But for now, such solutions are out of reach.
A second reason why some therapists may remain loyal to particular therapy models is political. Powerful institutions such as universities and professional associations hold privileged knowledge and impose the prevailing therapy models in our field. The entire profession, including clinicians, educators, researchers, and students are indoctrinated to the prevailing models, and the pressure is on to align with those models to advance in academic programs, obtain licensure and certification, and attain employment. Doing so creates a conflict of interest for some therapists by impeding curiosity, learning, and growth. This conflict of interest is also evident in the relationship many therapists have to the DSM-5. The fight for parity with psychiatrists has been a double-edged sword. On one hand, the struggle for parity has led to increased insurance reimbursements. Often all that is needed is to obtain insurance coverage is a DSM-5 diagnosis and matching treatment plan. But questions arise if this is the best way to serve our clients. The trend toward medicalization and evidence-based practice has often sacrificed the integrity of core values of many clinicians.
The DSM-5 engenders stigma
DSM-5 diagnoses are limited descriptions which highlight deficits, weaknesses, and problems and overlook capabilities, resources, and strengths. In effect, a DSM-5 diagnosis is a one-sided story that does not present the full picture of the client’s life. Prejudice toward individuals with mental health issues still pervades society as people with such conditions are often viewed as undesirable, incapable of participating in meaningful relationships, or unable to hold positions of authority. Some clients report feeling empowered by discovering they have a DSM-5 diagnosis, and a thorough assessment of the client’s sensibility toward their diagnosis. For many clients, however, the stigma of a diagnosis may be oppressive and in some instances worse than the mental illness.
Stigma affects all sectors of society. For example, the U.S. military has developed programs to combat stigma associated with mental illness and improve quality of care for soldiers who are increasingly dealing with posttraumatic stress disorder and other conditions. Organizations such as the National Alliance on Mental Illness and the National Institute of Mental Health have led efforts to overcome the stigma of mental illness. Anti-stigma campaigns have also been developed to fight discrimination against people with mental illness. Yet stigma of mental illness persists and is often reinforced by the pathologizing focus of the DSM-5.
In a cautionary statement, the DSM-IV-TR stated that any given diagnosis is intended to diagnose conditions, not individuals (APA, 2000, p. x). This salient warning is curiously absent from the DSM-5. Despite this caution to not label clients with DSM diagnoses, this hardly prevents the public, the media, and clinicians from doing so. Therapists serve an important function in combating this and other types of stigma because of their relevant education, training and experience, as well as their direct contact with clients, educators, health care providers and policymakers. A major part of the effort to address stigma involves calling attention to the underlying attitudes that reinforce stigma and then working to change those attitudes.
Sometimes therapists also buy into stereotypes about mental illness and the DSM-5 by using the same inaccurate language and biases as social institutions, the media and the general public. For example, some therapists may refer to a client as “a borderline” rather than a client who has been diagnosed with borderline personality disorder. Sometimes I refer to a client as “a schizophrenic” merely for economy of expression. I admit, however, I also sometimes slip back to my old ways and blur the distinction between the client and his or her disorder. When this happens, I remind myself that I am labeling the client and the disadvantages of doing so.
It may be cumbersome to change how we talk about our clients. But this is an ethical imperative if we are serious about changing how we think about our clients. Becoming aware of expressions of stigma is an important role for therapists, and learning how to challenge these expressions is a critical skill. Therapists can reinforce their own anti-stigmatizing efforts by helping colleagues, friends, and family members identify instances when they use diagnostic terms as insults or pejoratives. In addition, therapists can involve themselves in the efforts of organizations to fight stigma and call attention to inaccurate portrayals of people with mental illness in the media.
Perhaps most important, therapists have unique opportunities to help fight stigma in their clinical relationships with clients. Therapists can promote social justice in the fight against stigma one case at a time. This can be achieved by viewing clients as individuals with their own unique potentials rather than limited by a DSM-5 diagnosis. Once a diagnosis is given, it is common for clinicians to apply the halo effect: a cognitive bias in which an observer’s overall impression of an individual tends to influence the observer’s thoughts, feelings, and behaviors about that individual’s character. Ascribing any label to a person, such as “angry,” “stupid,” and “hopeless” may create a frame that blocks any opportunity for the individual to show an alternative. Once a client has been diagnosed with schizophrenia, it is common for therapists and others to attribute all or much of the client’s behavior to the diagnosis.
Stigma of mental illness has also been perpetuated by the DSM-5 due to its lack of emphasis on the critical role of culture and diversity in understanding human problems and the phenomena otherwise associated with mental illness. Multiculturalism addresses various domains, including race, ethnicity, gender, sexual orientation, disability, socioeconomic status, age, spirituality, religion, and family structure. The Cultural Formulation Interview is a new tool added to the DSM-5, but it has only been proposed for “further study” (APA, 2013, pp. 750–757) and it does not replace training and experience in multicultural competencies (Lee, 2013).
The DSM-5 rightly cautions that the boundaries between normality and abnormality differ across cultures (APA, p. 14). This acknowledgment suggests that mental illness is largely culture-bound. But this raises fundamental questions about the nature of mental illness itself. As Gergen (1985) has noted, our understandings of the world, including concepts of mental illness, are not driven by nature but, rather, by culture and historization:
In each case, constructions of persons or relations have undergone significant change across time. In certain periods, childhood was not considered a specialized period of development, romantic and maternal love were not components of human makeup, and the self was not viewed as isolated and autonomous. Such changes in conception do not appear to reflect alterations in the objects or entities of concern but seem lodged in historically contingent factors. (p. 267)
The pathologizing focus of the DSM-5 also distracts from many if not most of the problems of which most clients seek therapy. Research has found that the majority of problems that clients seek therapy for are relational in nature (Gottman, 1999). In addition to relationship problems, clients seek therapy for many others situational issues other than mental disorders. But a client will rarely be covered for these unless they are given an individual DSM-5 diagnosis — a mental disorder. To be covered by insurance, you have to be sick to receive mental health care.
More-of-the-same from the DSM
Since its first edition was published in 1952, the DSM has revised the diagnostic manual eight times (APA, 1968, 1974a, 1974b, 1975, 1987, 1994, 2000, 2013). The first edition of the DSM contained 106 disorders (APA, 1952). The second edition, published in 1968, was expanded to 150 disorders (APA, 1968). By 1994, the fourth edition totaled over 300 disorders (APA, 1994). The DSM-5 added 15 new disorders (APA, 2013). And new disorders are being “discovered” all the time. By decreasing thresholds, the DSM-5 criteria for many disorders have become so wide that clinicians are diagnosing and treating more people than ever with mild or even nonexistent conditions (Frances, 2013; Rogers, 2013). Drug companies have contributed to the problem by targeting DSM diagnoses when selling psychotropic drugs.
While efforts to improve the DSM appear to reflect the spirit of scientific inquiry, closer examination indicates most, not all, of the changes amount to a more-of-the-same solution for the problems I have thus far identified regarding this diagnostic system. Revisions of the DSM are sometimes made due to political and social pressure to remove archaic diagnoses.
Homosexuality is a dramatic example of a DSM diagnosis that reflected a culture’s misunderstanding rather than a mental illness. In 1952 the DSM-I listed homosexuality as a sociopathic personality disturbance (APA, 1952). Presented with data from researchers as well as pressure from gay activists and dissent within the APA, homosexuality was removed as a category of disorder in the seventh printing of the DSM-II (APA, 1974b) and replaced with “sexual orientation disturbance.” Did an entire group of people become “cured” of homosexuality with a single vote from the APA? If not, then where was the so-called mental disorder of homosexuality located? One can only reasonably conclude that it was located in the hands of those who created the diagnostic category in the first place.
Rather than provide an exhaustive review of changes in the DSM-5, I provide a brief review of the shift from a multiaxial system to a nonaxial system because unlike other changes, this significantly impacts use of the manual in daily practice. What was separated on Axis I, II, and III in the DSM-IV-TR are now combined in the DSM-5 because no distinction is drawn for purposes of listing diagnoses between medical disorders and mental health conditions. Whereas psychosocial problems were listed on Axis IV in the DSM-IV-R, these are now listed as V codes or 900 codes. The Global Assessment of Functioning (GAF), which was used on Axis V of the DSM-IV-TR, was removed due to a lack of conceptual clarity, lack of clinical utility, and poor reliability (APA, 2013, p. 16). The DSM-5 has introduced the World Health Organization Disability Assessment Schedule 2.0 (WHODAS 2.0; WHO, 2010) for “further study.” Information on the validity or reliability of the WHODAS 2.0 is not available in the DSM-5. Clinicians are also responsible for using the WHODAS 2.0, like any psychometric test, in accordance with their code of ethics and legal statutes.
A potential advantage of the nonaxial system is that it may reinforce the holistic perspective held by some therapists by no longer distinguishing between medical, emotional, and psychosocial factors. The nonaxial system also resolves some limitations and drawbacks of the multiaxial system. For example, it has been suggested that the multiaxial system had limited clinical utility (Paris, 2013). Some clients may also feel stigmatized by having an Axis II disorder. The nonaxial system may also present new challenges for some clinicians. In particular, some therapists may find it difficult to code in the less structured, nonaxial format of the DSM-5 each of the diagnoses and psychosocial factors that were coded on Axis II, III, and IV in the DSM-IV-TR.
The DSM-5 has made numerous diagnostic changes, structural modifications, and has revised its organization. Multiple online enhancements for the DSM-5 are provided at www.psychiatry.org/dsm5. I suggest therapists access this online resource due to anticipated changes in the DSM-5. To date, however, the DSM-5 has significant limitations, including that its diagnostic system lacks validity and reliability, does not inform treatment, and tends to engender stigma.
American Psychiatric Association. (1968). Diagnostic and statistical manual of mental disorders. Washington, D.C.: Author.
American Psychiatric Association. (1974a). Diagnostic and statistical manual of mental disorders (2nd ed.). Washington, D.C.: Author.
American Psychiatric Association. (1974b). Diagnostic and statistical manual of mental disorders (2nd ed., seventh printing). Washington, D.C.: Author.
American Psychiatric Association. (1975). Diagnostic and statistical manual of mental disorders (3rd ed.). Washington, D.C.: Author.
American Psychiatric Association. (1987). Diagnostic and statistical manual of mental disorders (3rd ed., revised). Washington, D.C.: Author.
American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, D.C.: Author.
American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (4th ed., Text Revision). Washington, D.C.: Author.
American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). Washington, D.C.: Author.
Bankole, K. (1998). Slavery and medicine: Enslavement and medical practices in antebellum Louisiana. London, UK: Routledge.
Duncan, B.L. (2014). On becoming a better therapist: Evidence-based practice one client at a time (2nd ed.). Washington, D.C.: American Psychological Association.
Duncan, B.L., Miller, S.D., Wampold, B.E., & Hubble, M.A. (2009). The heart and soul of change: Delivering what works in therapy (2nd ed.). Washington, D.C.: American Psychological Association.
El-Hai, J. (2005). The lobotomist: A maverick medical genius and his tragic quest to rid the world of mental illness. Hoboken, NJ: Wiley & Sons.
Frances, A. (2013). Saving normal: An Insider’s Revolt against Out-of-Control Psychiatric Diagnosis, DSM-5, Big Pharma, and the Medicalization of Ordinary Life. New York: William Morrow.
Glasser, W. (2004). Warning: Psychiatry can be hazardous to your mental health. New York: Harper.
Gottman, J.M. (1999). The marriage clinic: A scientifically based martial therapy. New York: W.W. Norton.
Guterman, J.T. (2013). Mastering the art of solution-focused counseling (2nd ed.). Alexandria, VA: American Counseling Association.
Kane, J.M, Robinson, D.G., Schooler, N.R., Mueser, K.T., Penn, D.L, Rosenheck, R.A., Addington, J., Brunette, M.F., Correll, C.U., Estroff, S.C., Marcy, P., Robinson, J., Meyer-Kalos, P.S., Gottlieb, J.S., Glynn, S.M., Lynde, D.W., Pipes, R., Kurian, B.T., Miller, A.L., Azrin, S.T., Goldstein, A.B., Severe, J.B., Lin, H., Sint, K.J., John, M.J., & Heinssen, R.K. (2015). Comprehensive versus usual community care for first-episode psychosis: 2-Year Outcomes from the NIMH RAISE early treatment program. The American Journal of Psychiatry, 174, 362- 372.
Largent, M.A. (2008). Breeding contempt: The history of coerced sterilization in the United States. New Brunswick, NJ: Rutgers Press.
Lee, C.C. (2013). The cross-cultural encounter: Meeting the challenge of culturally competent counseling. In C.C. Lee (Ed.), Multicultural issues in counseling (4th ed.) (pp. 13–19). Alexandria, VA: American Counseling Association.
Martin, C.V., Guterman, J.T., & Kopp, D.M. (2012). Extending the dialogue about s cience and humanities: A reply to Hansen. Journal of Humanistic Counseling, 51, 161–163.
Rogers, W.A. (2013). Avoiding the trap of overtreatment. Medical education, 48, 12–14.
Szasz, T.S. (2010). The myth of mental illness: Foundations of a theory of personal conduct. New York: Harper.
Vanheule, S., Desmet, M., Meganck, R., Inslegers, R., Willemsen, J., De Schryver, M., & Devisch, I. (2014). Reliability in psychiatric diagnosis with the DSM: Old wine in new barrels. Psychotherapy and Psychosomatics, 83, 313–314.
Williams, J., Gibbon, M., First, M., Spitzer, R., Davies, M., Bores, J., Howes, M., Kane, J., Pope, H., Rounsaville, B., & Wittchen, H. (1992). The structured clinical interview for DSM-III-R (SCID) II: Multi-site test-retest reliability. Archives of General Psychiatry, 49, 630–636.