The DSM and Its Limitations

The DSM, published by the American Psychiatric Association, is more than just an academic enterprise in psychiatric nosology, the medical science of classifying disease. It is the authoritative manual for classifying and diagnosing mental illness and impacts everyone who seeks treatment for mental illness in the US.

Only DSM diagnoses are covered by private insurance companies and government assistance programs such as Medicare. So if you’re having an existential crisis, which is not a diagnosis listed in the DSM, you had better hope your doctor is creative enough to diagnose you with something like “generalized anxiety disorder” or “anxiety disorder-not otherwise specified,” or else you’ll be forced to pay out of pocket for treatment.

The DSM is also used by other institutions like the legal system. Despite the manual’s explicit warnings against reliance on its classifications for forensic and legal decision-making, it is routinely used to determine whether defendants are mentally fit to stand trial or are eligible for the death penalty.

While the manual is pervasive in US institutions, its accuracy in capturing mental illness is questionable. The DSM relies on an algorithmic structure for many diagnoses — to be diagnosed patients must exhibit a subset of symptoms from a much larger list of symptoms.

This leads to high rates of heterogeneity — people diagnosed with the same disorder often look symptomatically very different from one another. Take an illness like major depressive disorder: a positive diagnosis requires a patient to have five out of nine listed symptoms, so two different patients diagnosed with depression may have only one symptom in common.

This variability highlights the thornier problem that diagnostic categories like depression or schizophrenia are largely socially-derived categories, not uniform, naturally occurring diseases as presumed by the medical model.

There are also remarkable similarities across DSM diagnoses. For example, anhedonia — the reduced ability to feel pleasure — is part of the diagnostic criteria for depression, schizophrenia, and substance abuse disorders. If you are diagnosed with one disorder, chances are you also meet the criteria for at least one other disorder. So patients diagnosed with a personality disorder (e.g., borderline personality disorder; narcissistic personality disorder) usually meet the criteria for approximately two more personality disorders.

Researchers argue that the high rates of multiple diagnoses suggest that instead of comprising distinct illnesses, the DSM diagnostic system may be “attaching multiple labels to differing manifestations of the same underlying condition.”

In the face of criticism, the DSM has been revised numerous times over the last six decades, but serious issues of validity remain, and many experts question whether the manual actually describes “diseases” akin to physical illnesses at all.

Not Medical Enough

Most of the field acknowledges the problems of the manual. But rather than conclude that the failures of the DSM show how limited the medical model approach to mental illness is, the dominant argument today is that the DSM is not medical enough.

The director of the NIMH, Thomas Insel, a neuroscientist known largely for his work on the social behaviors of rodents, has led the charge against the DSM, claiming that the manual’s reliance on subjective experience rather than objective measures is deeply flawed: “Unlike our definitions of ischemic heart disease, lymphoma, or AIDS, the DSM diagnoses are based on a consensus about clusters of clinical symptoms, not any objective laboratory measure.”

DSM diagnoses use patients’ subjective reports of their thoughts, feelings, and behaviors. A major depressive episode is diagnosed by asking patients whether they feel down, hopeless, or worthless (e.g., “Over the past two weeks, have you felt guilty or worthless nearly every day?”).

Insel compares this use of subjective experience to “creating diagnostic systems based on the nature of chest pain or the quality of fever,” and argues that subjective experience is too fuzzy and unscientific. Instead, he insists, the study of mental illness should rely on objective, observable facts. Oncologists don’t diagnose cancer using phenomenological symptoms and psychiatrists diagnosing depression shouldn’t either.

With Insel at the helm, the NIMH has rolled out a new research funding program, called the Research Domain Criteria (RDoC), to promote diagnosis and treatment based on objective, biological markers. RDoC aims to “bring the power of modern research approaches in genetics, neuroscience, and behavioral science to the problems of mental illness.”

Critically, Insel believes that the NIMH should start from scratch, and develop a new way of thinking about mental illness that eschews DSM categories in favor of constructs informed heavily by the biological sciences.

This is a major shift, and has caused quite a stir in the mental health research community. Researchers at SUNY Stonybrook say, “Our field is embedded in a larger system in which, for many researchers, obtaining grant support is the sine qua non of employment or tenure.” Because their careers rely on it, mental health researchers will adhere to the RDoC framework for funding purposes, which ultimately “runs the risk of again forsaking a more comprehensive conceptual model for a reductionist model.”

Allen Frances, a psychiatrist at Duke University, also criticizes RDoC’s reductionist account of mental illness: “It is a dangerous myth to assume that patients who meet criteria for ‘schizophrenia’ suffer only from a brain disease. Contextual forces play a large role in the onset of schizophrenia. A supportive environment, a decent place to live, and therapeutically encouraged engagement with school, work, and social activities are now, and always will be, absolute essentials.”

Critics of RDoC argue that the program privileges biological constructs over environmental, cultural, or “psychosocial” ones. This is certainly true, and it isn’t an accident.

Bringing Back Big Pharma

Psychopharmacology is in crisis. The data are in, and it is clear that a massive experiment has failed: despite decades of research and billions of dollars invested, not a single mechanistically novel drug has reached the psychiatric market in more than 30 years,” writes H.C. Fibiger, psychiatrist and Senior Vice President at Biovail Labs.

He is not exaggerating. In contrast to the boom in psychiatric drug development of the 1980s and 90s, the industry has come to a standstill, now largely repurposing old drugs for new illnesses.

Major field studies with thousands of patients conducted during the early 2000s on depression, bipolar disorder, and schizophrenia have found variable and lackluster effects of psychiatric medication for treating mental illness symptoms.

Steven Hyman, a former director of the NIMH, argues that “people suffering with the depressed phase of bipolar disorder often continue to experience severe symptoms even when they take multiple medications with serious side effects” and that for “the core social deficits of autism and the cognitive impairments of schizophrenia, there simply are no effective treatments.”

Remission rates for depression are around 30 percent, and people suffering from mild or moderate depression do no better on SSRI antidepressants than placebo pills. Echoing Hyman, Insel concludes that these disorders are “as much of a public health challenge now as they were a century ago.”

Although psychiatric drugs are one of the most profitable for pharmaceutical companies — one in five Americans take at least one psychiatric drug — the lack of promising new psychopharmacological treatments has led the industry to pull back their investments in the field. Forbes reports that “psychiatric drug development has become a dead end. GlaxoSmithKline, Novartis, and AstraZeneca have stopped trying to invent new psychiatric drugs. Pfizer, Merck, and Sanofi have de-emphasized them.”

This presents a major problem for psychiatric researchers because, after the NIMH, pharmaceutical companies are the largest source for funding.

Psychiatric researchers are acutely aware of this trend. Fibiger concludes, “In order to recapture industry’s investment in psychiatric drug development, major changes in psychiatry will need to take place,” and that “if we are successful in making the necessary changes, given the enormous level of unmet need left by existing psychiatric drugs, there is no doubt that industry will return and reinvest.”

Insel concurs: “The business case for psychiatric drug development is not promising” and that “in order to develop a long-term strategy for further development by the private sector, a new paradigm is needed.”

This is where RDoC comes in. By treating the DSM as the central cause of psychiatry’s failures — and eliminating it — the program aims to spur new research on the biology of mental disorders. For example, very few causal biological mechanisms of mental disorders have been identified — we still know little about what processes inside the body lead to mental illness.

This is a problem for drug companies because no one knows what biomarker their drug should target: “Pharmaceutical companies often compete on the same poorly validated targets, wasting time and resources . . . this is unsustainable as a business model,” says Lenox, professor of pharmacology and clinical neuroscience at the University of New England College of Osteopathic Medicine. Biomarkers will give researchers something for their drugs to target and make drug development less risky and more profitable.

The NIMH blames the DSM for the lack of identified biomarkers and argues that DSM’s “subjective” categories have led researchers astray. By focusing on “objective” measures, mechanisms, and biomarkers, the NIMH hopes to regain interest and, more importantly, funding from the pharmaceutical industry. To “jumpstart” this new understanding of mental illness, the RDoC is now requiring that all NIMH funding applications focus on a limited set of neuroscientific systems.

But the neuroscience systems proposed by the NIMH are grossly inadequate for capturing the complexities of mental illnesses. Scientists who have traditionally studied major depressive disorder, post-traumatic stress disorder, or bipolar disorder, are now scrambling to retrofit awkward neuroscientific constructs provided by the RDoC website, such as “negative valence systems” and “arousal/modulary systems” to mental illness in order to attract NIMH funding.

Although the NIMH’s RDoC funding program claims to work towards a understanding of mental illnesses free of the assumptions of the DSM, the program maintains the central one — that mental illnesses are “brain diseases” divorced from subjective experience and society’s values regarding what is and isn’t distressing, inappropriate, repugnant, and morally reprehensible.

Beyond Biology and the Medical Model

The analogy between mental and physical illness is deeply flawed. While it is certainly true that the subjective experience of cancer is irrelevant in its diagnosis, any objective markers found in depression are meaningless if you do not feel depressed. Scientists at Northwestern University are developing a blood test for depression. What if it comes back positive, but you feel fine? Alternatively, what if you are suicidal, but the test comes back negative? It would be odd to conclude that you are not depressed.

Subjective experience is central to mental illness. So while the modest effects of psychopharmacological drugs discovered in the past decade could indicate that better biological research is needed, they could also mean that the key to understanding and addressing mental illnesses lies elsewhere. Obviously biology matters, but the field needs to study mental illness on its own terms, rather than uncritically emulate the logic of the so-called hard sciences for the legitimacy and funding it brings.

Questioning the medical model doesn’t mean mental illness is a myth. Far from it. People’s distress is very real, and the field can no longer ignore the problems in its current understanding of mental illness. But the RDoC’s exclusive focus on biomarkers to draw in pharmaceutical investment will not address these shortcomings.

Instead, the RDoC will further strengthen the field’s economic and ideological ties to the profit-driven private sector. Its focus on “cutting-edge” neuroscience biomarkers will tether psychologists ever tighter to psychopharmacology and the prerogatives of big pharmaceutical companies.

The RDoC also hurts those suffering the most by ignoring the connection between mental illness and societal structures. As Frances says, “NIMH has had its attention so distracted by glorious dreams of a future research revolution that it has completely lost touch with the desperate suffering of schizophrenic patients in the present. It pays no attention to, and takes no responsibility for, the mess that is US mental health care.”

Over the past fifty years, the number of public psychiatric hospital beds in the US has decreased from 340 to 17 beds per 100,000 people, leaving poor patients nowhere to turn for adequate care. Many become homeless or end up in the prison system for petty crimes that could have been easily prevented if they had the proper community care. “These patients are suffering greatly not so much for lack of knowledge on how to care for them, but because of a lack of attention and inadequate resources,” says Frances. RDoC is part of a growing trend in which state and community services are short-changed in favor of a psychopharmacological model for treatment.

However, despite its deep problems, there is a potential silver lining in RDoC’s start-from-scratch approach. The program presents a unique opening for the field to be explicitly critical of its past assumptions and learn from its failures.

It would be a powerful move for psychiatry and psychology to take advantage of this crisis and not only move away from DSM diagnoses, but also the medical model itself and the overwhelming influence of pharmaceutical companies, toward an approach that incorporates subjectivity and recognizes the social and economic basis of mental illness. That would be a real change.