The Concept of Mental Illness, and Why the DSM Approach Is Wrong

CONCEPTS IN PSYCHIATRY

A discussion of psychopathology necessarily begins with the concept of pathology—what distinguishes it from normality. It entails examining what constitutes mental illness and, in turn, what constitutes mental health. This article assesses the concept of mental illness, and later we will discuss mental health.

Let’s begin with a discussion of what terms we wish to use and what they mean: illness, disease, disorder—they are not the same thing. We will provide some definitions without defending them in detail; we are aware that there are critiques that could be made, and alternative definitions. Another place, another time for all that debate. Here we will present some suggestions, based on the medical and scientific literature, a large literature in the history of medicine, as well as philosophy of medicine and psychiatry.

The modern concept of disease can be traced as follows: 3 centuries ago, the 17th century English physician and political revolutionary, Thomas Sydenham, taught that the concept of medical illness should be based on the Hippocratic tradition (long dropped by mainstream medicine for 2 millennia) of clinical symptoms and signs with a typical natural course—ie, a clinical syndrome. A century later, the 18th century Italian physician Giovanni Morgagni, the founder of modern anatomy and pathology, added to Sydenham by insisting that the clinical syndrome needed to correlate to a pathological finding in the body on autopsy. A century afterwards, after seeing tens of thousands of patients clinically and conducting thousands of autopsies, the 19th century Canadian physician William Osler popularized and established this “clinicopathological” tradition as the basis of diagnosis in scientific medicine.¹

So, the modern scientific approach to disease is to view it as a clinical syndrome based on pathology in the body. A clinical syndrome is not enough; a pathology in an organ is not enough—the 2 things have to exist together.

Immediately, we see one problem with modern psychiatric diagnosis based on the DSM system. It is nothing but clinical signs and symptoms. There is no pathology of the body included. Thus, it is like turning the clock back to 17th century revolutionary England; we are practicing psychiatry similarly to the practice of medicine in the 17th century.

Then there is the term illness. It is not exactly the same as disease. One can have a disease and not know it. One can think one has a disease, and not have it. The idea of illness reflects a person’s awareness of, and psychological reaction to, a bodily disease. If there is no bodily disease, there is no illness. If there is illness, there has to be a bodily disease. The 2 terms are not interchangeable, as often is done.

The term disorder is even less exact. It is, in fact, vague because it has no scientific meaning, which is why the DSM system insists on using it.² It is an English word, not a scientific term. It means, as the name implies, an absence of order. That is it. It does not entail a pathology of a bodily organ, nor even a specific clinical syndrome, nor psychological awareness or reaction to a bodily problem. It simply means “something is wrong.” It could be a bodily disease, or not; it could entail psychological awareness, or not.

In 1980, when the third edition of DSM (DSM-III) was created, which has been the core of future editions, the decision was made to use the word disorder for all DSM diagnoses. The prior second edition had used the word reaction for all diagnoses, based on the influence of a certain approach to interpreting psychiatry (based on the ideas of Johns Hopkins psychiatrist Adolf Meyer). The leaders of DSM-III wanted to remove any commitments to any specific theory in psychiatry, whether Meyer or Freud or biology or even standard medicine. Thus, the word reaction was banned, as was “complex,” “neurosis,” “disease,” and “illness.” The word disorder was chosen specifically because it was vague and meaningless. DSM itself is vague and meaningless, and it is intended to be so, although the profession has been in denial about this fact. Stated more subtly, the leaders of DSM-III defended their approach by saying it was atheoretical—which means it is meaningless. If you have no theory, then you have no idea what you are thinking or saying. Theories are not bad; false theories are bad.

In the last half century, defenders of DSM-III, IV, and 5 have come up with a rationale for their approach to defining their diagnoses under the rubric of mental disorders. We can summarize their basic concept as being based on the view that a disorder is present when there is a clinical syndrome which is associated with 1 or both of the following 2 features: “subjective distress” and/or “severe functional impairment.”

Mainstream modern psychiatry takes this approach, and most clinicians practice this way. Patients present to them with some kind of clinical symptoms—most commonly, anxiety or depression or inattention or executive dysfunction. The clinician then assesses whether patients are subjectively distressed; if they are unhappy or bothered by their symptoms, they meet the criterion. If the clinician assesses whether the symptoms are causing serious functional impairment, if there are notable problems in their life function, they meet the criterion. Bingo: DSM definitions are met and a mental disorder diagnosis is given.

Compare this process with modern scientific medicine, basically any other field of medicine except psychiatry: You go to the doctor with intermittent mild chest pain. It bothers you briefly for a few seconds, but then goes away; you are still able to work and go about your activities. The doctor does not say, “Well, you do not have much subjective distress, and certainly almost no impairment, thus I will not diagnose anything.” A week later, a massive heart attack gives the diagnosis. Almost all doctors would send you for an exercise stress test or coronary angiography, which might show an almost complete coronary artery blockage. Or perhaps not. Either way, in scientific medicine, you find out if a disease is present or not. And it does not matter one iota whether you have subjective distress or severe functional impairment. There is no functional impairment with hypertension, and no symptoms at all; there is no functional impairment and no symptoms with high cholesterol, but if ignored, a future heart attack will happen. Early cancer is not ignored because it has no symptoms and no effects on function. In other words, whereas it is possible (and common) to have an asymptomatic medical disease, it is impossible—by DSM criteria—to have an asymptomatic mental disorder.

Indeed, many psychology schools recognize that subjective distress is a poor index of underlying pathology. In existential psychology, the viewpoint is that subjective distress is a good thing; anxiety is not a symptom but an accomplishment, as Frankl put it.³ In phenomenological psychology, subjective experiences are viewed neutrally; they can reflect pathology or not for a range of reasons, unrelated to whether they feel distressing to the person; often people lack insight into their pathology.⁴ In social psychiatry, subjective distress is seen as a reflection of the imposition of society’s values on the person, but not as an inherent problem in the person himself.⁵ And of course in psychoanalytically-oriented psychology, patients with severe personality disturbance may lack distress altogether, while patients with neuroses suffer intensely despite relatively mild illness.⁶

Modern scientific medical diagnosis, in short, has nothing to do with subjective distress or functional impairment. Psychiatric diagnosis, in contrast, is limited to subjective distress and/or functional impairment. The 2 approaches are opposites.

This is why it is so false and in fact conceptually ignorant to say, as is often repeated, that the DSM approach is the application of the medical model to psychiatry. In fact, the DSM approach is antimedical and antiscientific.

This DSM approach has been used over decades to refuse to diagnose all kinds of potentially valid psychiatric diseases. For instance, the concept of a spectrum of manic-depressive illness, that would include mild manic symptoms including hypomania less than 4 days in duration (the current cut-off for which there is zero scientific evidence) has been rejected repeatedly on the grounds that those mild symptoms are not associated with subjective distress and/or functional impairment.

In contrast, the definition of attention-deficit/hyperactivity disorder (ADHD) has been broadened from its original meaning, which has the most scientific evidence, requiring onset in childhood usually around ages 7 to 9. Now it has been extended to age 12 for onset and diagnosis in adulthood has been allowed, based on the rampant presence of subjective distress and reported functional impairment, augmented by social media attention. Now fully one-half of psychiatric referrals in the primary care clinic where one of us works (SNG) are self-referrals from individuals who want to have the ADHD diagnosis and then receive stimulant and/or amphetamine medications.

A similar problem appears with the growing popularity of complex posttraumatic stress disorder (PTSD). The construct did not emerge from evidence of a distinct clinical entity but from attempts to capture subjective distress and impairment following chronic trauma. Its defining features—affect dysregulation, interpersonal problems, disturbances in self-perception—substantially overlap with borderline personality disorder, without clear evidence of a differentiable condition.⁷

In the unscientific and anti-medical approach of DSM to diagnosis, since subjective distress and functional impairment is central, we have a situation that is the opposite of the stigma of the past. Hyped by social media, many individuals now want to be diagnosed with a mental disorder (albeit this one, ADHD or complex PTSD, but not that one; they reject other diagnoses such as bipolar illness and borderline personality disorder). And the DSM system allows it.

We would argue that we can and should put aside something meaningless, like the concept of DSM mental disorder, and we should seek to catch up 3 centuries later with the rest of scientific medicine—if our pretenses to science are at all legitimate—and accept the concept of disease as explained here.

Readers might be thinking, if they accept this viewpoint theoretically, how it applies to modern psychiatry, though, given the perspective that we do not have any bodily pathology that can be diagnostically useful. That is a good question, long asked in psychopathology for over a century, with a range of answers. We will turn to those questions in the next article.

Dr Ghaemi is a professor of psychiatry at Tufts University School of Medicine and a lecturer on psychiatry at Harvard Medical School.

Dr Ruffalo is an assistant professor of psychiatry at the University of Central Florida College of Medicine in Orlando and adjunct assistant professor of psychiatry at Tufts University School of Medicine in Boston, Massachusetts.

References

1. Ghaemi SN. In the tradition of William Osler: a new biohumanistic model of psychiatry. Perspect Biol Med. 2023;66(4):520-534.

2. Decker HS. The Making of DSM-III: A Diagnostic Manual’s Conquest of American Psychiatry. Oxford University Press; 2013.

3. Frankl VE. Man’s Search for Meaning. Beacon Press; 1959.

4. Jaspers K. General Psychopathology. Johns Hopkins Press; 1997.

5. Sullivan HS. The Interpersonal Theory of Psychiatry. W.W. Norton & Company; 1953.

6. Akhtar S. Broken Structures: Severe Personality Disorders and Their Treatment. Jason Aronson; 1992.

7. Ruffalo ML, D’Agostino A. Why is it still so difficult to acknowledge the reality of borderline personality disorder? Harv Rev Psychiatry. In press.