Why Psychiatric Diagnosis Matters

Despite recent claims, diagnosis in psychiatry is scientifically meaningful.

Posted Aug 10, 2019

This article is co-authored by Ronald W. Pies, M.D., clinical professor of psychiatry at Tufts University Medical School and professor emeritus of psychiatry at the State University of New York Upstate Medical University, and Mark L. Ruffalo, D.Psa., psychoanalyst and instructor of medical education (psychiatry) at the University of Central Florida College of Medicine.

A college student is brought to the psychiatric emergency room due to bizarre behavior, disorganized thinking, and extreme emotional distress. His college roommate notified the dorm supervisor that he has been acting oddly over the past few weeks, talking to himself at all hours of the night, and expressing a fear that his electronic devices have been "tapped." Upon evaluation, the patient states that the KGB has implanted a radioactive thought monitoring device in his brain and that Russian agents have been following him for weeks. In fact, he believes there are spies right outside of the hospital room. He asks the interviewer if he is of Russian descent.

While this vignette may sound like a scene out of a movie, it is the reality in emergency departments around the world on a daily basis. The patient is suffering from acute psychosis and is probably in need of inpatient psychiatric hospitalization. Adequate treatment depends on a thorough investigation of the underlying cause of his psychosis and will vary depending on this etiological determination. Is he suffering from schizophrenia, delusional disorder, bipolar disorder, a substance-induced disorder, or another medical or psychiatric disease?

Mark L. Ruffalo, D.Psa.
The Diagnostic and Statistical Manual of Mental Disorders (5th ed.)
Source: Mark L. Ruffalo, D.Psa.

Despite the obvious importance of answering these diagnostic questions, authors of a recent study published in the journal Psychiatry Research contend that psychiatric diagnosis is "scientifically meaningless." The study, co-authored by psychologist Peter Kinderman and his colleagues at the University of Liverpool in England, was widely publicized and heralded as proof positive that psychiatric diagnosis (and psychiatry more generally) is hogwash.

In an interview published in Science Daily Kinderman went on to imply that psychiatric disorders—specifically, the categories described in the DSM-5—are not "real illnesses;" and that diagnosis in psychiatry is illusory. The paper's lead author, Dr. Kate Allsopp, claimed that psychiatric diagnosis serves only to stigmatize patients, and discounts the patient's life experiences and trauma.

As a psychiatrist who has spent much of his career considering the question of psychiatric diagnosis and a psychoanalyst who teaches in a psychiatry residency program, we feel obligated to address these assertions, since they have profound consequences for both patients and clinicians.

In essence, we believe that Allsopp and his colleagues do not provide a characterization of the DSM-5, so much as a caricature, and thereby misunderstand the manual's purpose and well-acknowledged limitations. This is not to say that we unequivocally or uncritically endorse the DSM's classification scheme; indeed, one of us (R.W.P.) has advocated different kinds of diagnostic schemes for clinical and research purposes, such as prototypical and categorical models (see Phillips et al., 2012).

Nevertheless, the claims of Allsopp et al.—particularly in the Science Daily interview—go beyond a reasoned critique of DSM-5 and psychiatric diagnosis, and border on what we would consider an "anti-psychiatry" polemic.

Below, we outline the basic claims made by Allsopp et al. and provide our rejoinders.

Claim #1: Diagnostic heterogeneity in DSM-5 is problematic.

While we would concede that, in some instances, "Diagnostic heterogeneity is problematic for both research and clinical practice" (Allsopp, Read, Corcoran, & Kinderman, 2019, p. 15), we would point out that diagnostic heterogeneity within a particular category does not, by itself, mean that the predictive validity of the particular condition is vitiated. To oversimplify, predictive validity in the psychiatric context refers to the ability of a diagnosis to maintain stability over time, and to predict rates of relapse, recovery, and response to treatment (Kendell & Jablensky, 2003).

For example, schizophrenia was recognized as "heterogeneous" as early as 1911, when Eugen Bleuler described "the schizophrenias," and is widely regarded multiform and heterogeneous to this day. Yet the general DSM construct of schizophrenia—allowing for slight differences in criteria from DSM-3-R to DSM-5—has considerable predictive validity (Mason, Harrison, Croudace, Glazebrook, & Medley, 1997).

Furthermore, as Dr. Kenneth F. Schaffner (2016) notes,

The notion that the entities that are fundamental in a scientific area need to be discrete and separable is an idea that works well in some sciences such as physics and chemistry. But these types of entities are rarely found in biology, where more 'polytypic' or 'polythetic' concepts reflect the variation in the entities that are fundamental in that science. And medicine and psychiatry are similarly affected by variation ... [yet] from a symptom aggregational view, the DSM-5 criteria [for schizophrenia] work fairly well in providing a diagnosis and a therapeutic plan (p. 40).

Claim #2: The DSM-5 neglects the role of trauma.

Allsopp et al. (2019) state,

By making reference to trauma or stressors only in one dedicated chapter, the DSM-5 implies that other diagnostic categories are unrelated to trauma … The consideration of social, psychological, or other adversities within diagnoses is therefore minimized; symptoms are constructed as anomalous or disordered, rather than potentially understandable in relation to a person's life experiences (p. 20).

This claim contains several misunderstandings. First, while the DSM-5 may not cite trauma per se as having an association with the majority of mental disorders, it does cite psychological and/or environmental stressors or precipitants as contributing to a number of diagnoses—if not as causal factors, then as possible risk factors or precipitants.

For example, the DSM-5 construct of Conversion Disorder contains two specifiers: "with psychological stressor" and "without psychological stressor." The accompanying text explicitly notes that "onset may be associated with stress or trauma, either psychological or physical in nature" (p. 319). The text goes on to state that "Assessment for stress and trauma is important, though absence of these should not preclude the diagnosis" (p. 320). Yet the authors do not take note of this in their table 7. The authors omit discussion of "childhood diagnoses" in which category they would put "Reactive Attachment Disorder." However, the DSM-5 classifies this within the Trauma- and Stressor-Related Disorders, and while "trauma" per se is not listed as an associated feature, "severe neglect" is.

Moreover, there is an inherent contradiction in the authors' position. They argue that the DSM-5's "limited reference to trauma implies that it affects only a limited number of diagnoses, despite increasing evidence to the contrary" (Allsopp et al., 2019, p. 21). But if the diagnoses themselves are invalid—and merely arbitrary labels without scientific foundation, how can evidence point to the role of trauma in these diagnoses?

The authors can't have it both ways. Either the diagnostic categories are more than arbitrary labels and really do have trauma-based antecedents (as the authors want to claim); or they are mere labels with no ontological import, in which case it is nonsensical to claim that trauma plays any role in their genesis. The authors' position is akin to arguing that unicorns don't really exist, but that genetic mutations in horses are associated with unicorn development!

Finally, Allsopp et al. seem to confuse "understandable" with "non-disordered." The two are not the same. An elderly man with coronary artery disease who has a sudden heart attack when a grizzly bear appears before him in the woods might well have had an "understandable" reaction, but a myocardial infarction is still pathological (disordered). A major depressive episode following the death of a loved one may be "understandable" but may still represent a disordered mood state, with all its implications for poor outcome, morbidity and mortality, etc. (see Pies, 2012).

We acknowledge (as the authors rightly point out) problems with the DSM-5's construct of normality. For example, the authors note, with respect to PTSD, "...there is no information [in DSM-5] about how to identify at what point someone has a 'disordered' response as opposed to one that is 'normal'" (Allsopp et al., 2019, p. 17).

But the fuzzy border between "normal" and "abnormal" prevails throughout the entire field of medicine—even in supposedly "objective" fields like oncology and pathology. Furthermore, the same charge could be brought against the authors' use of the term "understandable." What is "understandable" may vary greatly from culture to culture, family to family, and person to person. Allsopp et al. provide no objective criteria for judging the issue of "understandability."

Claim #3: The DSM-5 implies that its diagnostic categories should be construed as "causing" the symptoms listed.

Allsopp et al. (2019) write: "…for the majority of the DSM-5 diagnostic categories, the criteria suggest to clinicians that these difficulties are caused by the disorder (and implicitly that these disorders are associated with brain function)…" (p. 20).

This argument is commonplace in anti-psychiatry circles. It claims, in effect, that "psychiatrists say that hallucinations, delusions, and thought disorder define schizophrenia. Then when you ask them what causes these problems, they reply, 'schizophrenia.' Psychiatry merely attributes causal powers to their diagnostic labels." In fact, nothing in the DSM asserts that the patient's emotional or behavioral difficulties are "caused" by the specific disorder; rather, the disorder is simply identified by means of the given criteria, just as in general medicine. For example, when we state the body mass index (BMI) criterion for obesity as a BMI >30, we are not asserting that the high BMI is "caused by" the disorder of obesity. We are simply saying that obesity is identified by—or defined by—this BMI.

Moreover—with the exception of the dementias and related conditions—nothing in the DSM attributes the principle psychiatric disorders to abnormal brain function. Indeed, only in a few instances does DSM-5 imply an "association" with abnormal brain function. For example, in the text related to schizophrenia, it is noted that "differences are evident in multiple brain regions between groups of healthy individuals and those with schizophrenia" (pp. 101-102). But there are no causal inferences drawn from this fact. Indeed, the Manual states, "a diagnosis does not carry any necessary implications regarding the etiology or causes of the individual's mental disorder." (American Psychiatric Association, 2013, p. 25, italics added).
That said, it would be passing strange if Allsopp et al. are denying an association with psychiatric disorders and "brain function." If so, they might explain how, in the absence of brain function, any kind of emotional disorder could exist—or indeed, how consciousness itself could exist.

Claim #4: The DSM-5 categories "limit exploration" of other factors contributing to the person's condition.

Thus, on p. 20, the authors state, "and [the diagnostic categories] may therefore limit exploration further than identification of the disorder." The authors cite no evidence that use of the DSM's categories "limit exploration" of other factors. In fact, the DSM-5 explicitly encourages further exploration of the patient's condition, based on—among other things—cultural factors (see the section on culture-bound conditions). Moreover, on page 25, the manual describes the "approach to clinical case formulation," and states:

The case formulation must involve a careful clinical history and concise summary of the social, psychological, and biological factors that may have contributed to developing a given mental disorder. Hence, it is not sufficient to simply check off the symptoms in the diagnostic criteria to make a mental disorder diagnosis … the ultimate goal of a clinical case formulation is to use the available contextual and diagnostic information in developing a comprehensive treatment plan that is informed by the individual's cultural and social context (APA, 2013, p. 19, italics added).

Claim #5: Categorical diagnosis may obscure "individual experiences of distress."

This claim is a kind of corollary to claim #4. The authors write, "Furthermore, by focusing on diagnostic categories, individual experiences of distress and specific causal pathways may be obscured" (Allsopp et al., 2019, p. 21). The authors present us with a classic false choice: We must choose between categorical diagnosis, on the one hand, and recognition of individual experiences, on the other. In truth, the two are complementary. And the clinician's case formulation, discussed above, is aimed precisely at discovering the "individual experiences" shaped by the patient's "cultural and social context."

This article is separated into two parts. To view Part 2, click here.