Choosing our terms: The diagnostic words we use can be harmful


We are living in an era of increasing sensitivity to our diversity and the ways we interact, but also an era of growing resistance to change and accommodation. As clinicians, we hope to be among the sensitive and the progressive, open to improving our views and interactions. And as part of our respect for those we treat, we seek to speak clearly with them about our assessment of what is disrupting their lives and about their options.

Using the right words is crucial in that work. Well-chosen words can be heard and understood. Poorly chosen words can be confusing or off-putting; they may miscommunicate or be offensive. Maintaining the quality of clinician-patient communication requires special care, because one party is expert and the other may not be, and because only one party is identified as ill. Careful choice of words is also important among colleagues, who may not always mean the same things when using the same words.

Dr. Bruce M. Cohen, director of the Program for Neuropsychiatric Research at McLean Hospital, Belmont, Mass., and Robertson-Steele Professor of Psychiatry at Harvard Medical School, Boston Dr. Cohen

Dr. Bruce M. Cohen

In psychiatry, consumer knowledge and access are growing. There are effective standard treatments and promising new ones. But our terminology is often antique and obscure. This is so despite a recognition that some terms we use may communicate poorly and some are deprecating.

A notable example is “schizophrenia.” Originally referring to cognitive phenomena that were not adequately coherent with reality or one another, it has gone through periods of describing most psychosis to particular subsets of psychoses. Debates persist on specific criteria for key symptoms and typical course. Even two clinicians trained in the same site may not agree on the defining criteria, and the public, mostly informed by books, movies, and newspapers, is even more confused, often believing schizophrenia is multiple-personality disorder. In addition, the press and public often associate schizophrenia with violent behavior and uniformly bad outcomes, and for those reasons, a diagnosis is not only frightening but also stigmatizing.1

Many papers have presented the case for retiring “schizophrenia.”2 And practical efforts to rename schizophrenia have been made. These efforts have occurred in countries in which English is not the primary language.3 In Japan, schizophrenia was replaced by “integration disorder.” In Hong Kong, “disorder of thought and perception” was implemented. Korea chose “attunement disorder.” A recent large survey of stakeholders, including clinicians, researchers, and consumers in the United States, explored alternatives in English.4 Terms receiving approval included: “psychosis spectrum syndrome,” “altered perception syndrome,” and “neuro-emotional integration disorder.”

Despite these recommendations, the standard manuals of diagnosis, the ICD and DSM, have maintained the century-old term “schizophrenia” in their most recent editions, released in 2022. Aside from the inertia commonly associated with long-standing practices, it has been noted that many of the alternatives suggested or, in some places, implemented, are complex, somewhat vague, or too inclusive to distinguish different clinical presentations requiring different treatment approaches. They might not be compelling for use or optimal to guide caregiving.

Perhaps more concerning than “schizophrenia” are terms used to describe personality disorders.5 “Personality disorder” itself is problematic, implying a core and possibly unalterable fault in an individual. And among the personality disorders, words for the related group of disorders called “Cluster B” in the DSM raise issues. This includes the terms narcissistic, antisocial, histrionic, and borderline in DSM-5-TR. The first three terms are clearly pejorative. The last is unclear: What is the border between? Originally, it was bordering on psychosis, but as explained in DSM and ICD, borderline disorder is much more closely related to other personality disorders.

Notably, the “Cluster B” disorders run together in families, but men are more likely to be called antisocial and women borderline, even though the overlap in signs and symptoms is profound, suggesting marginally different manifestations of the same condition. The ICD has made changes to address the problems associated with some of these terms. ICD proposes personality “difficulty” to replace personality “disorder”; a modest change but less offensive. And it proposes seeing all, or at least most, personality disorders as being related to one another. Most share features of disturbances in sense-of-self and relationships with others. As descriptors, ICD kept “borderline pattern,” but replaced “antisocial” with “dissocial,” in an effort to be accurate but less demeaning. Other descriptors it proposes are negative affectivity, detachment, disinhibition, and anankastia, the last referring to compulsions.

These are notable advances. Can the field find even better terms to communicate hard to hear information, with words that are less problematic? In search of options, we surveyed clinicians at academic centers about the terms they preferred to avoid and the ones they prefer to use in talking with patients.6 Their practices may be informative.

Briefly summarized, these clinicians preferred not to use “schizophrenia” and very few used “antisocial,” “histrionic,” or “narcissistic.” Most avoided using “borderline” as well. Instead, they recommended discussing specific symptoms and manifestations of illness or dysfunctional behavior and relationships with their patients. They employed terms including “psychosis,” “hallucination,” “delusion,” “thinking disorder,” and “mood disorder.” They explained these terms, as needed, and found that patients understood them.

For Cluster B personality disorders, they spoke of personality traits and styles and specifically about “conduct,” “rule breaking,” “coping,” “self-focus,” “emotionality,” and “reactivity.” Those choices are not perfect, of course. Medical terms are often not standard words used in a conversational way. But the words chosen by these clinicians are generally straightforward and may communicate in a clear and acceptable fashion. It is also notable that the terms match how the clinicians assess and treat their patients, as observed in a separate study of their practices.7 That is, the clinicians advised that they look for and suggest treatments for the specific symptoms they see that most disrupt an individual’s life, such as delusions or mood instability. They are not much guided by diagnoses, like schizophrenia or borderline disorder. That makes the chosen terms not only less confusing or off-putting but also more practical.

Changing terminology in any field is difficult. We are trained to use standard terms. Clearly, however, many clinicians avoid some terms and use alternatives in their work. Asked why, they responded that they did so precisely to communicate more effectively and more respectfully. That is key to their treatment goals. Perhaps others will consider these choices useful in their work. And perhaps both the DSM and the ICD will not only continue to consider but will decide to implement alternatives for problematic terms in the years ahead, as they discuss their next revisions.

Dr. Cohen is director of the Program for Neuropsychiatric Research at McLean Hospital, Belmont, Mass., and Robertson-Steele Professor of Psychiatry at Harvard Medical School, Boston.


1. Lasalvia A et al. Renaming schizophrenia? A survey among psychiatrists, mental health service users and family members in Italy. Schizophr Res. 2021;228:502-9.

2. Gülöksüz S et al. Renaming schizophrenia: 5 x 5. Epidemiol Psychiatr Sci. 2019;28(3):254-7.

3. Sartorius N et al. Name change for schizophrenia. Schizophr Bull. 2014;40(2):255-8.

4. Mesholam-Gately RI et al. Are we ready for a name change for schizophrenia? A survey of multiple stakeholders. Schizophr Res. 2021;238:152-60.

5. Mulder R. The evolving nosology of personality disorder and its clinical utility. World Psychiatry. 2021 Oct;20(3):361-2.

6. Cohen BM et al. Diagnostic terms psychiatrists prefer to use for common psychotic and personality disorders. J Psychiatr Res. 2022 Sep 5;155:226-31.

7. Cohen BM, et al. Use of DSM-5 diagnoses vs. other clinical information by US academic-affiliated psychiatrists in assessing and treating psychotic disorders. World Psychiatry. 2021 Oct;20(3):447-8.

Recommended Reading

Immunodeficiencies tied to psychiatric disorders in offspring
Federal Practitioner
New insight into preventing antipsychotic-induced weight gain
Federal Practitioner
Cognitive remediation training reduces aggression in schizophrenia
Federal Practitioner
Antipsychotic cuts Alzheimer’s-related agitation
Federal Practitioner
Add-on antipsychotic beats switching meds in older adults with resistant depression
Federal Practitioner
New schizophrenia genes identified
Federal Practitioner
Clozapine may curb schizophrenia’s ‘most dreaded outcome’
Federal Practitioner
Erratic sleep, lack of activity tied to worsening schizophrenia symptoms
Federal Practitioner
New tool accurately predicts suicide risk in serious mental illness
Federal Practitioner
Young men at highest schizophrenia risk from cannabis abuse
Federal Practitioner