Rethinking the Diagnosis of Mental Disorders: Data-Driven Psychological Dimensions,Not Categories,as a Framework for Mental-Health Research,Treatment,and Training

Abstract

Generations of psychologists have been taught that mental disorder can be carved into discrete categories, each qualitatively different from the others and from normality. This model is now outdated. A preponderance of evidence indicates that (a) individual differences in mental health (health vs. illness) are a matter of degree, not kind, and (b) broad mental-health conditions (e.g., internalizing) account for the tendency of narrower ones (e.g., depression, social anxiety, panic) to co-occur. With these observations in mind, we discuss an alternative diagnostic system, called the Hierarchical Taxonomy of Psychopathology (HiTOP), that describes the broad and specific components of mental disorder. It deconstructs traditional diagnostic categories, such as those listed in the Diagnostic and Statistical Manual of Mental Disorders, and recasts them in terms of profiles of dimensions. Recent findings support the utility of this approach for mental-health research and intervention efforts. HiTOP has the potential to put mental-health research, training, and treatment on a much sounder scientific footing.

Keywords

classification diagnosis Hierarchical Taxonomy of Psychopathology (HiTOP)individual differences mental disorder

The American Psychiatric Association’s Diagnostic and Statistical Manual of Mental Disorders (DSM), now in its fifth edition (DSM-5; American Psychiatric Association, 2013), has for decades been the authoritative text for diagnosing mental illness. It is so authoritative, in fact, that the news media often refer to the DSM as the “bible” of clinical psychology. It frames most research activity, clinical services, and educational curricula in mental-health fields.¹

The DSM’s longevity and influence belie a shaky scientific foundation. Many DSM diagnoses have poor empirical track records, possibly because most of them were formulated decades ago, when standards of evidence were very different. For example, many of the diagnoses installed in the DSM’s third edition, which would become the template for subsequent DSMs, were based heavily on experts’ own experiences in the clinic (Spiegel, 2005).

Times are changing, to some extent, as the latest DSM revision efforts have weighted empirical evidence in addition to clinical experience (Kendler & Solomon, 2016). Still, there are plenty of ways that DSM committees do not strictly follow the objective data. Social and political forces inevitably sway these committees (Zachar et al., 2016). Also, change is conservative, such that outdated, even faulty, diagnoses tend to be impervious to data that invalidate them.

Many researchers are now demanding more of mental-disorder diagnoses. They are calling for concepts that are backed by empirical data, rather than expert opinion. In this article, we describe a new way to map the landscape of mental-health conditions, one that is more inductive and data based than deductive. It is gaining ground on the DSM and changing the way the field approaches mental disorder.

Diagnosing the Problem

Almost 70 years ago, the DSM’s architects decided to fashion mental-disorder categories in the mold of classical medical diagnoses. This categorical structure remains today in DSM-5. The manual is based on the assumptions that people are either members or nonmembers of each diagnostic category and that categories are distinct entities separated by relatively crisp boundaries.

These assumptions turn out to be mistaken. First, people rarely fit neatly into a single category. Those with one diagnosis tend to qualify for others at above-chance rates (Plana-Ripoll et al., 2019). This phenomenon is called comorbidity, and it has long been a source of frustration in mental-health research. Imagine a research team studying amygdala reactivity to social-evaluative threat in the context of social anxiety disorder (SAD), defined in DSM-5 by sensitivity to negative evaluation and avoidance of social situations. The team asks people with SAD and healthy counterparts to listen to a series of rejection messages while in a functional magnetic resonance imaging scanner and then compares the two groups’ amygdala response to this “threatening” stimulus. During data analysis, they stumble upon a problem: Almost all participants with SAD were also diagnosed with major depressive disorder, a DSM-5 condition marked by sadness, loss of interest in usual activities, appetite problems, and concentration difficulties. Perhaps, then, any observed correlation between SAD and amygdala hyperactivity could be attributable to depression instead. Comorbidity makes it difficult, if not impossible, to disentangle the effects of co-occurring conditions.

Second, people with a mental-disorder diagnosis are not qualitatively different from “healthy” people. Across the range of human experiences that define mental health and mental illness, people differ in degree, not kind (Haslam et al., 2020). When researchers dichotomize people’s naturally continuous characteristics, they discard valuable information. Take, for example, a study on how antisocial personality disorder relates to job performance in police departments. DSM-5 requires that at least three of seven diagnostic criteria—lawbreaking, deceitfulness, impulsivity, repeated aggression, disregard for others’ safety, irresponsibility, and indifference to others’ pain—are present in order for the diagnosis to be made. Assuming that all people without a diagnosis—that is, with two or fewer symptoms—perform generally the same way on the job is probably misguided. Imagine the differences in policing one could expect from someone who expresses two characteristics of antisocial personality disorder (e.g., deceitfulness and lack of empathy) as opposed to someone with none.

Professional bodies are heeding these limitations. Although still generally resistant to fully dimensional models, the American Psychiatric Association and the World Health Organization, which publish the DSM and International Classification of Diseases (World Health Organization, 2020), respectively, both incorporated some dimensional elements into the recent revisions of their diagnostic manuals (e.g., Narrow & Kuhl, 2011; Tyrer et al., 2019). The U.S. National Institute of Mental Health (NIMH), which funds a great deal of scientific work on psychological disorders, went further and jettisoned categorical diagnoses from their latest strategic plan (NIMH, 2020). This move was consistent with NIMH’s Research Domain Criteria initiative, a blueprint for mental-health research that foregrounds biological and continuously distributed components of psychopathology.

Empirical Classification

An empirical² classification system brings objective data to the fore. It rests on quantitative analysis of how symptoms naturally aggregate (i.e., co-occur), instead of clinical experience. There is, in fact, a long tradition of research into the quantitative architecture of mental-disorder signs and symptoms (Moore, 1930). Until recently, however, it had mostly taken place in the shadow of the DSM.

This body of evidence has identified numerous symptom clusters that can be assembled to form the “building blocks” of mental disorder. Some, such as fear of public speaking, are narrow in scope. Others, such as phobias of all “performance” situations (e.g., speaking, musical performance, athletics), are more complex, and others, such as the general propensity to social anxiety, are broader still. As we mentioned, research indicates that these symptom clusters are best conceived of as dimensions on which people differ quantitatively, not qualitatively. It is probably easy for most people to envision a continuum of social anxiety, marked on one end by individuals who crave the spotlight and on the other by those who are desperate to avoid scrutiny. The critical point is that there is space on this continuum for everyone in between (see Fig. 1).

Fig. 1.

Histogram showing the frequency of responses to a social anxiety questionnaire in a hypothetical data set of 2,000 randomly selected people. The x-axis represents the spectrum of social anxiety, ranging from gregariousness (at the left tail) to severe social phobia and avoidance (at the right tail). The y-axis shows the count of people who occupy each location on the social anxiety spectrum, according to scores on this questionnaire. We simulated these 2,000 scores to follow a normal distribution. The shading delineates different regions of severity that were defined according to how a person’s score ranks relative to others’. The bottom half of the distribution is coded by a single color; these scores are all presumed to reflect benign (or adaptive) levels of social anxiety. Meanwhile, above the mean, each standard-deviation increment is represented by a different shade. Roughly one third of all scores fall between the mean and 1 SD above the mean, another 14% span the area between 1 and 2 SD above the mean, and 2% are farther than 2 SD above the mean. As the figure’s key suggests, these regions might correspond to different approaches to symptom management. Around half of the scores (below the sample mean) reflect anxiety that requires no intervention. People with scores between 0 and 1 SD above the mean might benefit from more education about (and corresponding awareness of) anxiety and/or self-help practices. Higher scores could warrant more intensive intervention, such as psychotherapy, medication, or their combination. This characterization is just a coarse heuristic, intended to illustrate how mental-health professionals might use dimensional information, although it is modeled on research on how best to address clinical problems that run the full gamut of severity in the population.

When researchers zoom out to view the full universe of mental-health problems, they find that symptoms are related to one another hierarchically. This means that narrower concepts are nested within broader ones (see Fig. 2). This finding was not totally unexpected because it parallels the structure of other domains of human individual differences, such as cognitive ability and personality (e.g., Deary, 2012). As explained in the next section, broad concepts represent wide sets of symptoms that have a common theme (e.g., social anxiety). Further down the hierarchy, narrower concepts represent more specific symptom manifestations tied to particular contexts (e.g., having a conversation with an authority figure). All of these concepts, regardless of their location in the hierarchy, appear to be dimensional, such that people can fall anywhere on the spectrum from low to high severity.

Fig. 2.

Diagram illustrating the hierarchical structure of the domain of social anxiety symptoms. It resembles an “org chart” that details the relative positions people hold in a business or corporation. At the top, the social anxiety dimension represents the common thread through all fears of embarrassment and humiliation. At an intermediate level, there are different ways that social anxiety can be expressed. They all reflect social anxiety, but they are distinct manifestations. For example, as illustrated here, some people experience social anxiety only in performance situations, whereas others find one-on-one social interaction more challenging. At the most granular level of this hierarchy, there are different individual symptoms that constitute the basic units of the social anxiety domain. For instance, fear of being watched while working or otherwise performing and discomfort with eating, drinking, or writing in front of other people compose the performance-anxiety subdomain. Social anxiety can be understood at any of these levels of resolution. The different levels are interlocking parts of the same system.

The Hierarchical Taxonomy of Psychopathology

In 2015, a group of mental-health researchers and clinicians founded the Hierarchical Taxonomy of Psychopathology (HiTOP) consortium. Their objective was to (a) translate the available data on the dimensions of mental disorder into an evidence-based diagnostic rubric and (b) mobilize new research to continuously refine the system. They recently published a working model that encapsulates all the major axes of mental disorder that have been established so far (Kotov et al., 2017; see Fig. 3).

Fig. 3.

A map of the signs and symptoms of mental disorder, according to the Hierarchical Taxonomy of Psychopathology (HiTOP) model. As this diagram shows, the model is similar to a phylogeny. Broader psychological dimensions at higher levels of the hierarchy branch out to form narrower dimensions at lower levels. The thin vertical lines represent the nested relationship between broader and narrower concepts. This is a working model to the extent that some existing elements are provisional (i.e., based on relatively few studies) and others will be added as other domains of mental disorder are examined empirically (e.g., autism). As explained in the main text, diagnoses from the Diagnostic and Statistical Manual of Mental Disorders (DSM) are not part of the HiTOP model. This departure from the DSM’s approach to classification is represented in the figure by the solid line separating HiTOP from DSM concepts. Yet HiTOP is based on the same signs and symptoms that compose DSM diagnoses, and the color scheme symbolizes where the primary features of DSM disorders are likely to fall in the HiTOP structure. HiTOP includes empirical syndromes, which are similar in breadth to DSM diagnoses but, critically, are based on natural patterns of symptom and trait co-occurrence and may not resemble the diagnostic entities listed in the DSM. We do not yet have a firm impression of the number and nature of dimensions that will emerge at this level. As represented by the vertical line connecting the two lowest levels of the HiTOP structure, empirical syndromes represent the tendency to experience a set of related, fine-grain symptoms and maladaptive personality traits (see Fig. 3 in Kotov et al., 2017, for a list of symptom components and maladaptive traits). OCD = obsessive-compulsive disorder; ADHD = attention-deficit/hyperactivity disorder.

Figure 3’s vertical axis shows how clinical problems vary in breadth. Situated at the bottom of the HiTOP diagram, symptom components are tight-knit bundles of symptomatic behaviors. Maladaptive traits, which are relatively narrow personality dispositions established in research on personality disorder, occupy this same layer of the hierarchy. These homogeneous components—the most atomistic level of mental disorder—are too numerous to convey in Figure 3 (see Fig. 3 in Kotov et al., 2017, for a list of more than 100). By way of example, consider how Figure 2’s representation of social anxiety could be grafted onto Figure 3. Social anxiety is a constellation of symptoms that fits in alongside panic, separation anxiety, alcohol use, and other syndromes at the second-to-lowest tier of the hierarchy. Its most fine-grain elements, such as fear of eating in front of others and avoidance of situations in which one might meet new people, are symptom components that form the base of the hierarchy.

As this example suggests, symptom components and maladaptive traits can be thought of as the constituent parts of broader mental-health syndromes. Social anxiety, for example, is a syndrome that brings together various expressions of performance and social-interaction anxiety. A depression syndrome brings together mood disturbance, appetite problems, cognitive difficulties, and more. In the psychological literature, the term syndrome, which was borrowed from medicine, sometimes connotes a discrete entity. This is the meaning implied in the DSM (i.e., a binary disorder that someone either does or does not have). HiTOP presents a very different view. Syndromes are conceptualized as dimensional tendencies to experience related symptoms and maladaptive traits. In other words, each HiTOP syndrome is a continuously distributed concept, not category, that spans a range of severity (Krueger et al., 2018).

As are all other concepts in the HiTOP hierarchy, the syndromes are dimensional because that is how data analysis overwhelmingly has suggested they behave (see, e.g., Waszczuk et al., 2017). If convincing new evidence of categorical individual differences at that level (or others) materialized, then binary syndromes, akin to DSM diagnoses, would be included in the model.

The syndrome concepts are provisional pieces of the HiTOP model at the moment; there is a small, but growing, body of research on how fine-grain symptoms naturally co-occur to create syndromes. It is already clear, though, that the HiTOP syndromes are not replicas of DSM diagnoses. This is what the solid horizontal line in Figure 3 is meant to communicate. HiTOP is based on the same signs and symptoms of mental-health problems that make up DSM diagnoses, but HiTOP rearranges them into dimensional composites based on quantitative evidence. This emphasis on data-driven composites of symptoms and traits is why we use the label empirical syndromes for these concepts in Figure 3.

The HiTOP model posits that empirical syndromes are grouped within even more complex concepts at the subfactor and spectrum tiers of the hierarchy. Developmentalists in particular will be used to working with the internalizing and externalizing spectrum concepts, which have been a standard way to represent mental-health complaints among youth for more than half a century (Achenbach, 1966). At the apex of the hierarchy, a superspectrum accounts for the tendency for all mental-health problems to go together. It is sometimes called the p factor after its likeness to the g factor of intelligence (Caspi et al., 2014), which represents the tendency of people who do well on tests of one type of cognitive ability (e.g., verbal skills, mathematics, reasoning) to perform well on all the others, too. Scholars are currently pondering the meaning of the p factor. Among other possibilities, it might represent dysfunction, distress, or demoralization common to all forms of mental disorder (Caspi & Moffitt, 2018).

The HiTOP model might look novel at first glance, but it contains the same clinical phenomena that researchers are used to, just reorganized as dimensions based on empirical evidence. Familiar psychological disorders, such as depression, alcohol abuse, and eating pathology, are essentially dismantled and rearranged at different levels of the hierarchy according to data.

Who Is HiTOP for?

HiTOP is compatible with a wide range of research programs, and self- and informant-report measures of its symptom dimensions make it fairly easy to incorporate into most common study designs (see Kotov et al., 2017; see also HiTOP’s website at https://hitop.unt.edu/introduction). Such measures, which tap into the concepts presented in Figure 3, are typically published with detailed normative data that foster interpretation of scale scores. Many of them involve both broad and narrow scales so that investigators can focus on the level of the hierarchy that is best suited to their research question. In addition, the HiTOP consortium is working to develop a comprehensive assessment system that covers the entire model (Simms et al., 2021). These tools are all that investigators need to bring research projects and clinical activity into alignment with the empirical-classification movement.

HiTOP was not designed only for clinical practice and research (although it certainly has value for clinical practice—see Ruggero et al., 2019). It is suitable for any psychological research project that deals with mental-health problems. Basically, HiTOP is a valid alternative for any conceptual model or practical task that would have involved DSM symptoms or diagnoses. It is intended not only for clinicians and clinical researchers but also for psychologists who integrate mental-health problems into their research programs as a way to better understand other phenomena, such as neural circuitry, close relationships, organizational behavior, or cultural processes. Conway et al. (2019) reviewed ways the model has generated new insights in diverse research areas.

We realize that research habits can be entrenched. Categorical diagnoses are familiar and ubiquitous, and most of the field’s knowledge base is anchored to the DSM. Our position is that HiTOP provides substantial added value that makes the switch to a dimensional approach worthwhile for most researchers and clinicians. By portraying mental disorders in terms of dimensions, as opposed to categories, HiTOP preserves information about individual differences in mental health, thereby enabling more reliable and valid measurement (Markon et al., 2011). By deconstructing categorical diagnoses into their constituent parts, it sheds light on the aspects of mental disorder that have the most predictive power. By taking an empirical stance toward classification, it ensures that diagnostic concepts will evolve with new data, not ossify like the many decades-old diagnoses that persist in DSM-5.

Conclusion

The empirical approach to classification is on the verge of becoming mainstream. For the first time in decades, early-career psychologists are not pursuing research agendas on, say, DSM-5 generalized anxiety disorder, alcohol use disorder, or narcissistic personality disorder. The word is out that categories are a suboptimal way to represent mental disorder, and researchers are looking to other models for guidance.

HiTOP’s empirical system provides a more scientifically defensible perspective on mental disorder. It encodes the same symptoms that compose DSM categories, but it reconfigures them. Mental disorders are recast as dimensions, not types. Concepts are arranged in a hierarchy, a framework consistent with data pointing to nesting among mental-health conditions. Above all, the model responds to objective data, so it is designed to break away from clinical opinion and received wisdom when the evidence so dictates.

HiTOP is ready for implementation now, and it is backed by a solid evidence base. This is a chance for psychologists to move the field in the direction of a more credible, scientifically sound version of mental-disorder diagnosis and research.

Footnotes

Transparency

Action Editor: Robert L. Goldstone

Editor: Robert L. Goldstone

Members of the HiTOP Consortium Executive Board (listed alphabetically)

David C. Cicero, University of North Texas

Colin G. DeYoung, University of Minnesota

Nicholas R. Eaton, Stony Brook University

Miriam K. Forbes, Macquarie University

Michael N. Hallquist, University of North Carolina at Chapel Hill

Roman Kotov, Stony Brook University

Robert D. Latzman, Georgia State University

Camilo J. Ruggero, University of North Texas

Leonard J. Simms, University of Buffalo, The State University of New York

Irwin D. Waldman, Emory University

Monika A. Waszczuk, Rosalind Franklin University

David Watson, University of Notre Dame

Thomas A. Widiger, University of Kentucky

Aidan G. C. Wright, University of Pittsburgh

ORCID iD

Christopher C. Conway

Notes

References

Achenbach

T. M.

(1966). The classification of children’s psychiatric symptoms: A factor-analytic study. Psychological Monographs: General and Applied, 80(7), 1–37. https://dx-doi-org.web.bisu.edu.cn/10.1037/h0093906

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). https://doi.org/10.1176/appi.books.9780890425596

Caspi

Houts

R. M.

Belsky

D. W.

Goldman-Mellor

S. J.

Harrington

HL.

Israel

Meier

M. H.

Ramrakha

Shalev

Poulton

Moffitt

T. E.

(2014). The p factor: One general psychopathology factor in the structure of psychiatric disorders? Clinical Psychological Science, 2(2), 119–137.

Caspi

Moffitt

T. E.

(2018). All for one and one for all: Mental disorders in one dimension. American Journal of Psychiatry, 175(9), 831–844.

Conway

C. C.

Forbes

M. K.

Forbush

K. T.

Fried

E. I.

Hallquist

M. N.

Kotov

Mullins-Sweatt

S. N.

Shackman

A. J.

Skodol

A. E.

South

S. C.

Sunderland

Waszczuk

M. A.

Zald

D. H.

Afzali

M. H.

Bornovalova

M. A.

Carragher

Docherty

A. R.

Jonas

K. G.

Krueger

R. F.

. . . Eaton

N. R.

(2019). A hierarchical taxonomy of psychopathology can transform mental health research. Perspectives on Psychological Science, 14(3), 419–436.

Deary

I. J.

(2012). Intelligence. Annual Review of Psychology, 63, 453–482.

Haslam

McGrath

M. J.

Viechtbauer

Kuppens

(2020). Dimensions over categories: A meta-analysis of taxometric research. Psychological Medicine, 50(9), 1418–1432. 1–15. https://doi.org/10.1017/S003329172000183X

Kendler

K. S.

Solomon

(2016). Expert consensus v. evidence-based approaches in the revision of the DSM. Psychological Medicine, 46(11), 2255–2262.

Kotov

Krueger

R. F.

Watson

Achenbach

T. M.

Althoff

R. R.

Bagby

R. M.

Brown

T. A.

Carpenter

W. T.

Caspi

Clark

L. A.

Eaton

N. R.

Forbes

M. K.

Forbush

K. T.

Goldberg

Hasin

Hyman

S. E.

Ivanova

M. Y.

Lynam

D. R.

Markon

. . . Zimmerman

(2017). The Hierarchical Taxonomy of Psychopathology (HiTOP): A dimensional alternative to traditional nosologies. Journal of Abnormal Psychology, 126(4), 454–477.

10.

Krueger

R. F.

Kotov

Watson

Forbes

M. K.

Eaton

N. R.

Ruggero

C. J.

Simms

L. J.

Widiger

T. A.

Achenbach

T. M.

Bach

Bagby

R. M.

Bornovalova

M. A.

Carpenter

W. T.

Chmielewski

Cicero

D. C.

Clark

L. A.

Conway

DeClercq

DeYoung

C. G.

. . . Zimmermann

(2018). Progress in achieving quantitative classification of psychopathology. World Psychiatry, 17(3), 282–293.

11.

Markon

K. E.

Chmielewski

Miller

C. J.

(2011). The reliability and validity of discrete and continuous measures of psychopathology: A quantitative review. Psychological Bulletin, 137(5), 856–879.

12.

Moore

T. V.

(1930). The empirical determination of certain syndromes underlying praecox and manic-depressive psychoses. American Journal of Psychiatry, 86(4), 719–738.

13.

Narrow

W. E.

Kuhl

E. A.

(2011). Dimensional approaches to psychiatric diagnosis in DSM-5. Journal of Mental Health Policy and Economics, 14(14), 197–200.

14.

National Institute of Mental Health. (2020). NIMH strategic plan for research (NIH Publication No. 20-MH-8096). U.S. Department of Health and Human Services, National Institutes of Health. https://www.nimh.nih.gov/strategicplan

15.

Plana-Ripoll

Pedersen

C. B.

Holtz

Benros

M. E.

Dalsgaard

De Jonge

Fan

C. C.

Degenhardt

Ganna

Greve

A. N.

Gunn

Iburg

K. M.

Kessing

L. V.

Lee

B. K.

Lim

C. C. W.

Mors

Nordentoft

Prior

Roest

A. M.

. . . McGrath

J. J.

(2019). Exploring comorbidity within mental disorders among a Danish national population. JAMA Psychiatry, 76(3), 259–270.

16.

Ruggero

C. J.

Kotov

Hopwood

C. J.

First

Clark

L. A.

Skodol

A. E.

Mullins-Sweatt

S. N.

Patrick

C. J.

Bach

Cicero

D. C.

Docherty

Simms

L. J.

Bagby

R. M.

Krueger

R. F.

Callahan

J. L.

Chmielewski

Conway

C. C.

De Clercq

Dornbach-Bender

. . . Zimmermann

(2019). Integrating the Hierarchical Taxonomy of Psychopathology (HiTOP) into clinical practice. Journal of Consulting and Clinical Psychology, 87(12), 1069–1084.

17.

Simms

L. J.

Wright

A. G. C.

Cicero

D. C.

Kotov

Mullins-Sweatt

S. N.

Sellborn

Watson

Widiger

T. A.

Zimmermann

(2021). Development of measures for the Hierarchical Taxonomy of Psychopathology (HiTOP): A collaborative scale development project [Manuscript submitted for publication]. Department of Psychology, University of Buffalo, The State University of New York.

18.

Spiegel

(2005, January 3). The dictionary of disorder. The New Yorker, 3, 56–63.

19.

Tyrer

Mulder

Kim

Y. R.

Crawford

M. J.

(2019). The development of the ICD-11 classification of personality disorders: An amalgam of science, pragmatism, and politics. Annual Review of Clinical Psychology, 15, 481–502.

20.

Waszczuk

M. A.

Kotov

Ruggero

Gamez

Watson

(2017). Hierarchical structure of emotional disorders: From individual symptoms to the spectrum. Journal of Abnormal Psychology, 126(5), 613–634.

21.

World Health Organization. (2020). International statistical classification of diseases and related health problems (11th ed.). https://icd.who.int/

22.

Zachar

Krueger

R. F.

Kendler

K. S.

(2016). Personality disorder in DSM-5: An oral history. Psychological Medicine, 46(1), 1–10.

Rethinking the Diagnosis of Mental Disorders: Data-Driven Psychological Dimensions,Not Categories,as a Framework for Mental-Health Research,Treatment,and Training

Abstract

Keywords

Diagnosing the Problem

Empirical Classification

The Hierarchical Taxonomy of Psychopathology

Who Is HiTOP for?

Conclusion

Recommended Reading

Footnotes

Transparency

ORCID iD

Notes

References