The Hierarchical Taxonomy Of Psychopathology (HiTOP) consortium was formed in 2015 as a grassroots effort to articulate a classification of mental health problems based on recent scientific findings on how the components of mental disorders fit together. The consortium is developing the HiTOP model, a classification system, or taxonomy, of mental disorders, or psychopathology, aiming to prioritize scientific results over convention and clinical opinion. The motives for proposing this classification were to aid clinical practice and mental health research. The consortium was organized by Drs. Roman Kotov, Robert Krueger, and David Watson. At inception it included 40 psychologists and psychiatrists, who had a record of scientific contributions to classification of psychopathology The HiTOP model aims to address limitations of traditional classification systems for mental illness, such as the DSM-5 and ICD-10, by organizing psychopathology according to evidence from research on observable patterns of mental health problems. When the HiTOP model is complete, it will form a detailed hierarchical classification system for mental illness starting from the most basic building blocks and proceeding to the highest level of generality: combining individual signs and symptoms into narrow components or traits, and then combining these symptom components and traits into (in order of increasing generality) syndromes, subfactors, spectra, and superspectra. Currently, several aspects of the model are provisional or incomplete.
History of the quantitative classification movement through HiTOP Throughout the history of psychiatric classification, two approaches have been taken to deciding the content and boundaries of mental disorders that enter official diagnostic rubrics. A first one might be termed authoritative: experts and members of official bodies meet to determine classificatory rubrics through group discussions and associated political processes. This approach characterizes traditional classification systems, such as the DSM and the ICD.
A second approach might be termed empirical. In this approach, data are gathered on psychopathological building blocks. These data are then analyzed to address specific research questions. For example, does a specific list of symptoms delineate a single psychopathological entity or multiple entities? This approach is sometimes characterized as more "bottom up" (i.e., starting with raw observations and inferring the presence of diagnostic concepts), compared with the more "top down" approach (i.e., starting with a general clinical concept and deducing the symptoms that might define it) of official classification systems.
These approaches, although distinguishable, are not entirely separable. Some amount of empiricism and some amount of expert authority is inevitably present in both (i.e., authoritative classification approaches have relied on specific types of empiricism as part of their construction process, and an empirical approach begins with the expertise needed to assemble and assess specific psychopathological building blocks). Nevertheless, authoritative approaches tend to weigh putative expertise, disciplinary background, and tradition heavily. The consortium aims for an empirical rather than authoritative approach, but it has been argued that the HiTOP model is partly authoritative as it is grounded on a traditional but arbitrary statistical approach.
The empirical movement has a long history, beginning with the work of Thomas Moore, Hans Eysenck, Richard Wittenborn, Maurice Lorr, and John Overall, who developed measures to assess signs and symptoms of psychiatric inpatients, and identified empirical dimensions of symptomatology through factor analysis of these instruments. Others have searched for natural categories using such techniques as cluster analysis.
Similarly, research on patterns of emotional (also called affective) experience helped to identify dimensions of depression and anxiety symptoms. Factor analytic studies of child symptomatology found clusters of emotional and behavioral problems that remain in use in research and clinical assessment today. Finally, factor analyses of comorbidity among common adult disorders revealed higher-order dimensions of psychopathology that inspired a growing and diverse literature.
The most recent large-scale effort in this movement toward empirically based classification emerged in the spring of 2015. Forty scholars working in the classification of psychopathology started a consortium (now over 160 members with 10 workgroups) devoted to articulating an empirically based classification system of mental illness. Their initial proposed model – the Hierarchical Taxonomy of Psychopathology – has been claimed to provide a marked departure from DSM and ICD. The HiTOP model is based on structural studies that span from age 2 to 90 and include samples from many non-Western societies. However, Western samples are over-represented in this literature and very little research has been done with people over age 60. The HiTOP model does not account for individual level developmental processes that may lead to various disorder outcomes.
To update HiTOP as new structural and validation studies become available, the Consortium formed a Revisions Workgroup. This workgroup has designed a process for continuous evidence-based revision of the model. This process is intended to be nimble enough to keep pace with a rapidly growing literature on the structure of psychopathology, but not so fickle as to result in numerous changes without substantiated support.
HiTOP structure
Fundamental findings that shaped HiTOP Three fundamental findings shaped HiTOP. First, psychopathology is best characterized by dimensions rather than in discrete categories. Dimensions are defined as continua that reflect individual differences in a maladaptive characteristic across the entire population (e.g., social anxiety is a dimension that ranges from comfortable social interactions to distress in nearly all social situations). Dimensions reflect differences in degree (i.e., continua), rather than in kind (i.e., people are either in or outside of each category), as the evidence to date suggests that psychopathology exists on a continuum with normal-range functioning. These dimensions can be organized hierarchically from narrowest to broadest (see Figure). Specifically, dimensional description improves reliability and eliminates the need for "Other Specified" or "Unspecified" diagnoses, as every person has a standing on each dimension and thus is described. Nevertheless, some qualitative boundaries may exist in psychopathology. If categorical entities are identified and replicated, they would be added to HiTOP. Indeed, the term dimensional is not used in the name of the model, in recognition of openness to evidence on discrete entities.
Second, HiTOP assumes that the natural organization of psychopathology can be discerned in co-occurrence of its features. Classification that follows co-occurrence ensures coherence of diagnostic entities, so that related signs and symptoms are assigned together to tightly knit dimensions, whereas unrelated features are placed on different dimensions. Moreover, such constructs capture information about common genetics, risk factors, biomarkers, and treatment response shared by co-occurring forms of psychopathology. Third, psychopathology can be organized hierarchically from narrow to broad dimensions. Numerous studies have found that specific psychopathology dimensions aggregate into more general factors. Patterns of comorbidity are represented by higher-order dimensions. Accordingly, comorbidity is measured and expressed in scores that researchers and clinicians can use.
Organization of HiTOP model Consistent with these three fundamental findings, the HiTOP model consists of hierarchically organized dimensions identified in covariation of psychopathology features. Signs, symptoms, and maladaptive traits and behaviors are grouped into homogeneous components- constellations of closely related symptom manifestations; for example, fears of working, reading, eating, or drinking in front of others form performance anxiety cluster. Maladaptive traits are specific pathological personality characteristics, such as submissiveness. The leading conceptualization is that symptoms and maladaptive traits differ only in time frame. A symptom component reflects current functioning (e.g., past month), whereas the corresponding trait reflects functioning on the same dimension in general—that is, over many years.
Closely related homogenous components are combined into dimensional syndromes (e.g., social anxiety). Syndromes are composites of related components/traits, such as a social anxiety syndrome that encompasses both performance anxiety and interaction anxiety. Of note, the term syndrome can be used to indicate a category (for instance, some medical diseases such as Lyme disease are probably best thought of as natural discrete problems that someone either wholly has or wholly does not have), but here we use it to indicate a dimension. Importantly, HiTOP syndromes do not necessarily map onto traditional, categorical disorders like those found in DSM and ICD. Studies often have used categorical disorders to define HiTOP dimensions, but these categorical disorders are used as proxies and are not part of HiTOP as such. Rather than re-arranging DSM and ICD disorders, HiTOP aims to create a system based on signs and symptoms described in these manuals (as well as additional symptoms) and reorganize them based on how studies have found them to occur in combination.
Clusters of closely related syndromes form subfactors, such as the fear subfactor formed by strong links between social anxiety, agoraphobia, and specific phobias.
Spectra are larger constellations of syndromes, such as an internalizing spectrum composed of syndromes from fear, distress, eating pathology, and sexual problems subfactors. Six spectra have been included in HiTOP so far:
The thought disorder spectrum comprises maladaptive traits of peculiarity, unusual beliefs, unusual experiences, and fantasy proneness, as well as symptom dimensions of disorganization and reality distortion; also symptom dimensions of dissociation and mania are linked to this spectrum provisionally. The thought disorder spectrum includes some signs and symptoms of such disorders as schizophrenia and related disorders, mood disorders with psychosis, schizotypal personality disorder, and paranoid personality disorder, and provisionally dissociative disorders and bipolar disorders. The detachment spectrum comprises maladaptive traits of emotional detachment, anhedonia, social withdrawal, and romantic disinterest, as well as symptom dimensions of inexpressivity and avolition. The detachment spectrum includes some signs and symptoms of such disorders as schizoid personality disorder, avoidant personality disorder, schizotypal personality disorder, and schizophrenia and related disorders. The antagonistic externalizing spectrum comprises maladaptive traits of manipulativeness, deceitfulness, callousness, grandiosity, aggression, rudeness, domineering, and suspiciousness, as well as symptom dimensions characteristic of antisocial behavior, such as theft, fraud, destruction of property, and aggression. The antagonistic externalizing spectrum includes some signs and symptoms of such disorders as conduct disorder, antisocial personality disorder, intermittent explosive disorder, oppositional defiant disorder, histrionic personality disorder, paranoid personality disorder, narcissistic personality disorder, and provisionally borderline personality disorder.