High Hopes for the Diagnostic Model HiTOP

Psychological disorders are currently classified using the Diagnostic Statistical Manual (DSM), a book of diagnoses, now in its fifth revision, which contains a list of discrete disorders denoted by their symptoms. This classification system has served as useful common language for researchers, clinicians, and insurance companies since it was first published in 1952. Yet the DSM has been facing increased scrutiny and criticism, as new research has accumulated to reveal deep flaws within it.

Source: mohamed_hassan for Pixabay

First, the DSM system considers mental disorders as categories (you either have it or you don’t). The evidence, however, shows that behavior exists on a continuum from normal range to dysfunctional. No existing mental disorder has been shown to exist as a discrete, categorical entity. No sharp demarcation lines exist that separate ‘disordered’ from ‘non-disordered’ symptom level. In addition, imposing a categorical either/or structure on a dimensional construct results in the loss of much useful information. An 'all-or-nothing' system inevitably misses all the ‘something’ action in between.

Further, research on traditional DSM diagnoses has shown them to possess low inter-rater reliability, which means that different experts, looking at the same patient, will often arrive at different DSM diagnoses. This is in part because the symptoms of many different disorders overlap greatly and because current diagnostic categories suffer from a lack of coherence, with many disorders containing multiple, diverse symptoms and myriad pathological processes.

Moreover, the current diagnostic system is marked by high rates of co-occurrence (comorbidity) among different disorders. Clients who are diagnosed with different disorders in fact share multiple symptoms. In addition, medications (such as SSRIs) and therapy interventions (such as CBT) that work for one disorder often work for others. This suggests that the DSM diagnostic system may be cutting unitary phenomena into pieces without justification. Two disorders that share many of their characteristic symptoms may be one disorder, mislabeled.

Finally, many patients who show up at psychology clinics experiencing significant mental distress or impairment in fact fail to meet the criteria for any specific disorder. This means that the system, as currently construed, often fails to address the pragmatic needs of its target constituents.

HiTOP: A New Approach

In the past 20 years, a call within psychology to change the old DSM system has been gathering strength and momentum. These efforts have coalesced into what is known as the ‘quantitative classification movement,’ a group of scientists whose goal is to do away with the traditional reliance on a priori (non-tested) assumptions about mental disorders (such as the assumption that they are categorical) and replace the old categorical system with a new dimensional one based on findings about the actual organization of psychopathology. Such as system will provide a more useful guide for researchers, patients, and clinicians, and will enable psychologists to study and treat characteristics common to multiple conditions.

The new approach uses statistical procedures (like factor analyses) to identify constellations of co-occurring signs, symptoms, and maladaptive traits and behaviors, and organize them hierarchically based on patterns of association. In this model, certain mental health constructs (such as externalizing behaviors) are measured and regarded on a continuum in the same way physicians may regard blood pressure or viral load. Different levels of the construct are then empirically associated with different common outcomes and treatment is assigned based on these risks.

Research on the new approach has culminated in the formation of the Hierarchical Taxonomy of Psychopathology (HiTOP), a dimensional model of psychopathology. HiTOP looks to empirically identify psychopathology structures by "combining individual signs and symptoms into homogeneous components or traits, assembling them into empirically-derived syndromes, and finally grouping them into psychopathology spectra"—fundamental dimensions of maladaptive behavior. The spectra include both maladaptive traits and psychopathology symptoms, which parallel each other on different time scales. Traits capture typical and relatively stable (chronic) personality tendencies, while symptoms tend to reflect the current, acute picture. HiTOP research has so far discovered six spectra:

1. The ‘internalizing’ dimension accounts for the comorbidity found among disorders such as depression, anxiety, post-traumatic stress, and eating disorders, as well as sexual dysfunctions and obsessive-compulsive disorder.
2. The disinhibited externalizing dimension accounts for comorbidity among disorders such as substance abuse, conduct disorder, antisocial behavior, and attention-deficit-hyperactivity disorder.
3. The antagonistic externalizing dimension accounts for comorbid symptoms of personality disorders such as paranoid and borderline personality.
4. A thought disorder dimension ranges from normal reality testing to maladaptive trait psychoticism, to hallucinations and delusions, and encompasses symptoms and traits commonly seen in psychotic disorders and bipolar disorder.
5. A detachment dimension ranges from introversion to maladaptive detachment, to blunted affect and a-volition. It accounts for symptoms that are commonly diagnosed in avoidant personalities.
6. Finally, the somatoform dimension, related to conversion disorder and other presentations of psychologically based physiological symptoms, has been provisionally included in the model, as evidence is still equivocal about whether it is indeed unique from the internalizing dimension.

DSM Essential Reads

French Literature’s Most Famous Detective Criticizes the DSM

The Myth of Mental Health Diagnosis

Work is continuing on the model, with research looking to refine our understanding of the dimensions. Evidence exists that individual spectra may be combined into larger ‘super-spectra.’ For example, ‘emotional dysfunction’ combines the characteristics of internalizing and somatoform spectra, while ‘psychosis’ combines thought disorder and detachment spectra. Above these super-spectra may sit a general psychopathology dimension, (‘p factor’) containing features common to all mental disorders.

Researchers are also looking to fine-tune the spectra by examining their sub-factors. For example, the internalizing dimension has been found to include two distinct subgroups: distress factor (evident in depression, PTSD, and generalized anxiety) and fear factor (evident in panic disorder and OCD).

The HiTOP model is new and has generated the needed debate concerning its theoretical and practical limitations. HiTOP, the critics point out, “provides little specific guidance towards our ultimate goal, namely, a classification of mental disorders based on causal factors and mechanisms involved in the first development of psychopathology and its progression over time.”

Yet, research on the efficacy of the new approach has been accumulating, and the results are quite promising. HiTOP dimensions appear to account for psychopathology-related impairment better than DSM diagnoses, and have been shown to predict future onsets of disorders and symptom chronicity better than past and current DSM diagnoses. Research has shown that due to the current system’s deficiencies, community clinicians often do not select treatment according to DSM diagnosis, focusing instead on symptoms and presenting complaints, in effect aligning their decision-making with HiTOP descriptions rather than with traditional diagnoses.

Today, clients receive tests for various disorders with which they may be labeled. In the future, if the HiTOP model is to prevail, clients will receive tests for the various empirically derived psychopathology and personality dimensions to generate a psychological profile. Psychological profiles will replace diagnostic labels. Cutoff scores will replace categories. Clients’ profile scores (and their scores on each individual dimension) will correspond empirically to certain genetic vulnerabilities, environmental risk factors, neurobiological abnormalities, illness course, and functional impairment. These data will then be used to guide treatment.

Among researchers and clinicians in the field, the limitations of the current diagnostic system have been an open secret for many years now. A change of that system has been long overdue. The good news is that, with the growing noise around HiTOP, such change may be in view.