Alternative DSM-5 model for personality disorders

The Alternative DSM-5 Model for Personality Disorders (AMPD), introduced in Section III of the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5),^[1] is an alternative conceptual framework for the classification and understanding of personality disorders. It differs from previous DSM models of personality disorders, including the standard model in the DSM-5, in that it is based on a dimensional approach to personality pathology, whereas previous models have been characterized by rigid diagnostic criteria for each individual personality disorder.^[2] The alternative model, on the other hand, aims to better capture the complexity of personality pathology by assessing impairments in personality functioning and pathological personality traits. Designed to address limitations of the categorical system—such as excessive comorbidity and lack of diagnostic precision^[2]—the alternative model offers a nuanced perspective that aligns more closely with contemporary research and clinical practice. Its focus on the interplay between personality traits and functioning aims to improve diagnostic accuracy and treatment planning, though it remains a topic of ongoing debate and research.^[3] The alternative model features the following specified personality disorders, in alphabetical order: antisocial, avoidant, borderline, narcissistic, obsessive–compulsive, and schizotypal.^[2][4] This constitutes a reduction of entities,^[4] as the standard model contains the additional diagnoses of dependent, histrionic, paranoid, and schizoid personality disorders.^[5]

The cover of the DSM-5, which introduced the Alternative Model for Personality Disorders (AMPD).

Background

The categorical model of personality disorders in the DSM was originally introduced in the DSM-III, released in 1980, replacing the previously used narrative-based diagnoses with diagnoses based on clearly defined criteria.^[6] This categorical model persisted in subsequent editions – such as the DSM-III-R and DSM-IV – of the DSM.^[5] In the 2000's, when the upcoming DSM-5 was being planned and developed, there was a consensus that the categorical model was insufficient due to personality disorders being better understood as dimensional,^[7] while the categorical approach was understood to have had a negative impact on the development of conceptual and clinical aspects of personality disorders.^[8] Shifting the paradigm in psychiatric diagnosis in the direction of a dimensional approach was a significant intention behind the proposal to create a new DSM edition.^[9]

Early undertakings

The Personality and Personality Disorders Work Group^[10] (PPDWG)^[9] was responsible for the development of the chapter on PDs for the DSM-5.^[8] Originally named the Personality Disorders Work Group (PDWG), it was headed by chair Andrew E. Skodol beginning in 2007; the renaming occurred due to the understanding that the group was to shift to a dimensional approach.^[10] The paradigm shift had come to be manifested in the exclusion of the majority of individuals who worked on the DSM-IV from the early stages of development of the DSM-5, including from the work groups,^[9] with Larry Siever being the only one of the eleven members of the PPDWG who had been part of the DSM-IV PDWG.^[10]

During their initial work, there was disagreement among the members of the PPDWG in regards to several aspects of the establishment of a new system for PDs for the upcoming DSM.^[9][10] While initially opting for a broad approach towards the selection of a conceptual model for PDs, agreement in favor of a dimensional approach emerged;^[9] however, disagreements continued regarding which specific model should be used.^[9][10] For example, there was tension between the use of one of several established models and the creation of a new one with lesser basis in research.^[10] Another matter of debate was whether all, or merely some, of the categorical PD diagnoses should remain,^[9] with consideration given to their validity and utility – as well as to these in relation to their entrenchment in practice and to the thus possible benefits of retention of continuity with the extant categorical model.^[10]

Emergence and rejection

The PPDWG iteratively developed what became a dimensional–categorical hybrid model of personality disorders which was to replace the previously used categorical model.^[7] In 2009, Skodol made the first proposal of a model, consisting of ratings of impairment in functioning, as well as of six trait domains consisting of specific traits – with both impairment and traits rated numerically. The five herein included personality disorders – namely: antisocial, avoidant, borderline, obsessive–compulsive, and schizotypal – were described through narratives, the correspondence of a subject with which was to be rated along with a rating of the elevation of the traits associated with the PD.^[10]

During the evolution of the model, it was redesigned as a hybrid model anchored in impairment in areas of personality functioning and dimensional traits. Initially absent, narcissistic personality disorder was included was brought into the model following feedback from clinicians and researchers. The model also faced scrutiny against strict standards for evidence and utility.^[10] The Work Group and the DSM-5 Task Force recommended the transition to the model in its final form;^[7] however, in 2012, the American Psychiatric Association (APA) Board of Trustees voted against it, resulting in the new model being included in section III of the DSM-5 as the Alternative DSM-5 Model for Personality Disorders,^[7][8] with the DSM-IV categorical model being retained as the standard model in section II.^[11]

Core features

Flowchart showing an overview of the AMPD's PD assessment procedure.

In the alternative model, personality disorders are conceptualized based on level of impairment of personality functioning, as well as specific pathological personality traits present in the subject. Central to this model is the Level of Personality Functioning Scale (LPFS), which quantifies impairment on a spectrum from 0 (little to no impairment) to 4 (extreme impairment). Personality functioning is divided into two interrelated domains: self-functioning (encompassing identity coherence and self-direction) and interpersonal functioning (involving empathy and capacity for intimacy). Each of these four elements operates along a continuum, spanning from adaptive functioning to severe dysfunction. Additionally, the AMPD identifies five broad trait domains—Negative Affectivity, Detachment, Antagonism, Disinhibition, and Psychoticism—that manifest in varying degrees across distinct personality disorders.

The AMPD contains both criteria in general for what constitutes a personality disorder (PD), as well as criteria for specific personality disorders. The general criteria apply to any diagnosis of a PD, and as such, to any specified personality disorder, as well as to Personality Disorder - Trait Specified.^[2]

Criterion A: Level of personality functioning

See also: ICD-11 classification of personality disorders § Personality disorder

Table showing the areas of impairment of personality functioning at each level of impairment. Only one indicator of each area is shown, whereas the LPFS has three for each.

Supposed to capture fundamental problems specific to, and common among, personality disorders,^[12] the level of personalty functioning is assessed in accordance with a scale provided with the AMPD called the Level of Personality Functioning Scale (LPFS).^[6] It ranges from 0 ("Little or no impairment") to 4 ("Extreme impairment"), thus making the model truly dimensional and inclusive of the entire spectrum of personality functioning.^[13]

In the LPFS, personality functioning is conceptualized as consisting of self functioning, comprising identity and self-direction, and interpersonal functioning, comprising empathy and intimacy.^[6][13] These four elements "tend to form a single dimension of impairment in factor analysis".^[12] The functioning level of each of these four components is on a continuum ranging from optimal functioning to extremely dysfunctional. A diagnosis of a personality disorder requires that the impairment is moderate or greater (impairment ≥ 2).^[14][15]

The introduction of the LPFS in the alternative model was driven by findings that personality disorders share substantial commonalities. Assessment of the level of personality functioning as a core element in the diagnosis of personality disorder also has the benefit of clearly defining how severe the impairment must be and how it manifests in general functioning, rather than diagnosis relying merely on traits, which may not necessarily cause impairment significant enough to warrant a diagnosis. In addition to this, the LPFS captures a clearly defined core of what personality functioning is, and how impairment generally manifests, regardless of specified pathological traits.^[13]

Criterion A, specifically, has been criticized for being inadequate at capturing the core essence of personality disorders and differentiating them from other types of mental disorders.^[12] It has been suggested that criterion A should be removed from the AMPD in favor of instead measuring the impacts of criterion B traits.^[12]

Criterion B: Pathological personality traits

See also: ICD-11 classification of personality disorders § Prominent personality traits or patterns

The pathological traits of criterion B serve to describe the characteristics of any personality disorder, with identification of traits providing an explanation to specific reasons for impairment in personality functioning.^[1] 25 pathological personality traits, known as trait facets, are grouped into the following domains: negative affectivity, detachment, antagonism, disinhibition, and psychoticism.^[1][6]

These domains differ in prominence between specific personality disorders; for example, schizotypal personality disorder is characterized by traits of detachment and psychoticism, according to the criteria in the AMPD.^[2] The trait domains are explained below:

• Antagonism: Behaviors marked by disregard for others and which are conducive to disharmony; this includes lack of empathy, entitlement, and a tendency to manipulate or exploit for personal gain.^[16] This is a core trait domain of antisocial, borderline, and narcissistic personality disorders.^[17] This domain includes traits such as grandiosity, hostility, and attention-seeking.^[16]
• Detachment: A pattern of withdrawing from social interactions, emotional expression, and pleasurable experiences, leading to limited interpersonal connections and restricted affect.^[16] This is a core trait domain of avoidant, obsessive–compulsive and schizotypal personality disorders.^[17] This domain includes traits such as anhedonia, social withdrawal, and suspiciousness.^[16]
• Disinhibition: Inclination towards immediate rewards being given priority over long-term planning, resulting in impulsive actions with little regard for consequences.^[16] This is a core trait domain of antisocial and borderline personality disorders.^[17] Rigid perfectionism, a trait of obsessive–compulsive personality disorder, is considered a pathological trait; its absence is associated with disinhibition.^[16] This domain includes traits such as irresponsibility and risk taking.^[16]
• Negative affectivity: A disposition toward frequent and intense negative emotions, such as anxiety, sadness, anger, or shame, often leading to emotional instability, excessive worry, and difficulty managing distress.^[16] This is a core trait domain of avoidant, borderline, and obsessive–compulsive personality disorders.^[17] This domain includes traits such as emotional lability, submissiveness, and a pattern of anger and viciousness.^[16]
• Psychoticism: A pattern of unusual perceptions, thoughts, and behaviors that may seem eccentric or disconnected from shared social and cultural expectations.^[16] This is a core trait domain only of schizotypal personality disorder.^[17] This domain includes traits such as dissociation, depersonalisation and derealisation, as well as eccentric beliefs and odd speech.^[16]

The trait domains in the AMPD were developed to be consistent with the more broadly used Five Factor Model of personality, but with focus on the mal-adaptive ends of each of these personality factor spectra.^[18][19] The relationship between the Five Factor Model's openness and the AMPD's psychoticism has been subject to dispute. Studies on the topic have utilized different measures and definitions of the two concepts and have given mixed results, with some studies showing little to no connection while other studies report a weak correlation. Conceptualizations of openness differ from each other, with some focusing on traits such as self-actualization and open-mindedness, whereas others align more closely with schizotypal tendencies, the latter resulting in higher correlations with psychoticism.^[19]

A point of criticism leveled against the AMPD is that the model fails to define the amplitude of a trait required for its inclusion in the basis for a diagnosis to be made.^[20]

Other criteria

Criterion C: Pervasiveness & Criterion D: Stability

The alternative model requires that the impairments in the functioning of the personality are "relatively inflexible" and significantly affect the subject's overall functioning across time and in different types of situations. As such, these impairments cannot be only situational, and the pathology persists despite its patterns being mal-adaptive. Impacts of the pathology affect the subject's perceptions, thoughts, emotions, as well as way of relating to others and to self. Thus, the subject has trouble functioning in a multitude of important areas of life.^[21]

Criteria E, F, G: Differential diagnosis

See also: Differential diagnosis

The condition must not be attributable to other medical (criterion F) or psychiatric (criterion E) conditions, substances (criterion F), and the impairment in personality functioning and the expression of personality traits cannot be sufficiently accounted for by typical developmental stages or the individual's sociocultural background (criterion G).^[21] Criterion E requires a differential diagnosis in order to verify that personality disorder is the appropriate diagnosis. Other conditions, such as bipolar disorder^[22] and autism spectrum disorder^[23] can be difficult to differentiate from a personality disorder. At the same time, the AMPD manual states that "patients with other mental disorders should be assessed for co-morbid personality disorders",^[21] because there is no dichotomy between personality disorders and having another mental disorder, and because the presence of personality disorder is relevant to the clinical picture.^[21]

Diagnoses

Specific personality disorders

The six specific personality disorders included in the AMPD are based on PDs as described in the DSM-IV, though conceptualized through functioning level and pathological traits.^[6] Each of these PDs has its own short description, which captures the overall features of the respective disorder. For example, avoidant personality disorder (AvPD) is described in the following manner:

"Typical features of avoidant personality disorder are avoidance of social situations and inhibition in interpersonal relationships related to feelings of ineptitude and inadequacy, anxious preoccupation with negative evaluation and rejection, and fears of ridicule or embarrassment."^[24]

In the sections for the specific personality disorders, the general description is followed by^[17] descriptions of the characteristic manner in which the specific personality disorder impacts each of the elements of personality functioning,^[25] as well as descriptions of how mal-adaptive traits characteristic of the disorder commonly manifest themselves as part thereof. A certain number of impairments in elements of personality functioning, as well as of pathological personality traits, has to be fulfilled in order for the criteria to be met.^[17]

In addition to a description and criteria specific to the personality disorder, every specific personality disorder has a section dedicated to specifiers,^[17] i.e. traits which can be specified in addition to the required ones if they "may have clinical relevance".^[26] One example of this is narcissistic personality disorder (NPD) having suggested specifiers of additional antagonistic traits as well as traits of negative affectivity for malignant and vulnerable manifestations of NPD, respectively.^[27] The specificity of suggested specifiers varies between disorders; e.g. whereas suggestions concerning NPD are somewhat more specific, the AMPD states that non-required traits vary substantially in the case of AvPD.^[24] Level of personality functioning may optionally be recorded for any of the disorders.^[17]

Personality disorder - trait specified

The AMPD also contains a diagnostic category called Personality Disorder - Trait Specified (PD-TS). This diagnosis consists of pathology within at least two elements of personality functioning, as well as of at least one domain of pathological traits. PD-TS thus includes personality pathologies which do not align well with the established specific diagnoses in the AMPD.^[28] A study conducted among a sample of outpatients as well as non-patients deemed high-risk showed that slightly more than a fifth of people who met criterion A for personality disorder fit a specific personality disorder well, meaning that a great majority of people who are eligible for a diagnosis of personality disorder have a clinical picture better captured by PD-TS, unless diagnosed with several specific personality disorders and/or specifiers to those disorders, in case the criteria for at least one specific PD are met.^[29]

Both the categorical standard model and, to a lesser extent,^[29] the specific PD diagnoses in the AMPD share the weakness of being polythetic in nature; a diagnostic category may require only a subset of the total amount of criteria to be fulfilled in order for a diagnosis to be made. For example, the standard model requires that five out of nine criteria be met for a borderline personality disorder diagnosis,^[30] which means that two individuals diagnosed with it could exhibit substantially different symptom profiles, with only one overlapping criterion^[31] and 256 possible combinations of criteria met.^[32] The aforementioned problem is ameliorated by the introduction of the PD-TS diagnosis, which is specific with regard to the traits present in a subject.^[29]

Applications and usefulness

Comparison with the standard model

Although the AMPD has been criticized for possibly complicating communication and classification of personality disorder(s), due to its dimensional nature comprising a multitude of factors, a diagnosis based in a dimensional model may prove more useful because it provides a descriptive rather than purely categorical diagnosis.^[33] As such, an AMPD diagnosis has advantages such as giving a more in-depth description of the nature of an individual's mental impairment and suffering, in addition to documenting both the presence of a personality disorder and the specific ways in which it manifests,^[33] with the specified PD categories found in the AMPD maintaining a degree of compatibility with the standard model.^[33][2] This compatibility has been demonstrated through correlations between personality disorders in the AMPD and their standard model counterparts being strong^[34] for all AMPD PDs except obsessive–compulsive and schizotypal PDs, which had moderate^[34] correlations of 0.57 and 0.63, respectively;^[11] this was suggested to "likely to be as high as [the agreement] between two diagnosticians on DSM-IV (and now DSM-5 Section II) diagnoses".^[11]

Contrary to the aforementioned concern regarding possible complexities introduced by the AMPD, ease of use has also been described as being an advantage pertaining to the alternative model.^[11] In addition to this, although the standard model – in this case the one in the DSM-IV-TR, which is the same as the standard model in section II of the DSM-5^[35] – has been described as useful and easy to use by professionals, a study has shown that in the majority of domains measured, professionals rated the AMPD to be at least equal to or even superior to the standard model. The standard model was not evidently perceived as being superior to the AMPD, whereas there is significant evidence for the alternative model being perceived by professionals as having multiple advantages compared to the standard model.^[36]

Clinical utility and treatment relevance

An example of the AMPD being useful in practice is its use to assess one patient by several different treatment providers at different points in time. In this case, changes in the level of personality functioning, assessed in accordance with the LPFS, have and can be noted for inclusion in the basis for treatment.^[37] Research suggests that clinicians tend to provide consistent assessments of both personality functioning and mal-adaptive traits.^[15] While the level of personality functioning may vary across time, subject to treatment, research supports that mal-adaptive traits tend to be more consistent.^[37] While the benefit has not been as clear in the case of criterion A, clinicians have noted several benefits to the AMPD's criterion B, such as its ability to capture an individual's personal set of trait pathology, which is useful both for communication with the patient and for documenting an individual's clinical picture more comprehensively.^[1]

Although there is limited research and scientific knowledge regarding the relationship between AMPD criterion B pathological traits and treatment decision-making as well as the results thereof,^[1] there is some evidence indicating that negative affectivity, by virtue of being related to neuroticism, could predict SSRI-type^[38] antidepressants having an ameliorating effect.^[1] The AMPD has additionally undergone a trial in clinical settings as part of a psychological assessment conducted on patients before they undergo bariatric surgery, which has shown the AMPD's assessment of mal-adaptive personality traits to hold significance in the pre-surgery psychological assessment.^[37] There is empirical support for Five Factor Model traits – to which AMPD criterion B traits are related^[19] – being useful when deciding on treatment approaches.^[39]

Criticism and reception

Since being published in the DSM-5, the Alternative DSM-5 Model for Personality Disorders has given rise to, and been subject to, a multitude of research with the purpose of proving or disproving the AMPD's validity and usefulness in clinical settings, as well as how it compares to other models, such as the standard model in the DSM-5. The results of the research have largely supported the model's overall validity.^[37][15]

Regarding the main criteria (A and B) of the model, research has found a strong correlation between results from assessment regarding level of personality functioning, and mal-adaptive personality traits, respectively, which makes it difficult to separate in which way each of these influences the personality disorder.^[15]

The AMPD has additionally received criticism for the decision not to include four of the distinct diagnostic categories found in the standard model, which was not supported by a broad consensus of professionals at the time. Thus, an online survey was conducted among members of the Association for Research on Personality Disorders and the International Society for the Study of Personality Disorders to assess perceptions of the utility and validity of the DSM-IV-TR personality disorder categories,^[3] which are the same as the categories found in the DSM-5 standard model.^[35] The results indicated that most respondents regarded all personality disorders, particularly antisocial and borderline personality disorders, as valid. The only diagnosis regarding the inclusion of which the majority of respondents were not clearly supportive, was histrionic personality disorder. Likewise, most participants opposed the removal of any personality disorder from the classification system, with no disorder receiving majority support for deletion.^[3]

See also

• Psychiatry portal
• Psychology portal

• Global Assessment of Functioning – Scale to rate how well one is meeting various problems in living
• List of diagnostic classification and rating scales used in psychiatry
• Revised NEO Personality Inventory – Big Five personality trait inventory
• Structured Clinical Interview for DSM – Psychiatric diagnostic interview protocol