Martin Blay1,2*
Inès Benmakhlouf2Marion Zenou2Mélissa Amate2Arjin Uludag-Martin2François Chevrier3Marie-Laure Pomey3Marie Brisebarre3Paul Lebourleux3Oyhana Lopepe3Julie Cacouault3Saioa Lagadec3
Loïs Choi-Kain4,5
Mario Speranza1,6- 1Université de Versailles Saint-Quentin-en-Yvelines (UVSQ), Inserm, Centre de recherche en Epidémiologie et Santé des Populations (CESP), Université Paris-Saclay, Villejuif, France
- 2ADDIPSY, Addictology and Psychiatry Outpatient Center, Santé Basque Développement Group, Lyon, France
- 3Clinique Château Caradoc, Santé Basque Développement Group, Bayonne, France
- 4Department of Psychiatry, Harvard Medical School, Boston, MA, United States
- 5Gunderson Personality Disorder Institute, McLean Hospital, Belmont, MA, United States
- 6Versailles Hospital Center, University Department of Child and Adolescent Psychiatry, Le Chesnay, France
Personality disorders (PD) are common and debilitating psychiatric conditions, often characterized by severe interpersonal and self-dysfunction. Borderline personality disorder (BPD) is the most studied PD, with treatments like Good Psychiatric Management (GPM) demonstrating effectiveness. To address the current state of the personality disorder field, where most evidence-based treatments have been developed using a categorical approach, while the most empirically supported approach is the dimensional one, we developed an adaptation of GPM (GPM-extended) integrating concepts of dimensional personality dysfunction into an established therapeutic framework. This prospective “here-there” study will compare GPM-extended with standard GPM in two outpatient facilities. This methodological choice is due to organizational and institutional limitations. Participants (≥18 years) meeting BPD criteria per the SCID-II will be included. The GPM-extended group incorporates a dimensional diagnostic framework focusing on three personality dilemmas (rejection/abandonment fears, self-esteem dysregulation, and perfection/control issues), with tailored psychoeducation and treatment priorities. The standard GPM group will follow the validated GPM protocol, emphasizing interpersonal hypersensitivity and standard psychoeducation. Both groups will receive weekly individual therapy and group interventions for one year. Between-group differences at 1-year follow-up will be assessed using multiple regression models adjusted for baseline scores. The primary outcome is the change in BPD symptom severity (ZAN-BPD). Secondary outcomes include personality functioning, traits, and various clinical dimensions such as impulsivity, emotional regulation, and social functioning. All patients having initiated the program and having initial data will be included in the analysis, with multiple imputations for missing data and sensitivity analyses planned. Altogether, this study seeks to evaluate the feasibility and efficacy of integrating dimensionality into GPM, offering a pragmatic pathway to improve PD treatment and bridging gaps between evidence-based conceptualizations and treatments of PD.
Clinical trial registration: ClinicalTrials.gov, identifier NCT06913738.
1 Introduction
Personality disorders (PD) are among the most common psychiatric disorders. They affect around 10% of the general population, and the proportion of patients with at least one personality disorder reaches 50% in psychiatric settings (1). These disorders are associated with an increased risk for somatic comorbidities (e.g., (2)), psychiatric comorbidities (e.g., (3)) with a lower remission rate (e.g., (4)) and a lower efficacy of their treatment (e.g., (5)). Moreover, these disorders are associated with large consequences in term of social and professional functioning (6, 7) and in terms of societal costs (8, 9). Finally, personality disorders are also associated with stigmatization, both in general population and among health care professionals (10).
The most studied personality disorder is borderline personality disorder (BPD). BPD affects around 1.6% of the general population and is characterized by hypersensitivity to rejection and abandonment, resulting in significant difficulties in interpersonal relationships, emotion regulation and identity (11). Since its inclusion in the DSM-III (12), the BPD diagnosis has allowed the development of a large amount of research that shed light on the importance to consider, diagnose, and possibly treat personality disorders (11, 13), that were often considered as rigid, chronic and less important disorders compared to main disorders like major depressive or bipolar disorders (14). Indeed, major longitudinal studies have shown that BPD could improve over time (6), and evidence-based effective treatments have been developed over the last decades, each linked with a specific treatment conceptualization and framework, and having the overall same efficacy (15). The main therapies actually recommended are dialectical behavioral therapy (DBT), transference-focused psychotherapy, mentalization-based therapy (MBT) and schema-focused therapy (16). To note, these therapies are the most developed currently in the entire personality disorders field, as very few models have been developed for other PD categories. However, as these specialized models developed, an issue quickly emerged from a public health point of view, namely, the lack of accessibility and availability of these treatments for most BPD patients. Thus, given that these therapies are highly specialized and poorly available (17), to overcome this issue clinicians have developed new models, less complex and specialized but more accessible and practicable. These models are based on what is thought to be the core characteristics of effective treatments for borderline patients, and have been popularized through the dissemination of two mains models: Good Psychiatric Management (GPM) and Structured Clinical Management (18).
However, despite this robust clinical and scientific foundation supporting the diagnosis of BPD, ongoing debates persist about its reliability, along with that of other traditional personality disorder categories. Indeed, these diagnoses are often the target of critics (19, 20), because they are thought to have low reliability, with high level of diagnostic comorbidity (most patients have often more than one PD) and within-disorder heterogeneity (patients with the same PD can have different clinical issues) (14, 19). In this context, researchers have developed the so-called “dimensional” models, describing personality disorders using the combination of two criteria: level of personality functioning (criteria A) and personality traits (criteria B). The two main models currently used are the Alternative Model of Personality Disorders of the DSM-5 (AMPD) (21) and the personality disorders model of the International Classification of Disease 11th version (ICD-11) (22). More precisely, criteria A is divided in two main dimensions, self and interpersonal functioning, and criteria B is composed of 5 personality traits, mostly derived from the Big-5 literature: negative affectivity, antagonism, disinhibition, detachment, and psychoticism (for the AMPD)/anankastia (for the ICD-11)). However, these new models are currently only scarcely used in clinical services (23). Moreover, they are also not free of critics, particularly with regard to the clinical relevance of personality traits, considered by some authors to not represent the complex and dynamic self- and interpersonal processes characteristics of PD patients (24). Our hypothesis is that these two issues may be linked, as clinicians may be more reticent to use new models when the latter are not sufficient to describe what they see in their daily practice. Thus, these models need to be enhanced, possibly using more comprehensive and relevant conceptualizations of PDs patients’ daily issues, which may help the transition from classifications to clinical practice. Such adaptation could be even more important given that these models have not been used in the development and assessment of evidence-based treatments (contrary to classic categorical models), which furthermore limits their usefulness, even if new models are being developed trying to fill this gap (25).
We believe that GPM could be useful to solve this issue, both in terms of diagnostic assessment and treatment. We discussed the diagnostic assessment aspect in a recently published case report (24). Briefly, GPM, whether we consider the BPD model or its adaptation for narcissistic (NPD) (26) and obsessive-compulsive personality disorders (OCPD) (27), has allowed the development of dynamic and fundamentally interpersonal models aiming to better describe the borderline (interpersonal hypersensitivity model), narcissistic (intrapsychic coherence model), or obsessive-compulsive (model of overcontrol) patient’ psychopathology than classic categories or personality traits. Notably, these three conceptualizations encompass the main dilemmas an individual with PD can face in his/her pathology: fear of rejection and abandonment (for BPD), self-esteem dysregulation (for NPD), and control dependency (for OCPD). Based on these assumptions, we previously described how incorporating these three conceptualizations by describing them as non-exclusive ways personality can dysfunction in one individual, rather than restricting them to one category, may be useful to offer a more relatable conceptualization than classic personality traits, while being a more dimensional and less rigid way of assessing personality disorders (24).
However, the contribution of GPM may not be limited only to diagnostic assessment part but may also encompass the therapeutic dimension. Recent research have shown that BPD criteria may be a relevant marker of personality functioning, even if questions remain about whether this applies to specific BPD criteria, all criteria collectively, or the overall severity of BPD symptomatology (28, 29) Thus, if personality dysfunction is a core aspect of every personality disorder, and if BPD criteria are relevant markers of personality dysfunction, one could legitimately infer that all personality disorders may be treated by partly the same psychotherapeutic content, i.e., one developed and shown useful to treat patient fulfilling criteria for BPD, like GPM (30, 31). This makes even more sense when considering the improvement in interpersonal functioning – one hallmark of personality functioning - in GPM trials, and the overall focus on getting a life outside of treatment to build an integrated sense of identity – another hallmark of personality functioning - in the GPM treatment framework (32). Finally, even if the three adaptations of GPM consider different personality disorders and are partly different, a large part of the content remains the same: use of diagnostic disclosure, psychoeducation, case management, progress, psychodynamic and behavioral treatment, multimodality, safety planning, symptomatic medication management … This makes a dimensional GPM adaptation feasible beside being theoretically useful, with 1°) this common part being the cornerstone of the adaptation, and 2°) the specific parts of each dilemma being provided depending on the patient’s profile.
Altogether, like classic dimensional models, GPM offers the possibility for dimensional approach, with personality functioning assessed with the presence/absence of BPD criteria, and characteristics of self- and interpersonal dysfunction considered globally using GPM’s dilemma-based approach. However, unlike classic dimensional models, GPM has been empirically tested and found efficient to treat patients fulfilling criteria for BPD, and provides a more simple, easy-to-understand, psychoeducation-accessible, and daily-experience related approach of personality characteristics than classic dimensional models. In this context, we designed a dimensional adaptation of GPM (GPM-extended) for patients fulfilling criteria for BPD. In the present paper, we will describe the protocol of a pragmatic, real-life based, “here-there” study assessing the effectiveness of an outpatient treatment program based on GPM-extended compared to an outpatient treatment program based on classic GPM.
It is important to note that, in the present study, GPM is used not only as generalist treatment, but also as a broad clinical framework that organizes and structures treatment for patients with BPD. While GPM was initially designed as a pragmatic and accessible intervention for BPD, its foundations (diagnostic clarity, psychoeducation, case management, focus on real-life functioning…) naturally lend themselves to a modular approach in which additional therapeutic elements can be integrated when clinically indicated. Within this perspective, GPM provides the conceptual and organizational backbone of care, and complementary interventions (like psychoeducation or other evidence-based groups) are not considered as separate treatments, but rather as components selected and shaped by the GPM framework. This conception is central to the rationale of the present study, which compares a classic GPM program with a dimensional GPM-based adaptation that retains the same foundations while tailoring its components to patients’ core personality dilemmas.
2 Methods and analysis
2.1 Design and settings
Our study will use a “here-there” methodology, i.e., the inclusion of patients will be made based on which facility each patient is referred to. Two facilities will be included, one offering the outpatient GPM-extended program (ADDIPSY, Lyon, France), and the other offering the outpatient classic GPM program as it has been validated in the literature (Clinique CARADOC, Bayonne, France). Both facilities are private day-hospitals, offering outpatient programs for a wide range of psychiatric disorders. 2 psychiatrists and 3 psychologists (in ADDIPSY), and 3 psychiatrists and 3 psychologists (in CARADOC) will take part as therapists in the current study. All the psychometric evaluation, data collection and entry will be conducted by two trained clinical research assistants (IB and SL).
The estimated inclusion period for the present study is five years. Recruitment started on June 19th 2024 and is expected to be completed by June 2029. Each participant’s involvement lasts for one year, determined by the intervention period, which was established based on research standards for studying the effectiveness of psychotherapies in BPD. Data collection is anticipated to be completed by December 2029. The final results are expected to be available for publication by the end of 2030. Total duration of the study is estimated at six years and six months. However, the study may conclude earlier if the required sample size is reached ahead of schedule. The full SPIRIT schedule of enrolment, interventions, and assessments can be found in Table 1.
Table 1. SPIRIT schedule of enrolment, interventions, and assessments.
2.2 Objectives and hypotheses
The primary objective of this study will be to evaluate the efficacy of GPM-extended compared to classic GPM in improving the overall level of BPD symptoms in patients meeting the criteria for the latter. Our secondary objectives will be to compare the efficacy on other dimensions of interests, including sub-dimensions of BPD (i.e., affective, cognitive, impulsive, interpersonal), personality functioning and intensity of personality traits (according to the alternative DSM-5 model), level of disability and social functioning, narcissistic and obsessive-compulsive symptomatology.
Overall, we expect an improvement in all the outcomes in both groups, notably in borderline symptoms and level of personality functioning, at 1 year of treatment. We also expect a significant difference in both groups, with GPM-extended being superior to classic GPM, notably on borderline, narcissistic, obsessive-compulsive symptoms and overall personality functioning.
2.3 Selection criteria
The inclusion criteria for the present study will be 1°) being >18 years old, 2°) having a diagnosis of BPD made using the Structured Clinical Interview DSM–IV Axis II Personality Disorders (SCID-II (33)), 3°) having signed an informed consent and 4°) being affiliated with or beneficiary of the French social security system. On the other hand, exclusion criteria will be 1°) being < 18 years, 2°) having a comorbid psychotic disorder, intellectual deficiency, severe antisocial features, major substance use disorder making intensive psychotherapy impossible, anorexia nervosa with somatic risk, bipolar disorder in acute manic phase, 3°) being under protective measure, 4°) being unable to cooperate or complete self- or hetero-questionnaire and 5°) subject not affiliated with or non-beneficiary of a French social security system.
2.4 Ethics and dissemination
Written informed consent will be gathered for each patient by the clinical research assistants and the referent psychiatrists in each facility. This study protocol was submitted and validated by the Comité de Protection des Personnes Ile de France III (N°ID-RCB = 2024-A00475-42). Moreover, this study has been registered on ClinicalTrials.gov under the identifier NCT06913738 (first posted on April 6th 2025). To note, even though this study meets the WHO definition of a clinical trial, it was not registered before the inclusions began because the authors were unaware that registration was required. The authors confirm that all ongoing and related trials for this intervention are registered. Finally, all substantial modifications to the protocol (e.g., changes in eligibility criteria, outcomes, or analytical strategy) will be submitted by the coordinating investigator to the sponsor’s representative for approval. Following a favorable opinion, the sponsor will initiate the necessary regulatory procedures to obtain approval from the relevant ethics committee and/or competent authorities. Relevant parties, including investigators, regulatory bodies, trial registries, and journals, will be informed of approved amendments as required. A manuscript with the results of the primary study will be published in a peer-reviewed journal. If conducted, manuscripts reporting the results of ancillary analyses (e.g., mediation analyses) will also be submitted for publication in peer-reviewed journals. To note, for each manuscript, raw data and code for reconducting statistical analysis will be available upon request to the corresponding author (MB).
2.5 Key features of GPM-extended
Given that the present paper aims to present the protocol of our study, we will not describe all the characteristics and detail of our GPM-extended adaptation. A GPM-extended manual has been written for clinicians participating in the present study and is available upon request to the corresponding author. In this paragraph, we will rather discuss the specificity of our adaptation compared to classic GPM treatment (32).
GPM-extended builds on the foundational principles of GPM for BPD by incorporating a more dimensional and pragmatic approach. It emphasizes diagnostic clarity, treatment prioritization, personalized psychoeducation, and multimodal interventions tailored to severe personality dysfunctions. The diagnostic process focuses on BPD criteria and integrates dimensional tools like the Level of Personality Functioning Scale – Brief Form (LPFS-BF) (34) to assess personality functioning. This process is enhanced using the three main dilemmas a patient with severe personality dysfunction may face according to GPM (i.e., fear of rejection and abandonment, self-esteem dysregulation, and control dependency), to assess how one’s personality may dysfunction. The assessment of the presence/absence and relevance of each dilemma for each patient relies on both clinical and psychometric investigation, associated with repeated team discussion in the weekly supervision sessions. Clinical investigation uses DSM’s Section II and Section III specific criteria, and psychometric investigation uses the newly developed GPM Dilemmas Assessment Scale (GPM-DAS, Table 2) as well as other specific psychometric scales (i.e., Zanarini Rating Scale for Borderline Personality Disorder (ZAN-BPD), Pathological Narcissism Inventory Brief Form (PNI-B) and Pathological Obsessive-Compulsive Personality Scale (POPS). The complete dilemma assessment process can be found in Table 3.
Table 2. Content of the GPM-dilemmas assessment scale (GPM-DAS).
Table 3. Dilemma assessment process.
Once the main dilemmas are identified, both the clinician and the patient work on a clear prioritization of the treatment targets, with the most impacting dilemmas being the first one to focus on. Indeed, GPM-extended tailors psychoeducation and case management to each patient’s specificities. This prioritization is often reassessed by the patient and the clinician throughout the treatment, to ensure that treatment addresses the most impactful issues while maintaining flexibility to adapt to evolving needs.
Regarding treatment content, GPM-extended retains the core principles of GPM, such as active support, validation, focus on real-life situations, and fostering patient responsibility for change (32). It also retains the core components of a GPM treatment (i.e., alliance building, goal setting and achievement, anticipation of difficulties, and narrative work), and emphasizes the importance of group interventions (supportive group, psychoeducation, or specialized approaches like DBT or MBT) and pragmatic pharmacological management. However, it expands upon these characteristics with a dimensional approach that emphasizes individualized treatment focusing on the patient’s dominant personality dilemmas identified in the diagnostic process. Indeed, alliance building challenges, narrative work, and treatment main targets are different depending on the dilemmas. For example, in the case of borderline symptomatology, treatment focuses on building a meaningful life (e.g., stable social and professional roles) over intense relationships; in the case of narcissistic symptomatology, treatment focuses on the development of an integrated sense of self-worth by addressing grandiosity, shame, and avoidance challenges; and, in the case of obsessive-compulsive symptomatology, treatment focuses on improving control dependency, perfectionism, and rigid standards and fears through emotional and behavioral corrective experiences.
2.6 Interventions
Both treatment programs have a 1-year length, as in most standard clinical studies on psychotherapy in BPD. Patients that require it will be offered a second year of treatment if it is clinically relevant. Patients refusing to participate in the present study will be treated just as the participants but will not undergo the protocol-related psychometric evaluation. A schematic representation of the study and programs content can be found in Figure 1. To note, given that both participating facilities provide a wide range of general psychiatric care (including individual and group therapies, such as trauma-focused or addiction-focused treatments), participants are allowed to engage in these additional interventions, as long as they are not specifically designed to target borderline personality disorder or related personality dysfunction.
Figure 1. Schematic representation of the study groups.
2.6.1 Classic GPM group
The outpatient classic GPM group will consist in the association of both individual and group therapy:
- Individual = 1°) 1h weekly individual psychotherapeutic session conducted by a therapist trained in GPM, with a focus on interpersonal hypersensitivity and 2°) 30 min every 3 weeks of an individual psychiatric case management focused on medication stabilization by a psychiatrist trained in GPM.
- Group = 1°) Alongside the beginning of therapy, every patient will undergo >= 1 session of a 6-weeks group psychoeducation on BPD and 2°) Patients with high levels of emotion dysregulation (diagnosed when baseline Difficulties in Emotion Regulation Scale (DERS) total score is higher than the mean scores obtained in clinical populations, i.e. > 90 (Hallion et al., 2018)) will undergo >= 1 session of a 18-weeks DBT skills training group (either in one block or sub-divised in 3 blocks of 6 weeks each, corresponding to the main modules of DBT – mindfulness/distress tolerance, interpersonal efficacy and emotion regulation) during the year of therapy. To note, the exact moment when patients will be included will depend on the availability of skills groups).
The team will be trained and supervised (1 hour every week – online – and 1 day every month – in person) throughout the process by the main investigator of the study (MB). The latter will not practice as a therapist in the control group and will only serve as trainer and supervisor. It is important to note that the team did not have experience with GPM before the beginning of the project. All the content in this control group is manualized and validated, including the BPD psychoeducation program (35) and the DBT skills training program (36). Moreover, every clinician will be provided with the Handbook of Good Psychiatric Management for BPD for the conduction of individual psychotherapy (32).
2.6.2 GPM-extended group
The outpatient GPM-extended group will also consist in the association of both individual and group therapy:
- Individual = 1°) 1h weekly individual psychotherapeutic sessions conducted by a therapist trained in GPM, with a specialized focus on the main triggers/dilemmas identified with the patient and 2°) 30 min every 3 weeks of an individual psychiatric case management focused on medication stabilization by a psychiatrist trained in GPM.
- Group = 1°) Alongside the beginning of therapy, every patient will undergo >= 1 session of one or several of the 6-weeks group psychoeducation programs, depending on the main dilemmas of the patient (borderline, narcissistic and/or obsessive-compulsive, with each patient undergoing from 1 to 3 different programs, depending on the results at the GPM-DAS) and 2°) Patients with high levels of emotion dysregulation (diagnosed when baseline DERS-F total score is higher than the mean scores obtained in clinical populations, i.e. > 90 (Hallion et al., 2018)) will undergo >= 1 session of a 18-weeks DBT skills training group (either in one block or sub-divised in 3 blocks of 6 weeks each, corresponding to the main modules of DBT) during the year of therapy. Once again, the exact moment when patients will be included will depend on the availability of skills groups.
The team will be trained and supervised (1 hour every week – in person) throughout the process by the main investigator of the study (MB). The latter will practice as a therapist in the active group and will also be the trainer and supervisor. It is also important to note that, contrary to the classic GPM team, this team did have one year of experience with GPM before the beginning of the project. This discrepancy and its consequences will be discussed below in the Limitations section. Regarding the content, a GPM-extended manual will be provided for clinicians, including the specific content of therapy for each core triggers/dilemmas. These core triggers/dilemmas will be identified with the patient using both classic psychometric tools (e.g., PNI-B, POPS) and our newly developed GPM-DAS, designed to assess the three main dilemmas considered in GPM models. Moreover, the BPD psychoeducation program and the DBT skills training program will be the same as those used in the “classic GPM” group. The psychoeducation program for NPD is also manualized and has recently been published (37), and the psychoeducation program for OCPD has also been published online recently (38).
2.7 Measures
2.7.1 Psychometric assessment
Every measure will be assessed by dedicated clinical research assistants. The primary investigator won’t be involved in any psychometric assessment. Included patients will be administered, at the inclusion and 1 year follow-up several semi-structured interview and self-reported questionnaires, including socio-demographic characteristics and prescribed treatment information. Moreover, included patients will also undertake several self-report questionnaires at 4 and 8 months of therapy, for further mediation analyses (cf. 3.6). All the different measurements carried out at each stage of the EPSYLIA project can be found in Table 1.
Clinician-rated assessment. To assess categorical personality disorders, we will use the SCID-II (33, 39). The SCID-II has great inter-rater reliability and internal consistency (40, 41). Criteria will be rated if pathological, pervasive, and persistent, and we will use the recommended thresholds to establish diagnoses (e.g., for BPD, >= 5 out 9 criteria).
To assess other psychiatric disorders, we will use the Mini-international neuropsychiatric interview 7th version (MINI) (42). The MINI is a valid and reliable tools, with good psychometric properties (43), that assesses a wide range of past and present neuropsychiatric disorders, including depressive, bipolar, psychotic, anxiety, obsessive-compulsive, post-traumatic stress, substance use and eating disorders.
Finally, to have a clinician-rated measure of BPD symptoms intensity as a primary outcome, we will use the ZAN-BPD (44). Overall, the ZAN-BPD has good internal consistency (Cronbach’s α=0.85). It is a 9-items scale used to assess each of the nine criteria for BPD on a five-point anchored rating scale of 0 to 4. A total score (sum of all items) and 4 sub-scores (affective, cognitive, impulsive, and interpersonal (35)) can be calculated. To note, the ZAN-BPD was chosen for the primary outcome because classic GPM has been primarily evaluated as a treatment for BPD. Therefore, given that this study aims to compare an adaptation of GPM to classic GPM, we believe that a validated measure of BPD symptoms, like the ZAN-BPD, was required for the primary outcome. Moreover, only a small amount of trials used the dimensional framework as a primary outcome (45); thus, choosing a BPD measure as a primary outcome will allow for comparisons with the existing literature on psychotherapy efficacy in BPD. Finally, given that BPD symptom severity is strongly associated with the overall severity of personality dysfunction (28), changes in ZAN-BPD scores can serve as a proxy for improvements in personality functioning. However, we acknowledge that this choice has important limitations that will be discussed below.
Self-report personality assessment tools. To assess GPM dilemmas, we will use the GPM-DAS), which was specifically developed for this study. The GPM-DAS is an interviewer-rated 3-items scale, with each item assessing the presence or relevance of one dilemma— fear of rejection and abandonment, self-esteem dysregulation, or control dependency —on a 10-point Likert scale, over the last month. Each item is accompanied by a list of several examples illustrating how the dilemma may manifest in patients’ daily lives. The development of the GPM-DAS was based on a synthesis of the three major conceptual models of GPM (interpersonal hypersensitivity, intrapsychic coherence, and model of overcontrol) and informed by our clinical experience with these frameworks. Items were derived from clinically prototypical manifestations of each dilemma. reviewed by GPM experts and refined to retain those considered most representative and clinically informative. We opted for a 10-point Likert scale, and the order of items on the scale was carefully considered. Preliminary interpretative thresholds (0-2 (non-relevant), 3-5 (mildly relevant), 6-8 (moderately relevant), and 9-10 (strongly relevant)) were clinically determined to provide a nuanced understanding of the relevance of each dilemma. These thresholds will be further explored and validated within the context of this study, with future research aiming to refine their psychometric properties. These thresholds will also be used within the GPM-extended framework to help clinicians in prioritizing the most relevant dilemmas for each patient. It is important to note that the GPM-DAS was simultaneously developed in both French and English, ensuring broader applicability and facilitating its use across diverse clinical and research contexts. However, as this is a newly developed scale, it has not yet undergone formal validation. Even if the scale’s reliability and construct validity will be assessed within the present study, analyses using the GPM-DAS will be interpreted with great caution until further validation studies are conducted.
To assess personality functioning, we will use the LPFS-BF (34). The LPFS-BF has satisfactory internal consistency for total score (α = .69), and marginal to fair internal consistency for sub-scores (Self: α =.57; Interpersonal: α =.65). It is a 12-items scale used to assess personality functioning as described in the section III of the DSM-5 (21), with two sub-scales (self and interpersonal functioning), each item being rated on a 4-points Likert scale.
To assess personality traits, we will use the Personality Inventory for DSM-5—Brief Form (PID-5-BF) (46). The PID-5-BF is 25-items scale assessing the five personality traits of AMPD’s criterium B, (namely, negative affectivity, detachment, antagonism, disinhibition, and psychoticism), each item being ranked on a 4-points Likert scale, and has a satisfactory internal consistency (α ranging from.68 for antagonism to.81 for total score).
To assess self-reported borderline symptoms, we will use the Borderline Symptom List -23 items (BSL-23) (47). The BSL-23 assesses the global severity of BPD symptoms, each item being ranked on a 5-point Likert scale (from 0 (“never”) to 4 (“always”)) and has high internal consistency (α= 0.935-0.969).
To assess pathological narcissism, we will use the PNI-B (48, 49). The PNI-B is a 28-items scale assessing the two main facets of pathological narcissism (grandiosity and vulnerability), each item being ranked on a 5-points Likert Scale, and has great internal consistency (for grandiosity subscale: α=.83-.86; for vulnerability: α= .93). To our knowledge, no thresholds are currently available to distinguish narcissistic patients from clinical controls using the PNI-B, but mean PNI-B grandiosity and vulnerability scores found in general population in the original validation study were between 2.7 and 2.9 for grandiosity, and between 1.9 and 2.5 for vulnerability (49).
Finally, to assess obsessive-compulsive symptoms, we will use the POPS (50, 51). The POPS is a 49-items scale, each rated on a 6-point Likert scale, that assesses the overall severity and five facets of obsessive-compulsive personality (namely, difficulty with change, emotional over-control, rigidity, maladaptive perfectionism, and reluctance to delegate), with great internal consistency (for sub-scores: α=.89 to.93; for total score: α=.97) (50). To our knowledge, no thresholds are currently available to distinguish OCPD patients from clinical controls using the POPS, but the mean POPS total score found in general population in the original validation study was 163.13 (SD = 36.60) (51).
Other self-report assessment. To assess emotion regulation difficulties, we will use the DERS, in its original 36-items version (52). The DERS assesses emotion regulation difficulties in 6 different domains, ranging from awareness to acceptance of emotions or to ability to access to effective emotion regulation strategies. Each item is rated a 5-point Likert scale. The original DERS has great internal consistency (for sub-scores: α = .80 to.89; for total score: α = .93).
To assess impulsivity, we will use the original UPPS scale (53, 54). The UPPS is a 45-items scale assessing impulsivity on four dimensions (urgency, lack of premeditation and perseverance, and sensation seeking), each item being rated on 4-point Likert scale, and has good internal consistency (α from.77 to.83).
To assess entitlement, we will use the Psychological Entitlement Scale (PES) (55). The PES is a 9-items questionnaire assessing overall level of entitlement, each item being rated on a 7-points Likert scale, and has great internal consistency.
To assess alexithymia, we will use the 20-item version of the Toronto Alexithymia Scale (TAS-20 (56),). The TAS-20 assess alexithymia in three main dimensions (difficulty describing feelings, difficulty identifying feelings, externally oriented thinking), each item being rated on a 5-points Likert scale, and is a reliable and valid instrument with good psychometric properties (56).
To assess mentalizing abilities, we will use the Mentalization Scale (MentS) (57). The MentS scale is a 28-items scale assessing mentalizing abilities, with acceptable internal consistency in clinical samples (e.g., α = .75 for total score), and with three reliable subscales (self-mentalization, other-mentalization, and motivation to mentalize). Each item is rated on a 5-points Likert scale, and a total score can be calculated.
To assess non-suicidal self-injurious (NSSI) and suicidal behaviors, we will use the Suicidal Behaviors and Body Damage and Modifications Scale (SBBDM-S) (58), a 10-items self-report questionnaire recently developed (and currently not validated), that assesses the lifetime prevalence of suicidal behaviors, body modifications and NSSI.
To assess social functioning, we will use the Questionnaire de Fonctionnement Social (QFS) (59). The QFS is a 16-items questionnaire assessing the frequency and satisfaction of social behaviors, each item being rated on a 5-points Likert scale, and has good internal consistency (α = .65 -.83).
To assess disability and functional impairment, we will use the Sheehan Disability Scale (SDS) (60–62). In the SDS, the patient rates on a 10-point visual analog scale (graded by 5 levels described by words: not at all, slightly, moderately, very much, very much) the extent to which 3 domains of his/her life (work/school; social life and family life/family responsibilities) are impacted by his/her symptoms. The scores for the 3 domains can be summed to give an overall functional index ranging from 0 (no impairment) to 30 (severely impaired). The SDS has good psychometric properties, including internal consistency (e.g., for total score, α = .89).
To assess attachment styles, we will use the Experience in Close Relationship – Revised version (ECR-R) (63, 64). The ECR-R is a 36-items questionnaire with good psychometric properties assessing attachment anxiety and avoidance, each item being rated on a 7-point Likert scale (63).
To assess defense mechanisms, we will use the Defense Mechanism Rating Scales Self-Report (DMRS-SR) (65). The DMRS-SR is a 30-items questionnaire assessing the whole hierarchy of defense mechanisms, each item being rated on a 5-point Likert scale, and has great internal consistency (α =.75 to.90).
Finally, to assess working alliance, we will use the Working Alliance Inventory – Short Form Revised (WAI-SR) (66). The WAI-SR is 12-items questionnaires investigating three domains of therapeutic alliance (agreement between patient and therapist on 1°) the goals of the treatment and 2°) about the tasks to achieve these goals, and 3°) the quality of the bond between the patient and therapist), each item being rated on a 5-points Likert scale. The WAI-SR has high internal consistency (e.g., for total score: α = .95).
2.7.2 Fidelity assessment
The fidelity of the individual GPM intervention will be assessed by two independent certified GPM supervisors that will assess 2 videos of interview from each therapist involved in the study, one recorded in the first months of therapy and one in the last months. The supervisors will rate the interview using a scale developed based on the General Psychiatric Management Adherence Scale guidelines (67).
2.8 Statistical analysis plan
All the analyses will be conducted using R and R studio (68). The study’s principal investigator (MB) is responsible for drafting and implementing the statistical analysis plan.
2.8.1 Endpoints
The primary endpoint will be the difference in the overall intensity of BPD symptoms, as assessed by the ZAN-BPD, between the two groups after 1 year of treatment.
Secondary endpoints will be the difference in progression after 1 year of treatment on the various sub-scores of the ZAN-BPD (affective, cognitive, impulsive, interpersonal); the personality functioning, as measured by the LPFS-BF and its two sub-scores (self and interpersonal functioning); the intensity of the five personality traits of the alternative DSM-5 model, as measured by the PID-5-BF; the level of disability, as measured by the SDS and its three sub-scores (work/education, social life, family life); the level of social functioning, measured by the QFS and its two sub-scores (frequency and satisfaction); the number of suicide attempts and self-injurious behaviors, as measured by the SBBDMS; the self-reported intensity of borderline symptoms, as measured by the BSL-23; the intensity of narcissistic vulnerability and grandiosity, as measured by the PNI-B; the intensity of obsessive symptoms, as measured by the POPS and its five sub-scores (rigidity, excessive emotional control, inadequate perfectionism, reluctance to delegate, difficulty changing); the intensity of impulsivity, as measured by the UPPS and its four sub-scores (urgency, lack of perseverance, lack of premeditation, sensation seeking); the intensity emotional dysregulation, as measured by the DERS and its 6 sub-scores (lack of awareness of one’s emotions, lack of clarity about the nature of one’s emotions, lack of ability to engage in emotionally oriented activities); the intensity of alexithymia, as measured by the TAS-20 and its three sub-scores (description of feelings, identification of feelings, externally-oriented thoughts); the mentalizing abilities, as measured by the MentS and its three sub-scores (mentalizing self, mentalizing other, motivation to mentalize); the type of attachment styles, as measured by the ECR-R; the overall intensity of entitlement, as measured by the PES; the type of defense mechanisms, as measured by the DMRS-SR; the self-reported quality of therapeutic relationship, as measured by the WAI-SR and its sub-scores (goal, work, bond); the frequency of categorical personality disorder diagnoses, as measured by the SCID-II; and the frequency of current comorbidities, as measured by the MINI. In addition, differences in terms of partial response, complete response, and remission rates will be assessed, as will therapy discontinuation rates during the treatment year. To note, response and remission will be defined on the basis of the standards used by Ridolfi and colleagues (35): a “partial response” is defined as a decrease of > 4 but < 8 points in the total score on the ZAN-BPD scale, a “response” is defined as a decrease of > 8 points, and a “remission” is defined as moving from one of the higher stages of intensity to “mild symptoms” (with 1 < score < 9: mild symptoms; 10 < score < 18: moderate symptoms; 19 < score < +: severe symptoms).
2.8.2 Headcount calculations and patients included in the analysis
On the basis of several recent randomized controlled trials using the ZAN-BPD score as the primary outcome (69–71) and using their expected clinically significant difference and standard deviation, we estimated that 64 patients per group should be included to have 80% power to detect a clinically significant difference of 4 points (SD = 8) during the year of treatment, with a two-sided alpha risk of 5%. Assuming a lost to follow-up rate of around 20%, we estimated that a minimum of 77 patients should be included in each group. Finally, as the secondary analyses are exploratory, we have not planned any adjustment for multiple comparisons, and the statistical threshold retained is 0.05.
All patients having initiated the program and having initial data will be included in the analysis. If data are missing at 1 year, they will be imputed using multiple imputations. For this, we will use the R package “mice” (72). We also plan to perform sensitivity analyses, including an analysis of subjects with no missing data (“completers”).
2.8.3 Analyses
For the primary endpoint, we will use a multiple linear regression model, with the mean ZAN-BPD score at 1 year as dependent variable, and with a binary variable (taking 0 for the “classic GPM” group, and 1 for the “GPM-extended” group) and the baseline ZAN-BPD scale score as independent variables. Indeed, we want to adjust our model on the baseline ZAN-BPD scale score given that the main prognostic factor for treatment response in BPD patients is baseline symptom intensity (73). To note, other adjustment variables will be considered according to any inter-center differences detected on patient characteristics at baseline, to account for selection bias.
We will reproduce the same approach for all the secondary endpoints, namely, a regression model, taking the 1-year score of the variable as the dependent variable, and a binary variable (taking 0 for the “GPM classic” group, and 1 for the “GPM-extended” group) and baseline score of the variable (when available) as independent variables. The model will be adapted according to the type of dependent variable: linear regression model for a quantitative judgment criterion, logistic model for a binary dependent variable, quasi-Poisson and negative binomial model for a count. Adjustment variables will be added if relevant following the same rationale than for the primary endpoint.
Finally, we do plan to carry out exploratory analyses (including mediation) using data from all patients included, regardless of their group. These analyses will not, however, include a study of the influence of the type of therapy on the patient’s evolution. Rather, the aim will be to assess the factors influencing patient improvement during a GPM-based treatment, independently of the adaptation. This will not be the subject of the present study but rather of other targeted studies.
3 Discussion
In this article, we described the protocol we elaborated to evaluate our new dimensional adaptation of GPM (GPM-extended). In this discussion, we want to underline the potential benefits of our research alongside its possible methodological flaws.
Indeed, the EPSYLIA project, if carried out as planned, will produce new data regarding the efficacy of GPM-based intervention to treat patients fulfilling BPD criteria. This is all the more relevant given the small amount of research conducted on GPM efficacy, especially on long-term treatment, and the potential of diffusion this model has on a wide scale (74). Indeed, to date, only 5 randomized controlled trials have been conducted, one comparing DBT and GPM after 1 year of treatment (30, 31), two comparing a brief (10 sessions) individual version of GPM (still as a comparison treatment) to a personalized-treatment based on case formulation (75, 76), one compared the same brief individual version to treatment as usual (77), and one comparing a 6-week GPM-based psychoeducation group to a waiting list (35). All these studies found interesting results on the actual efficacy of GPM to improve borderline symptoms, as well as self and interpersonal functioning. However, only McMain’s study assessed the efficacy of a long-term GPM treatment. Thus, despite being observational, the EPSYLIA study could add further evidence on the efficacy of a long-term GPM-based treatment to treat patients with BPD.
On the other hand, we believe that our adaptation and the EPSYLIA project could also be a first step in solving the current conflicts at play in the personality disorders field. Indeed, as described in the introduction, classification of personality disorders is undergoing a paradigm shift from traditional diagnostic categories to dimensional systems. If these dimensional models are thought to represent a more valid approach of personality pathology, there are concerns regarding the deletion of the BPD category. Indeed, a large amount of research has been conducted among people who meet categorical BPD criteria, and, without the diagnosis, some experts fear that patients would lose access to specialized treatments for this condition, notably given the absence of treatment developed based on dimensional models (78). By building an in-between model, incorporating both the positive aspects of BPD category (e.g., existence of evidence-based treatment programs) and of dimensional models (e.g., consideration of the primary importance of self- and interpersonal functioning in personality assessment beyond diagnostic categories, and of within-disorder heterogeneity and high rates of comorbidity), we believe that the EPSYLIA project could be one of the first adaptation to try to pragmatically solve this issue.
Finally, beyond the comparison between classic GPM and GPM-extended, the present study will also invite a broader reflection on GPM itself. Rather than being limited to a generalist intervention, our study will provide valuable insights into how GPM can be also understood as a flexible organizing framework capable of supporting different levels of clinical complexity. In this view, GPM offers a coherent structure within which clinicians can either adopt a streamlined generalist approach or, when indicated, integrate additional therapeutic components drawn from other evidence-based treatment. What distinguishes this approach is not the specific techniques added, but the fact that they remain guided, prioritized, and contextualized through GPM principles and the patient’s individualized profile. Examining GPM in this enlarged function may help clarify how it fits into stepped-care approaches, where treatments must remain both scalable and responsive to heterogeneity in clinical presentations. Ultimately, such a perspective underlines the potential of GPM to connect accessibility and specialization, supporting a continuum of care that remains clinically coherent while adapting to the needs of diverse patients.
However, a large number of potential methodological issues must be underlined and will be considered during the study. A summary of potential biases and mitigation strategies can be found in Table 4. A first major risk of bias is the risk of selection bias linked to the “here-there” methodology, as patients treated in ADDIPSY may systematically differ from those treated in CARADOC for reasons unrelated to the intervention. Even though an adjustment for variables showing differences in baseline characteristics is planned, the risk of selection bias remains and will blurry the conclusions we will be able to draw from the present study. A second risk of bias is the risk of confounding bias lined to the fact that the primary investigator (MB) will have the triple role of therapist, trainer and supervisor in the “GPM-extended” group, in addition to double role of trainer and supervisor in the “classic GPM” group. The reasons for this multiplicity of roles are primarily organizational, as the primary investigator is a practicing psychiatrist in the structure where the “GPM-extended” group will be treated. We believe that there is an inherent risk of bias in being both therapist and trainer/supervisor, especially when considering that the primary investigator carries out this study aiming to show the superiority of GPM-extended over classic GPM. To mitigate this risk, we plan to have the “classic GPM” team complete the official online training of the Gunderson Personality Disorder Institute (79), an institute not directly involved in the study (even though Lois Choi-Kain act as a theoretical support in this study), alongside the training with the primary investigator. We also plan to regularly involve independent GPM, supervisors in the “classic GPM” team’s supervisions (approximately once every 1–2 months, depending on availability), to ensure that these supervisions are carried out properly. We believe that these elements will reduce the risk of bias in the training and supervision of the “classic GPM” team. Moreover, the PI will not take place in any psychometric assessment, and we also plan an independent fidelity assessment of individual GPM interventions. To note, treatment fidelity will only be assessed on a small number of recorded sessions due to organizational constraints. Even though these ratings will help ensure that gross deviations from the GPM model are detected, they will not provide the level of monitoring typically achieved in studies assessing a larger proportion of sessions. This limited fidelity sampling will therefore also be considered when interpreting the results.
Table 4. Summary of potential biases and mitigation strategies.
A third risk of bias is the risk of measurement bias linked to the reliability of clinician-rated assessments across the two sites. Even though both research assistants will receive structured training from clinicians experienced with the SCID-II and the ZAN-BPD, they will not complete an official certification-based program, and no formal interrater reliability procedures will be implemented. This raises the possibility of site-related discrepancies in the administration or scoring of these measures. To mitigate this risk, regarding the SCID-II, all diagnoses used for inclusion will be confirmed by the referent psychiatrist for the study in each site; and, regarding the ZAN-BPD, we added a concomitant self-report measure of BPD symptoms. Nevertheless, the risk of measurement bias remains, and such a methodological limitation should be considered when interpreting the findings. Four, we also believe that there is a significant risk of confounding bias linked to the fact that the “classic GPM” group is newly trained in GPM for this protocol, whereas the “GPM-extended” team is already trained and has one-year long experience with the model. Even though we believe such difference in terms of experience will be mitigated by training, supervision (including independent supervision), and fidelity assessment, there is still a risk of confounding bias that will be considered when analyzing the results. Five, another important limitation lies in the possibility that BPD criteria, while serving as a means to assess personality dysfunction (28), may also be inherently biased toward the specific subtype or dilemma characteristic of BPD. Consequently, the reliance on a BPD diagnosis made using the SCID as the primary inclusion criterion, as well as the choice of the ZAN-BPD as the primary outcome, introduce a risk of selection and measurement bias, particularly with respect to the interpretation of future findings on the efficacy of GPM and/or GPM-extended in improving dimensional personality functioning. In other words, even if significant improvements in personality functioning are observed with GPM and/or GPM-extended, these results should be interpreted cautiously, notably because GPM is already established as an effective treatment for individuals with BPD. Six, another limitation lies in the absence of formal validation of the GPM-DAS scale. Although the current trial will provide initial evidence regarding its reliability and validity, any interpretation of its scores will remain exploratory until further validation studies are conducted, to avoid measurement bias. Seven, a final source of confounding bias may stem from the fact that participants in both study arms will be allowed to engage in other forms of general psychiatric care (e.g., trauma-focused or addiction-focused treatments), if they do not target personality disorders. While this approach respects the ecological validity of real-world treatment settings, it also introduces the possibility that improvements in outcomes may partly reflect the influence of these ancillary interventions, rather than the study treatment alone. To mitigate this risk, we will try to limit and record such additional interventions as much as possible during the treatment year, notably to adjust for their potential influence, if needed. Despite these strong limitations, we believe that our study, if conducted rigorously, has the potential to contribute meaningfully to the existing literature on GPM. This is particularly notable given its long-term scope (1 year), its prospective controlled design, its incorporation of a more dimensional approach to the assessment of personality disorders, and its modular approach of the GPM framework.
Ethics statement
Written informed consent will be gathered for each patient by the clinical research assistants and the referent psychiatrists in each facility. This study protocol was submitted and validated by the Comité de Protection des Personnes Ile de France III (N°ID-RCB = 2024-A00475-42).
Author contributions
MBl: Conceptualization, Writing – review & editing, Investigation, Writing – original draft, Resources, Project administration, Formal analysis, Methodology. IB: Investigation, Conceptualization, Project administration, Writing – review & editing, Methodology. MZ: Project administration, Writing – review & editing. MA: Project administration, Writing – review & editing. AU-M: Project administration, Writing – review & editing. FC: Project administration, Writing – review & editing. M-LP: Project administration, Writing – review & editing. MBr: Project administration, Writing – review & editing. PL: Project administration, Writing – review & editing. OL: Project administration, Writing – review & editing. JC: Project administration, Writing – review & editing. SL: Project administration, Writing – review & editing. LC-K: Writing – review & editing, Methodology, Supervision, Conceptualization. MS: Supervision, Writing – review & editing, Conceptualization, Methodology.
Funding
The author(s) declared that financial support was not received for this work and/or its publication.
Conflict of interest
The authors declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declared that generative AI was not used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
References
1. Torgersen S. Epidemiology. In: The Oxford handbook of personality disorders. Oxford University Press, New York, NY, US (2012). p. 186–205. (Oxford library of psychology).
2. Moran P, Stewart R, Brugha T, Bebbington P, Bhugra D, Jenkins R, et al. Personality disorder and cardiovascular disease: results from a national household survey. J Clin Psychiatry. (2007) 68:69–74. doi: 10.4088/JCP.v68n0109
3. Oldham JM, Skodol AE, Kellman HD, Hyler SE, Doidge N, Rosnick L, et al. Comorbidity of axis I and axis II disorders. Am J Psychiatry. (1995) 152:571–8. doi: 10.1176/ajp.152.4.571
4. Skodol AE, Geier T, Grant BF, and Hasin DS. PERSONALITY DISORDERS AND THE PERSISTENCE OF ANXIETY DISORDERS IN A NATIONALLY REPRESENTATIVE SAMPLE: research article: personality disorders and anxiety disorders. Depress Anxiety. (2014) 31:721–8. doi: 10.1002/da.22287
5. Newton-Howes G, Tyrer P, and Johnson T. Personality disorder and the outcome of depression: Meta-analysis of published studies. Br J Psychiatry. (2006) 188:13–20. doi: 10.1192/bjp.188.1.13
6. Gunderson JG. Ten-year course of borderline personality disorder: psychopathology and function from the collaborative longitudinal personality disorders study. Arch Gen Psychiatry. (2011) 68:827. doi: 10.1001/archgenpsychiatry.2011.37
7. Skodol AE, Gunderson JG, McGlashan TH, Dyck IR, Stout RL, Bender DS, et al. Functional impairment in patients with schizotypal, borderline, avoidant, or obsessive-compulsive personality disorder. AJP. (2002) 159:276–83. doi: 10.1176/appi.ajp.159.2.276
8. Hastrup LH, Jennum P, Ibsen R, Kjellberg J, and Simonsen E. Societal costs of Borderline Personality Disorders: a matched-controlled nationwide study of patients and spouses. Acta Psychiatr Scand. (2019) 140:458–67. doi: 10.1111/acps.13094
9. Van Asselt ADI, Dirksen CD, Arntz A, and Severens JL. The cost of borderline personality disorder: societal cost of illness in BPD-patients. Eur Psychiatr. (2007) 22:354–61. doi: 10.1016/j.eurpsy.2007.04.001
10. Sheehan L, Nieweglowski K, and Corrigan P. The stigma of personality disorders. Curr Psychiatry Rep. (2016) 18:11. doi: 10.1007/s11920-015-0654-1
11. Gunderson JG, Herpertz SC, Skodol AE, Torgersen S, and Zanarini MC. Borderline personality disorder. Nat Rev Dis Primers. (2018) 4:18029. doi: 10.1038/nrdp.2018.29
12. Spitzer RL, Md KK, and Williams JBW. Diagnostic and Statistical Manual of Mental Disorders. Washington, DC 3rd ed. Washington, DC: American Psychiatric Association (1980).
13. Gunderson JG. Borderline personality disorder: ontogeny of a diagnosis. Am J Psychiatry. (2009) 166:530–9. doi: 10.1176/appi.ajp.2009.08121825
14. Lilienfeld SO and Latzman RD. Personality disorders: Current scientific status and ongoing controversies. In: Butcher JN and Hooley JM, editors. APA handbook of psychopathology: Psychopathology: Understanding, assessing, and treating adult mental disorders (Vol 1). Washington, DC: American Psychological Association, Washington (2018). p. 557–606. Available online at: http://content.apa.org/books/16053-023 (Accessed November 13, 2022).
15. American Psychiatric Association. The American Psychiatric Association Practice Guideline for the Treatment of Patients With Borderline Personality Disorder. 2nd ed. Washington, DC: American Psychiatric Association Publishing (2024). doi: 10.1176/appi.books.9780890428009
16. Storebø OJ, Stoffers-Winterling JM, Völlm BA, Kongerslev MT, Mattivi JT, Jørgensen MS, et al. Psychological therapies for people with borderline personality disorder. Cochrane Database Systematic Rev. (2020) 2020. doi: 10.1002/14651858.CD012955.pub2
17. Choi-Kain LW, Albert EB, and Gunderson JG. Evidence-based treatments for borderline personality disorder: implementation, integration, and stepped care. Harv Rev Psychiatry. (2016) 24:342–56. doi: 10.1097/HRP.0000000000000113
18. Gunderson J. The emergence of a generalist model to meet public health needs for patients with borderline personality disorder. AJP. (2016) 173:452–8. doi: 10.1176/appi.ajp.2015.15070885
19. Hopwood CJ, Kotov R, Krueger RF, Watson D, Widiger TA, Althoff RR, et al. The time has come for dimensional personality disorder diagnosis: The time has come for dimensional personality disorder diagnosis. Pers Ment Health. (2018) 12:82–6. doi: 10.1002/pmh.1408
20. Mulder R and Tyrer P. Borderline personality disorder: a spurious condition unsupported by science that should be abandoned. J R Soc Med. (2023) 116:1410768231164780. doi: 10.1177/01410768231164780
21. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. Fifth Edition. Arlington, VA: American Psychiatric Association (2013). doi: 10.1176/appi.books.9780890425596
22. World Health Organization. ICD-11 (2018). Available online at: https://icd.who.int/en (Accessed March 23, 2023).
23. Wright AGC, Ringwald WR, Hopwood CJ, and Pincus AL. It’s time to replace the personality disorders with the interpersonal disorders. Am Psychol. (2022) 77:1085–99. doi: 10.1037/amp0001087
24. Blay M, Benmakhlouf I, Duarte M, Perroud N, Greiner C, Charbon P, et al. Case reports: Using Good Psychiatric Management (GPM) conceptualizations in the dimensional assessment and treatment of personality disorders. Front Psychiatry. (2023) 14. doi: 10.3389/fpsyt.2023.1186524
25. Hutsebaut J, Willemsen E, Bachrach N, and Van R. Improving access to and effectiveness of mental health care for personality disorders: the guideline-informed treatment for personality disorders (GIT-PD) initiative in the Netherlands. Borderline Pers Disord Emot Dysregul. (2020) 7:16. doi: 10.1186/s40479-020-00133-7
26. Weinberg I, Finch E, and Choi-Kain L. Implementation of Good Psychiatric Management for Narcissistic Personality Disorder: Good Enough or Not Good Enough? In: Applications of good psychiatric management for borderline personality disorder: A practical guide. American Psychiatric Association Publishing, Washington, D.C (2019). p. 253–80.
27. Finch EF, Choi-Kain LW, Iliakis EA, Eisen JL, and Pinto A. Good psychiatric management for obsessive–compulsive personality disorder. Curr Behav Neurosci Rep. (2021) 8:160–71. doi: 10.1007/s40473-021-00239-4
28. Sharp C, Wright AGC, Fowler JC, Frueh BC, Allen JG, Oldham J, et al. The structure of personality pathology: Both general (‘g’) and specific (‘s’) factors? J Abnormal Psychol. (2015) 124:387–98. doi: 10.1037/abn0000033
29. Wright AGC, Hopwood CJ, Skodol AE, and Morey LC. Longitudinal validation of general and specific structural features of personality pathology. J Abnormal Psychol. (2016) 125:1120–34. doi: 10.1037/abn0000165
30. McMain SF, Links PS, Gnam WH, Guimond T, Cardish RJ, Korman L, et al. A randomized trial of dialectical behavior therapy versus general psychiatric management for borderline personality disorder. AJP. (2009) 166:1365–74. doi: 10.1176/appi.ajp.2009.09010039
31. McMain SF, Guimond T, Streiner DL, Cardish RJ, and Links PS. Dialectical behavior therapy compared with general psychiatric management for borderline personality disorder: clinical outcomes and functioning over a 2-year follow-up. AJP. (2012) 169:650–61. doi: 10.1176/appi.ajp.2012.11091416
32. Gunderson JG. Handbook of Good Psychiatric Management for Borderline Personality Disorder. Arlington, VA: American Psychiatric Pub (2014). p. 182.
33. First MB, Spitzer RL, Gibbon M, Williams JB, Lorna Smith B, Cottraux J, et al. Entretien clinique structuré pour les troubles de la personnalité de l’axe II du DSM-IV. (2013).
34. Hutsebaut J, Feenstra DJ, and Kamphuis JH. Development and preliminary psychometric evaluation of a brief self-report questionnaire for the assessment of the DSM–5 level of personality functioning scale: the LPFS brief form (LPFS-BF). Pers Disorders: Theory Research Treat. (2016) 7:192–7. doi: 10.1037/per0000159
35. Ridolfi ME, Rossi R, Occhialini G, and Gunderson JG. A randomized controlled study of a psychoeducation group intervention for patients with borderline personality disorder. J Clin Psychiatry. (2019) 81:41–6. doi: 10.4088/JCP.19m12753
37. Blay M, Duarte M, Benmakhlouf I, Amate M, Perroud N, Speranza M, et al. Psychoeducation for pathologic narcissism and narcissistic personality disorder: A review and proposal for a good psychiatric management-based six-week group program. J Psychiatr Practice. (2024) 30:249–58. doi: 10.1097/PRA.0000000000000797
39. First MB and Gibbon M. The Structured Clinical Interview for DSM-IV Axis I Disorders (SCID-I) and the Structured Clinical Interview for DSM-IV Axis II Disorders (SCID-II). In: Hoboken NJ, editor. Comprehensive handbook of psychological assessment, vol. 2 . John Wiley & Sons, Inc, US (2004). p. 134–43.
40. Lobbestael J, Leurgans M, and Arntz A. Inter-rater reliability of the structured clinical interview for DSM-IV axis I disorders (SCID I) and axis II disorders (SCID II). Clin Psychol Psychotherapy. (2011) 18:75–9. doi: 10.1002/cpp.693
41. Maffei C, Fossati A, Agostoni I, Barraco A, Bagnato M, Deborah D, et al. Interrater reliability and internal consistency of the structured clinical interview for DSM-IV axis II personality disorders (SCID-II), version 2.0. J Pers Disord. (1997) 11:279–84. doi: 10.1521/pedi.1997.11.3.279
42. Sheehan DV, Lecrubier Y, Sheehan KH, Amorim P, Janavs J, Weiller E, et al. The Mini-International Neuropsychiatric Interview (M.I.N.I.): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. J Clin Psychiatry. (1998) 59 Suppl 20:22–33.
43. Sheehan D, Lecrubier Y, Harnett Sheehan K, Janavs J, Weiller E, Keskiner A, et al. The validity of the Mini International Neuropsychiatric Interview (MINI) according to the SCID-P and its reliability. Eur Psychiatry. (1997) 12:232–41. doi: 10.1016/S0924-9338(97)83297-X
44. Zanarini MC. Zanarini rating scale for borderline personality disorder (ZAN-BPD): A continuous measure of DSM-IV borderline psychopathology. J Pers Disord. (2003) 17:233–42. doi: 10.1521/pedi.17.3.233.22147
45. Bomhof C, Löffler JP, Brugman S, Lodder P, van den End A, van Dijk MK, et al. The effectiveness of generalist (GIT-PD) versus specialist treatment (MBT/ST) for severe personality disorders (Personality Disorders Access to Effective Treatment, P-DAET): study protocol of a pragmatic randomised controlled non-inferiority multicentre trial. BMC Psychiatry. (2025) 25:1107. doi: 10.1186/s12888-025-07550-4
46. Combaluzier S, Gouvernet B, Menant F, and Rezrazi A. Validation d’une version française de la forme brève de l’inventaire des troubles de la personnalité pour le DSM-5 (PID-5 BF) de Krueger. L’Encéphale. (2018) 44:9–13. doi: 10.1016/j.encep.2016.07.006
47. Bohus M, Kleindienst N, Limberger MF, Stieglitz RD, Domsalla M, Chapman AL, et al. The short version of the borderline symptom list (BSL-23): development and initial data on psychometric properties. Psychopathology. (2009) 42:32–9. doi: 10.1159/000173701
48. Diguer L, Turmel V, Brin J, Lapointe T, Chrétien S, Marcoux LA, et al. Traduction et validation en Français du Pathological Narcissism Inventory. Can J Behav Science/Revue Can Des Sci du comportement. (2020) 52:115–20. doi: 10.1037/cbs0000140
49. Schoenleber M, Roche MJ, Wetzel E, Pincus AL, and Roberts BW. Development of a brief version of the pathological narcissism inventory. Psychol Assess. (2015) 27:1520–6. doi: 10.1037/pas0000158
50. Pinto A, Ansell EB, and Wright AGC. A new approach to the assessment of obsessive compulsive personality. In: Integrated paper session conducted at the annual meeting of the Society for Personality Assessment. Cambridge, MA: Cambridge, MA (2011).
51. Sadri SK, McEvoy PM, Pinto A, Anderson RA, and Egan SJ. A psychometric examination of the pathological obsessive compulsive personality scale (POPS): initial study in an undergraduate sample. J Pers Assessment. (2019) 101:284–93. doi: 10.1080/00223891.2018.1428983
52. Gratz KL and Roemer L. Multidimensional assessment of emotion regulation and dysregulation: development, factor structure, and initial validation of the difficulties in emotion regulation scale. J Psychopathol Behav Assessment. (2004) 26:41–54. doi: 10.1023/B:JOBA.0000007455.08539.94
53. Van der Linden M, d’Acremont M, Zermatten A, Jermann F, Larøi F, Willems S, et al. A french adaptation of the UPPS impulsive behavior scale. Eur J psychol Assessment. (2006) 22:38–42. doi: 10.1027/1015-5759.22.1.38
54. Whiteside SP and Lynam DR. The Five Factor Model and impulsivity: Using a structural model of personality to understand impulsivity. Pers Individ Differences. (2001) 30:669–89. doi: 10.1016/S0191-8869(00)00064-
55. Campbell WK, Bonacci AM, Shelton J, Exline JJ, and Bushman BJ. Psychological entitlement: interpersonal consequences and validation of a self-report measure. J Pers Assessment. (2004) 83:29–45. doi: 10.1207/s15327752jpa8301_04
56. Bagby RM, Parker JDA, and Taylor GJ. Twenty-five years with the 20-item toronto alexithymia scale. J Psychosomatic Res. (2020) 131:109940. doi: 10.1016/j.jpsychores.2020.109940
57. Dimitrijević A, Hanak N, Altaras Dimitrijević A, and Jolić Marjanović Z. The mentalization scale (MentS): A self-report measure for the assessment of mentalizing capacity. J Pers Assess. (2018) 100:268–80. doi: 10.1080/00223891.2017.1310730
58. Blay M, Hasler R, Nicastro R, Pham E, Weibel S, Debbané M, et al. Body modifications in borderline personality disorder patients: prevalence rates, link with non-suicidal self-injury, and related psychopathology. Borderline Pers Disord Emot Dysregul. (2023) 10:7. doi: 10.1186/s40479-023-00213-4
59. Zanello A, Weber Rouget B, Gex-Fabry M, Maercker A, and Guimon J. Validation du Questionnaire de fonctionnement social (QFS), un autoquestionnaire mesurant la fréquence et la satisfaction des comportements sociaux d’une population adulte psychiatrique. L’Encéphale. (2006) 32:45–59. doi: 10.1016/S0013-7006(06)76136-X
60. Arbuckle R, Frye MA, Brecher M, Paulsson B, Rajagopalan K, Palmer S, et al. The psychometric validation of the Sheehan Disability Scale (SDS) in patients with bipolar disorder. Psychiatry Res. (2009) 165:163–74. doi: 10.1016/j.psychres.2007.11.018
61. Leon AC, Olfson M, Portera L, Farber L, and Sheehan DV. Assessing psychiatric impairment in primary care with the sheehan disability scale. Int J Psychiatry Med. (1997) 27:93–105. doi: 10.2190/T8EM-C8YH-373N-1UWD
62. Sheehan DV, Harnett-Sheehan K, and Raj B. The measurement of disability. Int Clin psychopharmacology. (1996) 11:89–95. doi: 10.1097/00004850-199606003-
63. Favez N, Tissot H, Ghisletta P, Golay P, and Cairo Notari S. Validation of the french version of the experiences in close relationships– revised (ECR-R) adult romantic attachment questionnaire. Swiss J Psychol. (2016) 75:113–21. doi: 10.1024/1421-0185/a000177
64. Fraley RC, Waller NG, and Brennan KA. An item response theory analysis of self-report measures of adult attachment. J Pers Soc Psychol. (2000) 78:350–65. doi: 10.1037/0022-3514.78.2.350
65. Prout TA, Di Giuseppe M, Zilcha-Mano S, Perry JC, and Conversano C. Psychometric properties of the defense mechanisms rating scales-self-report-30 (DMRS-SR-30): internal consistency, validity and factor structure. J Pers Assessment. (2022) 104:833–43. doi: 10.1080/00223891.2021.2019053
66. Hatcher RL and Gillaspy JA. Development and validation of a revised short version of the Working Alliance Inventory. Psychother Res. (2006) 16:12–25. doi: 10.1080/10503300500352500
67. Kolla NJ, Links PS, McMain S, Streiner DL, Cardish R, and Cook M. Demonstrating adherence to guidelines for the treatment of patients with borderline personality disorder. Can J Psychiatry. (2009) 54:181–9. doi: 10.1177/070674370905400306
68. R Core Team. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing (2022). Available online at: https://www.R-project.org/ (Accessed April 10, 2023).
69. Gratz KL, Tull MT, and Levy R. Randomized controlled trial and uncontrolled 9-month follow-up of an adjunctive emotion regulation group therapy for deliberate self-harm among women with borderline personality disorder. psychol Med. (2014) 44:2099–112. doi: 10.1017/S0033291713002134
70. Juul S, Simonsen S, Poulsen S, Lunn S, Sørensen P, Bateman A, et al. Detailed statistical analysis plan for the short-term versus long-term mentalisation-based therapy for outpatients with subthreshold or diagnosed borderline personality disorder randomised clinical trial (MBT-RCT). Trials. (2021) 22:497. doi: 10.1186/s13063-021-05450-y
71. Priebe S, Bhatti N, Barnicot K, Bremner S, Gaglia A, Katsakou C, et al. Effectiveness and cost-effectiveness of dialectical behaviour therapy for self-harming patients with personality disorder: A pragmatic randomised controlled trial. Psychother Psychosomatics. (2012) 81:356–65. doi: 10.1159/000338897
72. Van Buuren S and Groothuis-Oudshoorn K. mice: Multivariate imputation by chained equations in R. J Stat Software. (2011) 45:1–67. doi: 10.18637/jss.v045.i03
73. Barnicot K, Katsakou C, Bhatti N, Savill M, Fearns N, and Priebe S. Factors predicting the outcome of psychotherapy for borderline personality disorder: A systematic review. Clin Psychol Review. (2012) 32:400–12. doi: 10.1016/j.cpr.2012.04.004
74. Kramer U. Good-enough therapy: A review of the empirical basis of good psychiatric management. Am J Psychother. (2024) 78:11–5. doi: 10.1176/appi.psychotherapy.20230041
75. Kramer U, Berger T, Kolly S, Marquet P, Preisig M, de Roten Y, et al. Effects of motive-oriented therapeutic relationship in early-phase treatment of borderline personality disorder: a pilot study of a randomized trial. J Nerv Ment Dis. (2011) 199:244–50. doi: 10.1097/NMD.0b013e3182125d19
76. Kramer U, Kolly S, Berthoud L, Keller S, Preisig M, Caspar F, et al. Effects of motive-oriented therapeutic relationship in a ten-session general psychiatric treatment of borderline personality disorder: a randomized controlled trial. Psychother Psychosom. (2014) 83:176–86. doi: 10.1159/000358528
77. Kramer U, Grandjean L, Machinea JB, Beuchat H, Ranjbar S, de Roten Y, et al. Emotional processing as mechanism of change in brief good psychiatric management for borderline personality disorder: results of a randomized controlled trial. BMC Psychiatry. (2024) 24:921. doi: 10.1186/s12888-024-06370-2
78. Blay M, Duarte M, Charbon P, Poulet E, Debbané M, Weibel S, et al. Should we be so radical as to overlook facts? J R Soc Med. (2023) 116:228. doi: 10.1177/01410768231181252
79. Gunderson J and Choi-Kain L. HMS Postgraduate Education. General Psychiatric Management for Borderline Personality Disorder (2018). Available online at: https://postgraduateeducation.hms.harvard.edu/general-psychiatric-management-borderline-personality-disorder (Accessed January 26, 2022).
Keywords: borderline personality disorder, categorical, dimensional, good psychiatric management, personality disorder, psychotherapy
Citation: Blay M, Benmakhlouf I, Zenou M, Amate M, Uludag-Martin A, Chevrier F, Pomey M-L, Brisebarre M, Lebourleux P, Lopepe O, Cacouault J, Lagadec S, Choi-Kain L and Speranza M (2026) Protocol for a cohort comparison treatment study evaluating the efficacy of a dimensional adaptation of GPM (GPM-extended) compared to a classic outpatient GPM treatment for patient fulfilling criteria for BPD: the EPSYLIA project. Front. Psychiatry 16:1687987. doi: 10.3389/fpsyt.2025.1687987
Received: 18 August 2025; Accepted: 23 December 2025; Revised: 13 December 2025;
Published: 27 January 2026.
Edited by:
Francesca Strappini, Foro Italico University of Rome, ItalyReviewed by:
Robert Gregory, Upstate Medical University, United StatesNadège Bourvis, Assistance Publique Hopitaux De Paris, France
Copyright © 2026 Blay, Benmakhlouf, Zenou, Amate, Uludag-Martin, Chevrier, Pomey, Brisebarre, Lebourleux, Lopepe, Cacouault, Lagadec, Choi-Kain and Speranza. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Martin Blay, bWFydGluLmJsYXk1QGdtYWlsLmNvbQ==