Introduction
Personality disorders (PD) are defined as “an enduring pattern of inner experience and behavior that deviates markedly from the expectations of the individual’s culture” 1. They have a large social impact on the general population: 7.8% of the overall world population suffer from a PD during their lifetime 2, while in high-income countries the prevalence of PD increases to 9.6% 2. Although PD are widespread everywhere, their incidence in terms of diagnoses seems to be higher in the more developed countries.
Regarding the clinical population, PD is diagnosed in 40-50% of psychiatric inpatients, with borderline personality disorder (BPD) alone reaching 30% 3. According to Gunderson et al. 4, the prevalence of BPD in psychiatric inpatients is 15-28%, taking into consideration different services (psychiatric clinics, hospitals and general health facilities). It has been hypothesized that the emergence of BPD in the clinical population may depend on the greater tendency of these patients to seek help, in combination with the severity of the disorder 5. Indeed, some studies indicate BPD, along with antisocial personality disorder (AnPD), as the most diagnosed PD in their severe form 5. Therefore, given the high prevalence of patients diagnosed with severe BPD, the social impact of this psychopathology is very high.
Regarding the treatments adopted to deal with PD, the guidelines of Western countries focus mainly on Cluster B and, in particular, BPD. In fact, the American Psychological Association (APA) has published guidelines concerning only the treatment of BPD 6, as have Switzerland, Denmark, Finland and Catalonia; Great Britain has adopted guidelines for BPD and AnPD; and only Sweden, Germany and the Netherlands have published generic guidelines for all PD 7.
Looking at the guidelines of Europe and USA as a whole, there is general agreement in: (1) considering psychotherapy as the primary treatment 7; (2) emphasizing the importance of having specifically trained clinicians; (3) reporting data relating to the effectiveness of some types of psychotherapy; and (4) considering hospitalization useful in acute cases. The first two features, moreover, are confirmed by Evans et al. 8, which links the effectiveness of treatments to their clarity and consistency, as well as the availability of well-supported and adequately trained clinicians 8. In addition, all guidelines integrate the psychological treatment with a pharmaceutical intervention, which, focusing on behavioral traits influenced by neurochemical dysfunctions of the central nervous system, limits the impact of symptoms during the therapeutic process 9. Therefore, most of the guidelines state the importance of using multimodal treatments, in spite of some differences: in the American guidelines, this implies the combination of individual, group and couple psychotherapy 6; other countries do not concern only the clinic and recommend a more articulate process involving different professional figures (psychologists, psychiatrists, nurses, social workers) 7,10-11.
It follows that although psychotherapy-based outpatient treatment is the first indication of approach in all guidelines, some of these also admit residential treatments not related to acute illness; in these contexts a multimodal approach is used. For example, in Switzerland, it is possible to spend a period of 8-12 weeks in hospital or semi-hospital, strictly in preparation for outpatient treatment 12; in Great Britain, in addition to day services, there are therapeutic communities, which are specialized residential facilities dedicated to the treatment of BPD 10. The Italian National Health System also provides for therapeutic communities for severe PD 13. Even in these contexts, a multidisciplinary residential treatment is indicated.
Among the different experiences of residential facilities, the therapeutic communities undoubtedly present some specificities. Among these, the literature identifies three main characteristics 14. The first is the involvement in shared activities and community life 14, which helps patients to learn and maintain the ability to share the living space with other people and join in with structured activities. The second is the sense of belonging to a group, which is a fundamental motivation for human beings. In therapeutic communities, this aspect is strongly promoted because it may offer the opportunity to build stable relationships through time, as well as to benefit and consolidate the sense of being part of a collective 15. Lastly, the sense of agency, through which patients perceive themselves as the main agents of their actions and feel in control of their treatment, favors the development of their sense of responsibility without judgment 14-15.
In addition to the different contexts of care, the evaluation of treatment outcomes is another important issue. In fact, for each type of treatment an effect is expected, and this also applies to PD 16. However, such effects need to be defined and measured for several purposes, such as to justify mental healthcare expenditure, advance scientific research, give the clinician feedback on the patient’s progress 17 and communicate to the patient the extent of his change.
Currently, the literature offers evidence of therapeutic outcomes pursued in the outpatient setting and focused mainly on psychotherapy, whereas there seems to be less evidence regarding the effects of residential multimodal treatments specifically based on several psychosocial interventions. It is therefore necessary to take stock of the situation in this regard. With this in mind, our attention will focus on the expected outcomes from the more complex treatments, characterized by maximum intensity (achievable in a residential regime), long duration and multimodality, intended for patients with severe pathology. These people are characterized by frequent hospitalizations, a lot of suicide attempts and self-harming acts. Often these are patients with other comorbid psychiatric pathologies, unemployed or in prison. Their symptoms are therefore frequent, persistent and extreme and the social and occupational impairment is very high 18.
A core question related to evaluation of the treatment outcomes in PD concerns their duration. In general, considering the scientific literature, long-term interventions are indicated 19-20,3 to achieve a long-lasting effect across the lifespan of the patient 18. Starting from this assumption, residential settings would seem particularly suited to providing intensive and continuative care. Specifically, they would be able to induce structural changes in the patients’ inner world, in their strategies and behavioral patterns, with all-around monitoring and support that aims to provide a stable structure for the guests. In this kind of context, psychotherapeutic treatments are provided within a structured course during which people with PD can receive continuous support between sessions, within a coherent theoretical framework 21. A high therapeutic and rehabilitative intensity is desirable for the more critical clinical pictures 22, especially with younger patients who often dropped out or did not respond to other forms of less intensive treatment. By using such an approach it is possible to obtain reliable results, although the patient’s presence is required for a long-term period because changes in personality structure are resistant in time and hard to implement 23. In any case, indications about duration and frequency of treatment are not homogeneous. First of all, there is no agreement on the definition of short-term treatment (according to Danish guidelines it means less than 12 months; according to British guidelines, less than 3 months). The definition of long term is equally vague, ranging from 6 months to 1-3 years, and in some cases both short and long treatments are allowed, which is the case in Denmark 11 and Sweden 25. Furthermore, the duration and frequency of treatment often depend only on the duration and structure of the psychotherapy (which varies according to the theoretical orientations). This lack of agreement on the duration of treatment makes systemic analysis of the question difficult.
A further important question about the assessment of the treatments deals with which specific effects of the outcomes to evaluate. Indeed, according to the instructions of the UK National Institute for Health and Care Excellence 10, better outcomes can be achieved by combining group and individual approaches and integrating them with social care 9, which is the primary focus of long-term treatments in residential settings. In this context, the outcome evaluates changes within the manifestation of personality traits and behaviors of a patient. The outcome will necessarily depend on several specific elements for each kind of treatment: the severity of the situation prior to its inception (life history, social risk factors, severity and quality of the clinical picture), the impact of the treatment on the individual and the environment in which the treatment will take place 18. These factors are crucial for the three pivotal elements that determine the impairment due to a PD and correspond to the symptomatic manifestation of the disorder, the level of impairment in psychosocial functioning and the subjective distress of the patient’s condition 1.
While the symptomatic picture can be defined by clinical criteria 6,10,11,12,24,25, the signs of psychosocial functioning can be identified by the presence or the ability to: (a) maintain at least one good relationship able to provide emotional support; (b) work or study full-time or taking care of a home or a child adequately 19. In PD, because intrapsychic dynamics have a substantial negative impact also on social functioning 26-27,18, these aspects should be included when evaluating the outcome of any treatment for PD. The intrapsychic dynamics of PD are described differently by different theoretical orientations. However, the intrapsychic characteristic considered to be common to all PD is a lack of identity formation and integration 18,27. Overall, analyzing internal mechanisms, dysfunctional behaviors and relational and social difficulties separately is not an easy undertaking, given the high degree of interaction among these components 18,28. Consequently, current assessment tools also show this difficulty. Indeed, they often provide an accurate categorical diagnosis but give little information on other aspects related to this type of disorder: for example, how interpersonal functioning really affects patients or the reasons underlying dysfunctional acting (self-harm, for instance) 28.
Objectives
The aim of this work is to review which outcomes of complex treatments for PD are taken into account and in what form. The focus is on a given clinical group in residential settings or in contexts that can be assimilated in terms of intensity, duration and complexity of the treatment.
The choice of this particular clinical group is determined by the fact that multimodal settings show critical elements that are highly specific. These elements may not be observed when considering an outcome evaluation that is less complex and integrated, for example, psychotherapy alone. In addition, some non-specific features could have a large effect on treatment outcomes, such as staying in a shared context within which people who are experiencing a psychiatric condition are required to interact according to the rules of an appropriate social functioning. Assuming this, in order to find what the common assessment areas are within multimodal residential treatments and to distinguish them from unimodal treatments, this work includes different therapeutic and rehabilitative models.
Methods
We conducted a systematic review of relevant studies with the purpose of identifying which are the methods and tools to assess outcomes of therapeutic interventions in multimodal contexts. Our review has been conducted according to the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines 29. We included studies published from the year 2000 (year of publication of DSM-IV-TR, which we used as the first reference for the diagnosis definition of PD) until 2022. The research has been conducted through databases of books and journals, such as Pubmed, Scholar, PsycINFO, PsycCRITIQUES, JSTOR, Web of Science, Wiley Online Library Journals, Springer, OVID Emcare, Embase, Cochrane Library and CINAHL.
Inclusion criteria for our research were: (a) the presence of a PD diagnosis; (b) treatment administered in a residential setting (hospital settings and therapeutic communities) or a complex multimodal treatment; and (c) all material must be written in English. By “complex multimodal treatment” we mean all those intensive, complex and integrated treatments, both clinical and rehabilitative.
Another critical element that guided our study selection was the length of the treatment. Indeed, the guidelines do not indicate exact or unanimous timings. According to the British guidelines, short treatments last less than 3 months 10. In contrast, the Danish guidelines set the distinction between short and long treatment at 12 months 11. However, assuming that residential protocols and multimodal treatments are intended for seriously ill patients, whose personality structure is particularly resistant to change, the duration of treatment should include a minimum threshold. We therefore decided to take a 3-month period as a reasonable minimum limit. This threshold also appears to be in accordance with the British guidelines 10.
In conclusion, among the results of our research, we excluded: (a) studies that focused on outpatient psychotherapy only; (b) studies carried out on offenders held in prisons, due to the impossibility of an overall treatment that allows them to freely experiment with the recovery of the social dimension; (c) studies with patients younger than 18 and older than 65 years, who may be assessed on specific criteria related to age; (d) the presence of cognitive disabilities; (e) studies conducted in hospital wards, in which the outcome focuses only on medical-specific symptoms that require intensive and urgent care and on stabilization of the condition in order to proceed with the discharge (among these are studies on samples with double diagnoses or severe addiction issues requiring attention exclusively on remission of addiction symptoms; (g) treatments shorter than 3 months; and (h) studies focusing their outcome on the assessment of a single symptom (e.g., self-harm, eating disorders).
The research included three required components: PD; residential or multimodal treatment; and outcome assessment. Therefore, the literature review was carried out using the following keywords both individually and in combination: PD, outcome, assessment, result, therapeutic community, residential treatment and multimodal treatment.
Results
Our research led to 672 papers retrieved across all databases, plus 46 additional records identified through other sources; 6 of the 718 papers were removed because the full text could not be retrieved. After further examination, a total of 23 papers were considered relevant and included in our review. Within the 23 articles selected, two were reviews 30-31, 20 were clinical trials and one was a longitudinal study 32.
Samples
Some of the selected studies included diagnoses other than PD, such as mood disorders 22, psychotic disorders 33, post-traumatic stress disorder (PTSD) 34; or multiple diagnoses 32,35. Despite the presence of other diagnoses in the samples, these studies were considered relevant because the outcomes evaluated are also considered for patients with PD.
Moreover, sample size varies across the studies, from 22 patients 34 to 582 22 (Tab. I).
Furthermore, considering the composition of the samples with respect to gender, there is a prevalence of the female gender, which in some trials reaches 100% 34,36-38. Only in Young et al. 32 was there a prevalence of male gender (60%). The composition of the samples, therefore, is rather heterogeneous, as is their size.
Settings and duration
Since our interest is in high intensity multimodal treatments, different kinds of settings are included in the review. Likewise, the facilities in which the treatments take place are different from each other. In four cases they are therapeutic communities 32-33,39-40, whereas one is a clinic 34. Otherwise, the treatments take place in hospital and outpatient settings. Therefore, also with respect to the structures, a certain variability is observed, which increases considering that often a study includes different samples, placed in different contexts, in order to receive different treatments. This is because many of them are comparison studies.
Hence, concerning trials, 13 involve multiple variously combined settings, which can be summarized as follows: (1) step-down program, inpatient and outpatient treatments 39,41-42; (2) step-down program and inpatient treatment 43-44; (3) step-down program and outpatient treatment 20,36-37,45; and (4) inpatient and outpatient treatments 22,34,38,46. Then, there are treatments for inpatients only 33,35,40 and for outpatients only 47-48. Lastly, Bohus et al. 36 use only a step-down program. Among the different residential facilities, four of these are therapeutic communities 32-33,39-40.
Consequently, given the variety of contexts and programs, the duration of treatments ranges from 12 weeks 37 to 30 months 33-34. It is possible to note that most of the treatments, in outpatient, residential or mixed, provide for a total duration of more than 12 months.
The studies are grouped according to the setting and the duration (see Tab. II).
A particular case is represented by the work of Gunderson et al. 22, a long multicenter study on numerous samples (patients with BPD, n = 175; patients with cluster C PD, n = 312) in which the areas of global and social functioning are investigated. They report that alongside the improvements observed in some areas, social functioning remains impaired. However, while the results of Gunderson et al. 22 are based on data collected using scales adopted also in other contexts, the therapeutic approaches used with patients are not specified. In particular, it is not clear whether these were multidisciplinary and complex, as suggested by international guidelines.
Tools
Despite the differences in duration and setting, it is possible to identify the main outcomes assessed through the tools used. Indeed, we decided to proceed by analyzing the tools to trace the constructs, considering it the most objective way to reach the goal (Tab. III).
First of all, the studies pay attention to the preliminary diagnosis, which, if not accurate, could invalidate the results. Because of this, excluding Bohus et al. 36 and Werbart et al. 33, several tools are used to check the presence of Axis I and II disorders. The most common are the Structured Clinical Interview for DSM-III-R (SCID) and the Structured Clinical Interview for DSM-III-R Personality Disorders (SCID-II). However, other kinds of interviews are also used, such as the Mini International Neuropsychiatric Interview (MINI) used by Antonsen et al. 20,45 and Pabst et al. 34 Furthermore, Antonsen et al. 45 integrate information collection with the Severity Indices of Personality Problems (SIPP-118), whereas Javaras et al. 38 use the Diagnostic Interview for Personality Disorders (DIPD-IV) and the Revised Diagnostic Interview for Borderlines (DIB-R). The Borderline Syndrome Index (BDI) is used by Chiesa et al. 42,47, Jones et al. 48 and Young et al. 32 to define the diagnosis.
Still with regard to the preliminary diagnosis, we also count the National Adult Reading Test (NART), aimed to obtain the intelligence quotient 41, and the Posttraumatic Diagnostic Scale (PDS), used by Pabst et al. 34 to investigate this particular diagnostic question.
Regarding evaluation of the treatments, which is the core question of the current review, we can distinguish several different areas. First of all, strictly clinical features are the most evaluated. Indeed, excluding Gunderson et al. (2011) 22 and Javaras et al. (2017) 38, all remaining studies evaluate the clinical improvement or relapse by assessing the presence or absence of symptoms. To achieve this, a variety of tools suitable for gathering information on individual symptoms or symptom groups are used, such as the Symptom Checklist (SCL-90-R), a self-report questionnaire that covers many symptoms of psychiatric distress, used in 11 of the 23 studies considered. More specifically, the constructs investigated with ad hoc tools are: (1) depression, individually assessed in Bohus et al. 36-37, Jones et al. 48, Pabst et al. 34, Sollberg et al. 35 and Vermote et al. 46; (2) anxiety 35-37,46,48; (3) aggression 40; (4) anger 35-37,46; (5) self-esteem 45,48; (6) hopefulness 32; (7) dissociative experiences 34,37; (8) identity diffusion 33,35,46; (9) self-harm and attempted suicide 35-36,39-41,45-46; (10) mentalization skills 45; (11) interpersonal problems 20,37,45; and (12) meaningful relationships 46. The range of symptoms investigated is therefore wide and the tools used are numerous.
A further notable area deals with the outcomes related to psychosocial functioning. The most popular tools to collect this information are the Global Assessment of Functioning (GAF), the Global Assessment Scale (GAS), the Social Adjustment Scale (SAS) and the Work and Social Adjustment Scale (WSAS), used in 12 studies 20,22,32-33,37,39,41,43-45,48-49. Javaras et al. 38, however, make use of the Background Information Schedule (BIS), whereas Pearce et al. 40 provide information on improvement in psychosocial functioning through the Social Function Questionnaire (SFQ). Despite their diversity, these tools collect information on important domains of the patient’s life, such as work, accommodation, home management, social and private leisure activities and close relationships. Certainly, this second area mainly concerns rehabilitation and includes both relational aspects, such as the ability to build a friendship network or to have close familiar relationships, and also more individual skills, such as the ability to take care of one-self, manage free time, contact territorial services autonomously, maintain a good level of physical health and adhere to treatment. Furthermore, even keeping a job or reaching an objective related to studies or professional career and settling in stable residential accommodation can be considered rehabilitation goals. Finally, to assess the general sense of empowerment and particularly the sense of empowerment to manage problems related to physical and mental health, Young et al. 32 use specific tools: the Making Decisions and Empowerment Scale (MD&ES), the Health Orientations Scale (HOS), the Mental Health Recovery Scale-Revised (MHRS-R) and the Mental Health Confidence Scale (MHCS).
A third kind of assessed results deals with the patients’ recourse to the available services. To collect these data, Beecham et al. 39 employ a variant of the Client Service Receipt Inventory (CSRI), which gives information about accommodation support, hospital services, employment services, general practitioner (GP) contacts, drug and alcohol services, community-based psychiatrist, psychologist and psychiatric nurse, social worker, education classes and legal services. In a similar way, visits to primary services and emergency departments are recorded using specially constructed questionnaires, both in Pearce et al. 40 and Chiesa et al. 42,47.
Hence, it is observed that some of the scales used cover several areas: this is the case of the Strauss-Carpenter Outcome Scale (SCOS), which investigates both clinical features (symptom severity) and other aspects related to social functioning and is scored by the staff 33.
Finally, it is possible to notice how, starting from validated tools, specific scales have been built to fit the needs of particular contexts: the Cassel Community Adjustment Questionnaire is designed to collect information of a clinical nature 42,47, while the HAS scale, also resulting from an adaptation, is designed to be used by nurses, providing a complete profile of behavioral functioning and evaluating adaptation to the hospital environment and rules 49. In addition, Pearce et al. 40 developed two self-report questionnaires specifically for the current study, one related to suicidal acts and self-harm and the other to provide data on visits to primary care and emergency departments.
Discussion
Summarizing the results of the review, it substantially emerges that the areas evaluated at the end of complex treatments are, in order of frequency, clinical, psychosocial functioning and use of services. In particular, the most relevant feature is fragmentation of the treatment methods, both in terms of duration and structuring. The overall duration of the treatments, with each of them integrated and complex, varies between 12 weeks and a few years. The variability in the structuring concerns the setting (outpatient, residential or mixed mode), as well as the single tasks that concretely characterize the therapeutic process of the patients. Therefore, it is clear, as already highlighted by Bateman and Fonagy 31 in their review, that until now there has been no universally widespread program of treatment that is acknowledged or has demonstrated decisive superiority with regard to setting or type of treatment compared to others. In fact the guidelines, both in European countries and the USA, are not homogeneous in indicating the timing and procedures of treatments, which are, however, very complex. What has been suggested so far, and that represents the point of convergence of a long analysis combined with extensive expertise in the field, is that there are some characteristics that any treatment for BPD (not generalizable to other PD) should include: structured psychosocial treatments, made in close partnership with the patient by active, supportive and validating therapists; focus on managing life situations that help to connect feelings to event and actions; pharmacotherapy limited to managing specific symptoms 9; and a particular focus, during psychotherapy, on increasing patient mentalization 50.
With regard to differences in duration, these are also likely to reflect the economic constraints imposed by national health systems. This kind of health policy, however, does not consider that, in the long term, social costs arising from non-adequately treated PD could most likely far outweigh the benefits of immediate savings. Indeed, there is a risk that the cost for patients with PD will be borne by the health systems (and community in general) throughout their lives, whereas it would probably be less expensive to try to prevent such chronicity 51-54.
A second point concerns the areas that define a treatment as multidisciplinary: clinical work, the individual’s recovery and social inclusion. Analysis of the indicators used to monitor therapeutic progress of integrated treatments shows that the number of scales used in each study for such testing is very small. Indeed, there are studies based on the use of only three 38,42,47 or four measuring instruments 22,33,39,44. Limitation in the use of such available instruments seem not to be adequate for the wide range of areas to investigate when evaluating the outcomes of complex treatments on patients with severe impairment. On the other hand, consistent with what has been observed with regard to the variety of treatments, the measuring instruments chosen by the authors also vary greatly: among 63 total instruments, 41 are used only in single studies. The fact that the choice of measuring instruments is also based on the theoretical reference framework undoubtedly makes the picture even more complicated. Diversification goes so far as to create scales according to the needs of individual services, such as the CCAQ 42,47, the HAS 49 or the questionnaires created by Pearce et al. 40. Certainly, tools designed to best meet the specific needs of a context can be very useful for that specific type of therapy, but not be equally suitable for other therapeutic contexts. Likewise, choosing scales according to the type of theoretical reference that is applied in the treatment is extremely reasonable. However, this makes it very difficult to compare results. The problem of comparing results had already been identified by Bateman and Fonagy 31, who emphasized as critical precisely the multiplicity of theoretical approaches and also the difficulty of identifying samples of patients homogeneous with each other. Among the weaknesses highlighted by Bateman and Fonagy 31 also remains that relating to the lack of controlled studies, with as many as eight not having a control group. On this matter, however, it should not be forgotten that even single-case studies can be extremely significant. As highlighted by Dazzi et al. 55, they constitute a kind of bridge between clinical practice and research. Being indeed closer to clinical practice than the randomized clinical trial, they are better able to capture the complexity and bring out some elements of the process, in addition to the results 55. Wampold and Imel 55 also observed that, in randomized controlled trials, there is great difficulty in determining rigorously the mediations and the mechanisms of change that interpose between the delivery of psychotherapy and the observable outcomes. However, the problem of comparability of results remains open to future solutions.
Another critical element, linked to the previous one, concerns the fact that the complex and integrated approach emphasizes the central role of individual psychotherapy 57-58: see also Bateman and Fonagy 31 for a detailed review. According to the relevance of this theme, Oud et al. 58 analyzed the most effective specialized psychotherapies in the treatment of BPD. In support of the crucial role of individual psychotherapy in the treatment of BPD, it has been shown that psychotherapy changes not only mental functions but also brain structure 58. Indeed, the authors, through a magnetic resonance imaging study, demonstrated an increase in gray matter critical areas of the patient’s brain (anterior cingulate cortex, inferior frontal gyrus and superior temporal gyrus) implicated in emotion regulation and higher-order functions, such as mentalizing. In this regard, it should be noted that although there is evidence of changes in psychotherapeutic treatment, it appears to be a lack of tools for evaluating strictly and exclusively psychotherapy as a process. The reason for this shortage seems to be the complexity of this kind of research. It is not possible here to deal extensively with this issue, for which reference is made to the works of Dazzi et al. 55 and Wampold and Imel 56. However, it is worth mentioning the main critical elements: the existence of several psychotherapies of different orientation, hence the problem of identifying common factors and specific factors, and the difficulty of separating elements closely involved with the patient-therapist relationship from other elements.
It follows that it is even more difficult to specify the measurement of complex and layered care processes to which different factors and different professionals play a role. In complex treatments, interventions are indeed multidisciplinary and, as mentioned, are composed of psychotherapy and rehabilitation, aimed at social reintegration. With regard to the interaction of these different combined interventions and the impact they have on each other, there are no systematic studies. However, the remark of Fonagy et al. 60 on the relationship between psychotherapy for BPD and the social environment is extremely interesting. The authors assume that psychotherapy (mainly intended as an opportunity to increase the mentalization ability) can show its main relapse in the social life of the patient. As patients gradually becomes more able to understand themselves and others, they receive, from social situations, positive reinforcement of their acquisitions and an encouragement to further improve. This would also produce, over time, a reduction of the symptomatology. The complex interdependence between sociality and the reduction of psychopathology proposed by Fonagy et al. 60 makes a real contribution to the discussion on the relationship between psychotherapy and psychosocial functioning and makes it even more urgent to study carefully the individual factors involved. In the wake of this, it remains to investigate how the contribution of other professionals (professional social workers, nurses and other figures working in rehabilitation) in turn reflects on the overall recovery.
Further complexity to an already rich picture is added, as already mentioned, by the degree of customization of treatments in complex contexts: while remaining within the treatment protocols, indeed, the team need, in therapeutic and rehabilitative practice, to make the treatment as suitable as possible for the patient. This problem is detected by Capone et al. 30 in a review dedicated to the outcomes of treatments on borderline patients in therapeutic communities, where it is underlined how the personalization of treatment makes measurement and standardization difficult. Finally, patients with PD often have other comorbid diseases, which complicates both the treatment protocols, which must take into account the two diagnoses, and also the study protocols, which sometimes present mixed samples and in which it is even more difficult to attribute the observed results to specific factors.
Conclusions
The issues described above highlight two needs that open up new challenges. The first concerns a deep understanding of PD: these disorders are so complex and, thinking about BPD, have such different manifestations that it is difficult for clinicians to currently develop a unique treatment. Probably, a diagnostic approach that is not only categorical, such as that of DSM, but also attentive to a dimensional analysis of the individual patient could encourage greater understanding of individual functioning and could be useful to set the treatment accordingly 61. Indeed, in order for a therapeutic plan to be effective and not be iatrogenic, it is appropriate to consider that everyone has, at the base, their own style of personality and are sensitive to specific issues depending on their life story 61. Likewise, while looking at the theme from another point of view, Guidano 62 points out that, although within a life cycle that in some respects unites everyone, each one will follow a unique development itinerary that cannot be superimposed on anyone else 62,63. Of course, such considerations on the uniqueness of the individual and his/her life themes cannot be neglected in planning the treatment of a psychiatric disorder. Finally, in the formulation of a therapeutic project, the severity of the individual case must also be taken into account 61,64. It follows that the uniqueness of functioning and development, even pathological, cannot therefore be ignored. However, it remains a challenge for the future to combine the needs of categorization of research with the need to fully understand individuality, staying within a shared framework.
The second challenge concerns measuring instruments. In accordance with the foregoing, they should be suitable for three purposes: help in understanding the functioning of the patient from all points of view (assessment); measuring treatment outcomes (what and how much has changed); and evaluating the processes (how and why the change took place).
The review shows, for example, that only one study involves the use of a specific tool for the detection of PTSD (the PDS) 34. Nevertheless, BPD is increasingly being configured as the result of a complex PTSD 65. Among the useful tools for understanding the patient, it may therefore be interesting to systematically insert something that identifies the presence of complex PTSD; all the more so considering that understanding this link between BPD and complex PTSD has led to the hypothesis that an effective treatment may be the one that addresses the two disorders together 66.
Tests to detect neuropsychological dysfunction (such as attention and memory associated with typical BPD impulsivity) may also be appropriate. Recent studies show that they could provide useful data both in terms of prognostics, that is, evaluation of the degree of impulsiveness of the patient (linked to the greater or lesser propensity for anticonservative and self-harming behaviors), and also for the evaluation of subsequent recovery 67. This is confirmed by imaging studies, which suggests that deficits in impulse control may be predictive of the outcome of treatment and perhaps also of involvement in psychotherapy 68.
Regarding the measurement of individual outcomes, as mentioned above, the review reveals that many scales are used, especially for clinical outcomes. Although psychosocial recovery is also an outcome, it is not considered with the same frequency as clinical outcomes. Instead, it might be interesting to always combine the clinical evaluation with tools to measure the rehabilitation dimension (including access to services after treatment). Indeed, if functioning in the social environment becomes the measure of psychotherapeutic success 60, the data relating to social reintegration (e.g., the ability to at least partially maintain a stable job) become extremely significant. Therefore, given the nature of these long-term processes, it would be desirable to follow up regularly for some years, aimed precisely at verifying these outcomes.
Finally, the evaluation of processes is perhaps the most complex aspect to deal with. The difficulty involved in evaluating the psychotherapeutic process has already been mentioned. Let us just add that, although it is a costly task and therefore difficult to fit into an evaluation battery, it would be an opportunity to move in this direction. That psychotherapy acts also on brain plasticity, allowing a recovery of mental functions, is known 57-58 but it remains to understand exactly how. In the same way, since we discuss layered and complex interventions, it might be interesting to examine not only the processes that take place in the psychotherapist’s room but also all those triggered by other qualified professionals working towards rehabilitation targets.
It is therefore advisable that, in order to learn more about these issues, new research and new tools be developed.
Conflict of interest
The authors declare no conflict of interest.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Ethical consideration
Not applicable.
Authors’ contributions
All the Authors conceived of the present idea. Maggioni wrote the manuscript with the relevant support from Risso, Caslini and Bailo. Pellegrini and Rezzonico supervised the project.
Figures and tables
| Sample size | ||
|---|---|---|
| Inpatients | Step-down | Outpatients |
| Antonsen et al., 2014 (n = 59) | Antonsen et al., 2014 (n = 54): control | |
| Antonsen et al., 2016 (n = 42) | Antonsen et al., 2016 (n = 37): control | |
| Beecham et al., 2006 (n = 32) | Beecham et al., 2006 (n = 29): control | Beecham et al., 2006 (n = 47): control |
| Bohus et al., 2000 (n = 24) | ||
| Bohus et al., 2004 (n = 31) | Bohus et al., 2004 (n = 19): control | |
| Chiesa, 2000 (n = 81) | ||
| Chiesa et al., 2004 (n = 49) | Chiesa et al., 2004 (n = 45): control | Chiesa et al., 2004 (n = 49): control |
| Chiesa et al., 2009 (n = 68) | ||
| Chiesa et al., 2017 (n = 45) | Chiesa et al., 2017 (n = 87): control | Chiesa et al., 2017 (n = 30): control |
| Chiesa & Fonagy, 2000 (n = 46) | Chiesa & Fonagy, 2000 (n = 44): control | |
| Chiesa & Fonagy, 2003 (n = 40) | Chiesa & Fonagy, 2003 (n = 40): control | |
| Javaras et al., 2017 (n = 61) | Javaras et al., 2017 (n = 164): control | |
| Jones et al., 2013 (n = 25) | ||
| Pabst et al., 2014 (n = 11) | Pabst et al., 2014 (n = 11): control | |
| Pearce et al., 2017 (n = 70) | ||
| Sollberg et al., 2015 (n = 44) | ||
| Vermote et al., 2010 (n = 31) | Vermote et al., 2010 (n = 13): control | |
| Werbart et al., 2012 (n = 56) | ||
| Young et al., 2014 (n = 323) | ||
| Duration of treatment | ||||
|---|---|---|---|---|
| 3 months/12 weeks | 4-6 months | 7-12 months | Up to more than 12 months | |
| Inpatients | Bohus et al., 2004Sollberg et al., 2015Young et al., 2014 | Chiesa et al., 2004Chiesa et al., 2017Chiesa & Fonagy, 2003Vermote et al., 2010 | Beecham et al., 2006(11-16 months)Chiesa, 2000(6-14 months)Chiesa & Fonagy, 2000(11-16 months)Pabst et al., 2014(30 months)Pearce et al., 2017(3-18 months)Werbart et al., 2012(av. 30 months) | |
| Step-down | Bohus et al., 2000* | Antonsen et al., 2014(18 weeks + 4 years + 2.5 years)Antonsen et al., 2016(av. 32 months)Beecham et al., 2006(11-16 months + 12-18)Chiesa et al., 2004(6 months + 12-18 + 6-9)Chiesa et al., 2017(6 months + 24)Chiesa & Fonagy, 2000(6 months + 12-18 + 9)Chiesa- Fonagy, 2003(6 months + 12-18 + 9) | ||
| Outpatients | Bohus et al., 2004 | Vermote et al., 2010 | Antonsen et al., 2014(av. 2 years)Antonsen et al., 2016(av. 25 months)Beecham et al., 2006(av. 24 months)Chiesa et al., 2009(av. 18.2 months)Chiesa et al., 2017(24 months + 6/12)Jones et al., 2013(18 months)Pabst et al., 2014(30 months) | |
| *In the study by Bohus et al., 2000 the duration of treatment for the post-discharge part is not specified. | ||||
| Scale | Frequency |
|---|---|
| Structured Clinical Interview for DSM-IV Axis II Disorders (SCID-II) | 12 |
| Symptom Checklist-90-R (SCL-90-R) | 11 |
| Global Assessment of Functioning (GAF) | 7 |
| Global Assessment Scale (GAS); State-Trait Anxiety Inventory (STAI) | 5 |
| Beck Depression Inventory (BDI); Social Adjustment Scale (SAS); Structured Clinical Interview for DSM-IV Axis I Disorders (SCID-I); State-Trait Anger Expression Inventory (STAXI) | 4 |
| Borderline Syndrome Index (BSI); Structured Clinical Interview (SCID); Mini International Neuropsychiatric Interview (MINI); Hamilton Depression Scale (HAMD) | 3 |
| Cassel Community Adjustment Questionnaire (CCAQ); Client Satisfaction Questionnaire-8 (CSQ-8); Revised Diagnostic Interview for Borderlines (DIB-R); Fragebogen für Dissoziative Symptome (FDS), Hamilton Anxiety Scale (HAMA); Inventory of Personality Organization (IPO); Lifetime Parasuicide Count (LPC); Self Harm Inventory (SHI); Work and Social Adjustment Scale (WSAS) | 2 |
| Adult Attachment Interview (AAI); Beck Depression Inventory II (BDI-II); Background Information Schedule (BIS); Brief Symptom Inventory 18 (BSI-18); Borderline Symptom List (BSL-23); Circumplex of Interpersonal Problems (CIP); Clinical Outcomes in Routine Evaluation (CORE-OM); Client Service Receipt Inventory (CSRI); Dissociative Experiences Scale (DES); Diagnostic Interview for DSM-IV Personality Disorders (DIPD-IV); EuroQol-5D (EQ-5D); Generalized Anxiety Disorder 7 (GAD-7); General Health Questionnaire 12 (GHQ-12); Global Social Adjustment (GSA); Hospital Adjustment Scale (HAS); Health of the Nation Outcome Scales (HONOS); The Hope Scale (HOPE); Health Orientations Scale (HOS); Hopkins Symptom Checklist 25 (HSCL-25); Inventory of Interpersonal Problems (IIP); Inventory of Interpersonal Problems Circumplex Scales (IIP-C); Index of Self-Esteem (ISE); Integration/Sealing-over Scale (ISOS); Making Decisions and Empowerment Scale (MD&ES); Mental Health Confidence Scale (MHCS); Mental Health Recovery Scale-Revised (MHRS-R); Modified Overt Aggression Scale (MOAS); National Adult Reading Test (NART); Patient Health Questionnaire 9 (PHQ-9); Purpose In Life Test (PIL); Posttraumatic Diagnostic Scale (PDS); Reflective Functioning Rating Scale (RF); Rosenberg Self-Esteem Scale (RSE); Standardized Assessment of Personality Abbreviated Scale (SAPAS); Social Adjustment Scale-Self-Report (SAS-SR); Strauss-Carpenter Outcome Scale (SCOS); Social Functioning Questionnaire (SFQ); Spiritual Involvement and Beliefs Scale (SIBS); Severity Indices of Personality Problems (SIPP-118); UCLA Loneliness Scale (ULS) | 1 |
PDF
