Background
Borderline Personality Disorder (BPD) is one of the most prevalent mental disorders [
1]. In the general population 1 to 2% of adults are diagnosed with BPD. In psychiatric populations up to 10% of outpatients and 20% of inpatients are diagnosed with BPD [
1,
2]. BPD is also associated with high psychosocial and socio-economic costs [
3,
4]. The economic burden of disease associated with BPD is higher than that associated with depression, and comparable to that of patients with schizophrenia [
1,
4]. BPD is also associated with high psychiatric comorbidity, particularly depression, anxiety disorders, eating disorders, substance abuse [
5‐
7] and various other personality disorders [
5,
6,
8‐
13], often in combination with high levels of acting-out (e.g., suicidality) [
14] and/or functional impairment [
15‐
17]. The lifetime risk for completed suicide associated with BPD may be as high as 10% [
14]. Together, these findings emphasize the need for the development of effective treatments for this severe disorder.
In recent years, a number of evidence-based specialist treatments for BPD have been developed and evaluated [
1]. These include dialectical behaviour therapy [
18], schema therapy [
19], transference-focused psychotherapy [
20,
21], Systems Training for Emotional Predictability and Problem Solving (STEPPS) [
22], and Mentalization-Based Treatment (MBT) [
23]. Various reviews [
1,
24,
25] and treatment guidelines [
26‐
28] recommend these treatments for patients with BPD.
The present study focuses on MBT as developed by Bateman and Fonagy in the United Kingdom [
23,
29,
30]. MBT is a promising psychodynamic treatment that is rooted in attachment and mentalizing approaches. Briefly, mentalizing refers to the capacity to interpret the self and others in terms of internal mental states such as feelings, emotions, wishes, desires, attitudes and values. This capacity is typically acquired in attachment relationships, and is associated with feelings of self-agency, affect regulation and resilience in the face of adversity. A growing body of research suggests that impairments in mentalizing can be seen as a core feature of BPD, as patients with BPD typically fail to make sense of their own internal experiences and those of others, particularly in contexts characterized by high levels of arousal. This results in emotional instability, impulsive behaviour, and the use of self-defeating strategies in an attempt to cope with these feelings (such as self-harm, substance abuse and promiscuity).
So far, two types of MBT have been empirically investigated: intensive outpatient MBT (MBT-IOP) [
29] and day hospital MBT (MBT-DH). Both MBT-IOP and MBT-DH consist of a treatment phase and a maintenance phase, each lasting a maximum of 18 months. The treatment phase of MBT-DH, the focus of the current study, consists of a day hospital treatment (five days per week) that includes daily group psychotherapy, weekly individual psychotherapy, individual crisis management from a mentalizing perspective, art therapy twice a week, mentalizing cognitive therapy and writing therapy. The maintenance phase in MBT-DH consists of a one-day follow-up treatment program combined with intermittent individual follow-up appointments, with the frequency reduced over time (step down).
At the time the current study was designed (2007), there was only one randomized controlled trial (RCT) investigating the efficacy and cost-effectiveness of MBT-DH compared to treatment as usual (that is, standard psychiatric care) conducted by the developers of MBT [
31]. This study randomized 38 BPD patients to either MBT-DH or TAU, which consisted of standard treatment offered in the UK in general psychiatric services and comprised (a) regular psychiatric review with a senior psychiatrist when necessary (on average twice a month); (b) inpatient admission when necessary, with discharge to non-psychoanalytic psychiatric partial hospitalization focusing on problem solving; followed by (c) outpatient and community follow-up as standard aftercare [
31]. Results showed that MBT-DH was superior to TAU on all major outcome variables, that is, depressive symptoms, suicide attempts and self-harm, number of inpatient days, and social and interpersonal functioning. These results were maintained during the 18-month follow-up period [
32]. Five years after discharge from MBT, the MBT-DH group still showed superiority over TAU on suicidality, diagnostic status, service use, use of medication, global functioning scores above 60 (on the Global Assessment of Functioning [GAF] Scale), and vocational status [
33]. For example, 74% of the patients in the TAU condition had made at least one suicide attempt, in comparison with only 23% in the MBT-DH group. And at the end of the follow-up period, 13% of the MBT-DH patients met the diagnostic criteria for BPD, compared to 87% of the TAU group. Before treatment the total health related-costs for the MBT-DH group ($44,947) and the TAU group ($52,563) were comparable; after 18 months of treatment the costs were reduced to $27,303 in MBT-DH and $30,976 in TAU. During the 18-month follow-up, costs further diminished sharply. After 18 months follow-up, the total health-related costs in the MBT-DH group were one-fifth of that for patients in the TAU condition: $3,183 for MBT-DH compared to $15,490 for TAU [
32].
Since this original trial, two other trials focusing on MBT-DH have been published. An RCT in Denmark investigated the efficacy of MBT-DH compared with a less intensive manualized supportive group therapy combined with psycho-education and medication treatment in patients diagnosed with BPD [
34]. In total, 58 patients were randomly allocated to MBT-DH and 27 patients to the specialist combined treatment. Results showed that both the intensive combined MBT treatment and the less-intensive supportive group therapy led to significant improvements on a variety of psychological and interpersonal measures, e.g., general functioning, depression, social functioning and number of diagnostic criteria met for BPD, with moderate to large effect sizes (
d = 0.5 to 2.1). Contrary to the expectations of the researchers, however, MBT-DH was superior only on therapist-rated GAF [
34]. No follow-up or cost-effectiveness data are yet available from this trial.
Additional evidence for the effectiveness of MBT-DH comes from a naturalistic study by Bales and colleagues [
35] in The Netherlands. These authors investigated the effectiveness of 18-months of manualized MBT-DH in 45 patients with severe BPD and a high prevalence of comorbid Axis-I and Axis-II disorders. Results showed significant improvements in symptomatic distress, social and interpersonal functioning, and personality pathology and functioning; with moderate to large effect sizes (
d = 0.7 to 1.7). These authors also showed that care consumption, defined as additional treatments and admissions during the last year before entry into and during MBT treatment, reduced significantly during and after treatment. Yet, the lack of a control group limits the possibility to draw conclusions from this study about the effectiveness of MBT-DH.
Hence, although there is some promising evidence supporting the efficacy and cost-effectiveness of MBT-DH, given the small number of studies, more research is urgently needed, particularly in light of the limitations of existing trials. First, one of the two RCTs was conducted by the developers of MBT [
31], and thus researcher allegiance may have influenced this study. Second, it is unclear whether results from trials conducted in the UK and Denmark may generalize to The Netherlands, given the large differences in health care systems between these countries. For instance, standard psychiatric care may be more effective in The Netherlands than in the UK because of differences in the allocation of health care and the clinical training of health workers. For instance, standard care in the Netherlands includes more and thus more expensive evidence-based treatments compared to the UK, and there is more funding available per patient. This assumption leads to the expectation that differences between a specialist treatment such as MBT-DH and TAU may be smaller in The Netherlands, as they may be in Denmark, which may explain the lack of substantial differences between MBT and specialist standard care in the trial of Jørgensen et al [
34]. This assumption is supported by recent findings that highly structured treatment programmes are associated with considerable effects in BPD patients, which are often comparable to the effects of specialist treatments such as MBT [
33,
34,
36]. For instance, in a randomized trial of BPD patients, Bateman and Fonagy [
29] found that MBT-IOP outperformed a manualized structured clinical management programme only at long-term follow-up in terms of effects on suicide attempts, severe incidents of self-harm, symptom severity, depression, interpersonal functioning and social adjustment. Hence, with regard to generalizability, a concern is the need for trials comparing MBT-DH to a credible TAU [
29].
Further, the study by Jørgensen and colleagues [
34] suffered from a number of important methodological limitations, such as a skewed randomization (with the majority of patients being randomized to MBT-DH) and the fact that the same therapists conducted treatments in both conditions (i.e., MBT-DH and supportive therapy), which may have led to “spill-over” effects.
Finally, none of the existing trials focused on the purported mechanism of change in MBT, that is, changes in attachment and mentalizing. Given the growing evidence for the role of common factors in psychotherapy, there is a pressing need to provide evidence for presumed mechanisms of change in current evidence-based treatments such as MBT [
37].
Research aims and hypotheses
The primary aim of the present study was to investigate the efficacy and cost-effectiveness of MBT-DH in comparison to a specialist TAU in The Netherlands. We expect both MBT-DH and specialist TAU to be effective on both primary and secondary outcomes. Yet, we expect MBT-DH to outperform TAU, particularly at 36 months follow-up. After 18 months of treatment we expect the costs in the MBT programme to be comparable to TAU; after 36 months we expect MBT-DH to outperform TAU (following Bateman & Fonagy [
32]). Second, we will investigate purported mechanisms of change in MBT, focusing on changes in attachment and mentalizing. Results of this trial are expected to inform mental health professionals, patients, and policy makers.
Discussion
Although MBT-DH is a promising psychodynamic treatment for BPD [
1,
24‐
28], so far only two RCTs have compared MBT-DH to specialist TAU [
31,
34]. Given the relative paucity of data supporting the efficacy and cost-effectiveness of MBT-DH, the possible influence of researcher allegiance in one of the trials [
31], potential spill-over effects in another trial [
34], and problems with the generalization of findings to mental health systems in other countries, more research is urgently needed. Further, the only health economic study of MBT-DH did not use state-of-the-art cost-effectiveness methodologies [
32]. To assess the efficacy and cost-effectiveness of MBT-DH in The Netherlands, the current study aims to compare MBT-DH with specialist TAU using state-of-the-art cost-effectiveness analyses. Further, because the current study is a multi-site trial, and treatments in both treatment arms are conducted by different therapists, the risk of spill-over effects is minimal.
We expect both MBT-DH and TAU to lead to significant reductions in primary and secondary outcome measures, including symptomatic distress and interpersonal functioning. However, we expect MBT-DH to outperform TAU, particularly at 36 months follow-up.
To date, no studies have examined the purported mechanism of change in MBT in the context of a randomized trial and there is a pressing need to provide evidence for presumed mechanisms of change in current evidence-based treatments [
37]. Aside from providing new data on the efficacy and cost-effectiveness of MBT-DH, this trial also promises to provide a better understanding of the mechanisms of change in MBT.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
EMPL drafted the first version of the manuscript and maintained the lead in the writing process. DW was responsible for the coordination of the study and collecting the data. JJMD leads the research project. JJMD, MJK and JP developed the study design and wrote the study proposal. JP developed the statistical design and randomization procedure, manages the data flow and will perform the statistical analyses together with EMPL. HVE is responsible for the cost-effectiveness analyses. AJVB was coordinator at the MBT-unit during the study. PL, JJVB, HVE and MN made substantial contributions to the revision of the manuscript. All authors provided comments, read and approved the final manuscript.