BPD, part 2: schema therapy and how the approaches compare

In short

• BPD has an unusually crowded field of therapies. A 2020 Cochrane review found 75 randomised trials with more than 4,500 participants, testing more than 16 types of therapy.
• Schema therapy grew out of cognitive therapy. Its founder, Jeffrey Young, set out to help people with long-standing, chronic patterns that classic CBT did little for. It explains BPD through childhood needs that went unmet, 18 "schemas" and rapidly shifting "modes".
• Its signature tools are unusual for CBT: "limited reparenting", a warm and active therapist stance; imagery rescripting of memories; and chairwork. These are professional techniques, not self-help.
• Schema therapy works best as individual plus group sessions. In the largest trial (495 people, five countries) the combination had a large effect, while a "mostly group" version did no better than usual treatment.
• Schema therapy and DBT have now been compared head to head twice, in 2024 and 2026. Neither trial found a significant difference on the main outcome. "No difference found" is not the same as "proven equal".
• No brand wins. Meta-analyses: psychotherapy for BPD beats usual treatment, but the margin is modest, and no specialist therapy has been shown to beat another. Well-organised "general" management (GPM) was no worse than DBT in a trial.
• BPD is not a life sentence. Over 10 years, 86% of former inpatients had a lasting remission of symptoms, but only about half reached full recovery, including work and relationships.
• The brand name matters less than you might think. What effective approaches share: a clear model, structure, enough time, and a trained clinician with supervision.

Timeline

The main events at a glance, from the first books on cognitive therapy for personality disorders to the head-to-head trials of the 2020s.

1. 1990

   Beck, Freeman and colleagues publish Cognitive Therapy of Personality Disorders

   Cognitive line

2. 1993

   Marsha Linehan publishes her DBT manual (covered in Episode 11)

   Third wave

3. 1995

   The STEPPS group programme is created at the University of Iowa

   Cognitive line

4. 1999

   Bateman and Fonagy run the first trial of what became MBT, with 38 people

   Evidence

5. 2003

   Young, Klosko and Weishaar: Schema Therapy: A Practitioner's Guide

   Cognitive line

6. 2006

   Giesen-Bloo trial (schema therapy vs TFP) and the BOSCOT trial (CBT for BPD)

   Evidence

7. 2009

   NICE guideline on BPD; DBT vs GPM trial in Toronto

   Evidence

8. 2014

   Schema therapy tested for other personality disorders (323 people)

   Evidence

9. 2017–2020

   Meta-analyses by Cristea (2017) and Oud (2018), and the Cochrane review (2020)

   Evidence

10. 2022

    Largest schema therapy trial: 495 people, five countries

    Evidence

11. 2024

    First head-to-head trial of schema therapy vs DBT (Germany); new APA guideline

    Evidence

12. 2026

    Second head-to-head trial, BOOTS (the Netherlands)

    Evidence

What the lines mean. "Cognitive line" here means approaches that grew out of Beck's cognitive therapy; "Third wave" is DBT; "Evidence" covers trials, meta-analyses and guidelines. MBT and TFP are psychodynamic approaches and appear on the timeline only through their trials. Whether schema therapy belongs to the third wave or to the "second", cognitive wave is disputed sources disagree; see Episode 1.

How to read: evidence labels

Every claim carries a label that shows how solid it is.

The label comes first in brackets, followed by the source, for example proven by research; Storebø et al., 2020. Almost all the trials discussed here were run by the developers' own groups; more on that in "So who won?".

What BPD is, and why it has so many therapies

A short reminder of the diagnosis, and of why a whole "tournament" grew up around it.

Borderline personality disorder (BPD) is a lasting pattern of instability in emotions, self-image and relationships, with impulsive behaviour. Episode 11 covers the diagnosis itself and Marsha Linehan's DBT. The key point here: BPD was long seen as almost untreatable, which is exactly why so many specialist approaches appeared.

The scale shows in the numbers. The 2020 Cochrane review gathered 75 randomised trials with 4,507 participants, covering more than 16 types of psychotherapy; DBT and MBT were the most studied documented fact; Storebø et al., 2020.

It helps to sort these approaches into three families (interpretation; following Choi-Kain et al., 2017):

One note on the diagnosis. In ICD-11, in force since 2022, personality disorder is described by severity (mild, moderate, severe) and by prominent traits; a "borderline pattern" is kept as an extra specifier documented fact; Psychiatric News, 2021. In 2022 the UK's NICE said it was reviewing its BPD guideline because of this change documented fact; NICE, 2009, 2022 update. Nearly all the studies below used the older criteria, DSM-IV and DSM-5.

Schema therapy: Beck's student and the "hard" cases

Where the method came from, and why it counts both as part of CBT and as a school of its own.

Schema therapy was created by the American psychologist Jeffrey Young. He earned his graduate degree at the University of Pennsylvania and completed a postdoctoral fellowship at its Center for Cognitive Therapy, the centre of Aaron Beck, the founder of cognitive therapy documented fact; Young's biography on PESI; Wikipedia. Young later founded the Schema Therapy Institute in New York documented fact; Young et al., 2003, publisher's author note.

Young and his followers tell the origin story like this. Some clients improve quickly with classic cognitive therapy: find the thought, test it, feel better. For others the problem isn't a single thought but the shape of their whole life: the same relationships, the same pain, the same choices, again and again. A standard course did little for them, and that is who Young built his approach for the author's own account; Young et al., 2003, via Wikipedia: Schema therapy.

The core text is Schema Therapy: A Practitioner's Guide by Young, Klosko and Weishaar (2003); a Russian translation came out in 2020 documented fact; Young et al., 2003; Young et al., 2020.

CBT or not?

It depends on what you look at (interpretation).

• For "it's CBT": it grew out of cognitive therapy, works with core beliefs (see Episode 2) and uses cognitive and behavioural techniques. A review by Jacob and Arntz (2013) calls schema therapy a major development of CBT for personality disorders the author's own account; Jacob & Arntz, 2013.
• For "it's a separate school": it borrows from attachment theory, Gestalt and psychodynamic work, has its own "modes" model, imagery and chair techniques, its own society (ISST) and its own certification documented fact; ISST.

Classic CBT mostly works with what is happening now. Schema therapy pays far more attention to childhood and to feelings as they arise in the session.

Early maladaptive schemas: 18 themes about self and others

What a schema is, how to spot one in everyday life, and what the data say about the model.

An early maladaptive schema is, in the International Society of Schema Therapy's definition, a broad, pervasive theme about oneself and one's relationships with others that develops in childhood the author's own account; ISST. A schema forms when a child's temperament meets painful experiences with family or peers. Afterwards it feels familiar and "right"; people may unwittingly bend events so the schema stays "true"; and it can lie dormant for a long time until the right situation wakes it up the author's own account; ISST.

An everyday example. A partner hasn't replied to a message for two hours. One person thinks, "probably busy". Another goes, within a minute, to "he's leaving, people always leave me". If that reaction repeats for years across different relationships, schema therapy would see something like the "abandonment" schema.

According to Young, the goal of therapy is to help a person learn to meet their core emotional needs the author's own account; ISST. The 2003 book lists five groups of needs: secure attachment to others; autonomy, competence and a sense of identity; freedom to express needs and emotions; spontaneity and play; and realistic limits and self-control the author's own account; Young et al., 2003.

18 schemas in 5 domains

A domain is a group of schemas linked to one unmet need the author's own account; ISST.

The domains and schemas follow ISST the author's own account; ISST. The phrases in the third column are our illustrations, not questionnaire items.

What the data say about the model itself

The model is a working map, not a proven blueprint of the mind (interpretation). In 2018 Bach, Lockwood and Young himself analysed responses from more than a thousand adults in Denmark, from clinical and non-clinical samples. The data fitted four domains better than five: disconnection and rejection; impaired autonomy and performance; excessive responsibility and standards; and impaired limits. The authors call for replication proven by research, one study; Bach, Lockwood & Young, 2018. Psychometric work on the short form of the questionnaire also shows very high overlaps between some schemas, so some of them may partly duplicate each other proven by research; Bach et al., 2017.

Three ways of coping with a schema

When a schema is triggered, people cope with it somehow. Young describes three coping styles the author's own account; Young et al., 2003; ISST:

From the outside these look like opposites: procrastination and perfectionism. In the model they can be responses to the same schema, and all three keep it alive in the long run the author's own account; Young et al., 2003. This is a clinical observation by the authors, not a law tested in experiments (interpretation).

Modes: why BPD needed "states"

A schema is a lasting theme; a mode is what is happening to a person right now.

A person with a severe personality disorder has many schemas active at once, and their state can change within minutes. Working through 18 schemas one by one is impractical, so the mode model was developed the author's own account; Jacob & Arntz, 2013.

A mode (schema mode) is a here-and-now state: an emotion plus a way of coping with it, set off by an "emotional button" and leading to an over- or under-reaction the author's own account; ISST. The first mode model was built specifically for BPD; mode models were later described for most other personality disorders too the author's own account; Jacob & Arntz, 2013.

Five modes in BPD

In Kellogg and Young's (2006) model, five modes are most typical of BPD the author's own account; Kellogg & Young, 2006; Young et al., 2003:

In the model, the healthy adult mode is weak in people with BPD, and therapy builds it up the author's own account; Kellogg & Young, 2006.

There is a questionnaire for modes, the Schema Mode Inventory (SMI): 118 items, 14 modes; it was tested on 863 participants and its scales showed good reliability proven by research; Lobbestael et al., 2010. It is protected by its authors' rights, so we don't reproduce the items.

Inside the schema therapist's room

How it's done in therapy: the therapist's stance, the techniques, and what they cost. This is a description for understanding, not instructions.

Kellogg and Young describe treatment for BPD in three phases: bonding and emotional regulation; changing modes; and developing autonomy. They name four mechanisms of change: limited reparenting, experiential work (with imagery and dialogue), cognitive work and psychoeducation, and breaking behavioural patterns the author's own account; Kellogg & Young, 2006.

Limited reparenting

ISST calls limited reparenting the heart of schema therapy: the therapist builds a secure attachment and, within the limits of a professional relationship, partly meets the client's core needs that went unmet in childhood the author's own account; ISST: Limited Reparenting. In practice that means warmth, predictability, open care and protection from the "punitive parent", together with firm limits on the modes that sabotage therapy. The word "limited" is the important one: this happens inside professional boundaries; the therapist does not become a substitute parent.

That is a clear departure from the image of the reserved, neutral therapist: a schema therapist openly shows that they care. Nobody has yet managed to measure the contribution of this stance separately from the rest of the therapy (interpretation; Jacob & Arntz, 2013).

One piece of it has been tested. In a trial by Nadort and colleagues (62 people), half the therapists were randomly assigned to be available by phone outside office hours. It added nothing to the outcome; after 18 months of schema therapy, 42% of participants had recovered from BPD proven by research; Nadort et al., 2009.

Empathic confrontation

Empathic confrontation means warmth and firmness at once the author's own account; ISST. The therapist acknowledges where a pattern came from and how it once helped, and names plainly what it costs now. For example: "I understand why switching off was the only way to get through what happened back then. And I can see that when you switch off now, we lose the chance to help the part of you that's hurting" (a composite example).

Imagery rescripting

Imagery rescripting is a technique for changing the meaning of a painful memory, rather than simply reliving it. In Arntz's updated protocol (2025) the logic is this: the memory is recalled up to its worst moment, but instead of going through every detail, the question becomes what the person needed back then. In early sessions the therapist "steps into" the image and protects the child; later, the client does it themselves from the adult position the author's own account; Arntz, 2025.

The protocol has safety rules: keeping the threat in the image under control is the central task; there are ways to bring the person back to the here and now if they start to dissociate; and the therapist doesn't suggest details or confirm or deny whether an event really happened the author's own account; Arntz, 2025.

What we know about its benefits. For post-traumatic stress disorder (see Episode 9), rescripting as a standalone method is supported by meta-analyses, but it hasn't been shown to beat other evidence-based treatments proven by research; as reviewed in Arntz, 2025. Its specific contribution within schema therapy for BPD has not been isolated (interpretation; Jacob & Arntz, 2013). In a small study by Weertman and Arntz (21 people with personality disorders), 24 sessions focused on childhood memories worked about as well as 24 sessions focused on the present proven by research, small sample; Weertman & Arntz, 2007.

What patients say: in a German qualitative study, 21 people with BPD described strong emotions and exhaustion at first, and later a better grasp of their schemas, more stability and relief. Distrust, a fast pace and too little time to talk afterwards got in the way; a clear structure, preparation, starting with easier memories and a stop signal helped proven by research, qualitative study; Schaich et al., 2020.

Chairwork

Chairwork comes from Gestalt therapy and psychodrama: modes are "seated" on different chairs, and the person speaks from each one. A typical version is a dialogue with the punitive parent in which, from another chair, the healthy adult finally answers back the author's own account; Young et al., 2003, via Wikipedia: Schema therapy. In a qualitative study, 29 people with BPD often felt sceptical, ashamed or awkward at first, and pain and tiredness in the short term; in the long run, all participants described understanding their modes better and treating themselves less harshly proven by research, qualitative study; Josek et al., 2023.

Familiar CBT tools

Schema therapy also uses techniques familiar from CBT: listing a schema's pros and cons, testing the "evidence" for it, a dialogue between the schema and the healthy side, flashcards for difficult situations, a schema diary, role-play and behavioural homework between sessions the author's own account; Young et al., 2003.

Does schema therapy work? The key studies

From the first trial in 2006 to the large international one in 2022: what the data show, and where they are weak.

Three years, twice a week: schema therapy vs TFP

In 2006 in the Netherlands, 88 people with BPD at four centres were randomly assigned to schema therapy or transference-focused psychotherapy (TFP, Kernberg's psychodynamic approach). Both ran twice a week for three years documented fact; Giesen-Bloo et al., 2006.

Both produced significant improvements. But after three years, 45.5% in the schema therapy group met the recovery criterion on a BPD severity interview, versus 23.8% in TFP; the relative risk of recovery was 2.18. Reliable clinical improvement: 65.9% vs 42.9%. The risk of dropping out was higher in TFP proven by research, one trial; Giesen-Bloo et al., 2006.

On cost: over four years, schema therapy came to about €37,800 per person versus €46,800 for TFP proven by research; van Asselt et al., 2008. The intensive therapy was cheaper not in itself, but compared with another intensive therapy and counting all care costs.

The limitation: this compared two active therapies, not therapy against usual care, and it was run by schema therapy's developers (interpretation).

Group schema therapy as an add-on

In a small US trial by Farrell and colleagues (32 people), 30 group schema therapy sessions over eight months were added to usual individual therapy documented fact; Farrell et al., 2009. According to an ISST press release, dropout was 0% vs 25%, and by the end of treatment 94% vs 16% no longer met BPD criteria the author's own account; ISST press release, 2009 — not checked against the paper's tables. The sample is tiny, so treat these figures with care.

The largest trial: 495 people, five countries

The largest schema therapy trial came out in 2022: 495 people at 15 centres in Australia, Germany, Greece, the Netherlands and the UK, with two years of treatment documented fact; Arntz et al., 2022. It compared three options:

proven by research; Arntz et al., 2022

How to read d: it's a standardised difference between groups; about 0.2 is considered small, 0.5 medium, and 0.8 or more large. Both forms of schema therapy together beat usual treatment with d = 0.73, and the combined format beat the mostly-group one with d = 0.84. Over two years, 74% stayed in treatment in the combined arm, 62% in the group arm and 64% in usual treatment proven by research; Arntz et al., 2022.

The surprising takeaway: group work alone was not a cheap substitute. What worked was group plus individual work.

And the hard side: three people in this trial died by suicide, one in each arm documented fact; Arntz et al., 2022. Even the best therapies are no guarantee of safety, which is why a crisis plan is always needed.

Not only BPD

Schema therapy has also been tested for other personality disorders: avoidant, dependent, obsessive-compulsive, paranoid, histrionic and narcissistic. In the Netherlands in 2006–2011, 323 people at 12 centres were assigned to schema therapy, usual treatment, or clarification-oriented psychotherapy. After three years more people had recovered with schema therapy, and fewer dropped out proven by research; Bamelis et al., 2014. Sources give the exact rates differently: around 80% versus roughly 50–60% (sources disagree; 2014 press release via ScienceDaily: about 80%, 60% and 50%; Markowitz's 2015 commentary: 81%, 52% and 51%). The differences showed up in interviews but not in participants' self-reports proven by research; Bamelis et al., 2014.

Other roads: CBT-PD, STEPPS, MBT, TFP and GPM

Shorter, cheaper, in a group, through mentalising or through the relationship with the therapist: what the other approaches offer and what we know about them.

CBT for personality disorders (Davidson): 16 sessions in a real health system

Back in 1990, Beck, Freeman and colleagues described cognitive therapy for personality disorders: each disorder has its own profile of core beliefs and of "overdeveloped" and "underdeveloped" behavioural strategies documented fact; Beck, Freeman et al., 1990. On that basis the Scottish psychologist Kate Davidson built CBT for cluster B personality disorders, borderline and antisocial. It stays within the classic cognitive model: a shared case formulation, work on core beliefs about self and others, and building new strategies the author's own account; Davidson et al., 2006. Formulation, or conceptualisation, is covered in Episode 4.

Her question was practical: can people with BPD be helped within an ordinary public health system, without three years of twice-weekly sessions? The BOSCOT trial (Glasgow, London, Ayrshire) enrolled 106 people; they were offered up to 30 sessions over a year and attended 16 on average documented fact; Davidson et al., 2006.

The result was modest and honest proven by research; Davidson et al., 2006; Davidson et al., 2010:

• The share of people with at least one episode of suicidal behaviour over two years did not differ between groups (43% vs 54%), but on average there were about half as many such episodes in the CBT group.
• After six years the difference held only as a trend, without statistical significance; hospital stays were shorter on average in the CBT group.
• Depression, social functioning and quality of life did not change much; quality of life stayed low in both groups.

The authors warned against overstating the result, while noting that CBT can produce clinically important change in a relatively small number of sessions in real clinical settings the author's own account; Davidson et al., 2006.

STEPPS: bringing family and friends in

STEPPS (Systems Training for Emotional Predictability and Problem Solving) was created in 1995 at the University of Iowa by Nancee Blum and colleagues documented fact; University of Iowa, 2017. Its logic: BPD affects not just one person but their whole system of relationships the author's own account; Ness, 2009.

How it works the author's own account; Ness, 2009; NHS Education for Scotland:

• 20 weekly group sessions of about two hours, with two facilitators.
• Each session opens with a short self-rating, so people can see their progress on a chart.
• Participants learn to notice how an emotion builds across five levels, and practise skills for managing emotions and behaviour: distraction, challenging negative thoughts, everyday problem-solving, sleep, eating, avoiding substances.
• The "reinforcement team": the person chooses family, friends and professionals, who learn the programme's language. In a hard moment they don't argue or rescue; they help the person remember which skill fits.
• STEPPS is designed as an add-on to existing treatment, not a replacement.

A detail from the developers: the levels of emotion were first pictured as a tornado, but when the programme was introduced in the Netherlands it turned out Dutch has no word of its own for "tornado", so the image became a pot on the stove, from cold water to boiling the author's own account; University of Iowa, 2017. The same university reports that more than 25,000 people in the Netherlands have completed the programme and about 900 therapists have been trained the author's own account; University of Iowa, 2017.

The data. In the trial by Blum and colleagues (165 people), adding STEPPS led to greater improvement in overall BPD symptoms, impulsivity, mood and functioning, and fewer emergency visits; there were no differences in self-harm, suicide attempts or hospitalisations proven by research; Blum et al., 2008; NIMH, 2008. In the Cochrane review STEPPS beat usual treatment at the end of the course (SMD −0.39, three trials), but the effect was not confirmed at follow-up proven by research; Storebø et al., 2020. A 2023 systematic review (20 studies, three of them randomised) calls the results promising but too early for firm conclusions proven by research; Ekiz et al., 2023.

MBT: learning to read states, your own and other people's

MBT (mentalisation-based treatment) was developed in the UK by Anthony Bateman and Peter Fonagy and is psychodynamic in origin. Mentalising is the capacity to understand yourself and others in terms of the mental states behind behaviour the author's own account; Bateman & Fonagy, 2016. Not "he didn't reply, so he doesn't care", but "he didn't reply — maybe he's tired, maybe he's hurt, maybe he hasn't seen it". In the authors' model, this capacity "switches off" in BPD exactly under the stress of close relationships the author's own account; Bateman & Fonagy, 2016. In a later version of the theory the authors talk about epistemic trust, the capacity to trust others and learn from them (interpretation; Bateman et al., 2018).

The data proven by research:

• The first trial, in 1999, was small: 38 people in a day hospital (Bateman & Fonagy, 1999). Eight years after it began, suicidal behaviour was found in 23% vs 74% of the comparison group, and the BPD diagnosis persisted in 13% vs 87%; but the sample was 41 people, and overall social functioning remained impaired (Bateman & Fonagy, 2008).
• Outpatient MBT vs structured clinical management (SCM), 134 people, 18 months: both groups improved substantially, with steeper drops in crises and hospital admissions under MBT (Bateman & Fonagy, 2009).
• In an independent Dutch trial (95 people), day-hospital MBT did not beat specialist usual treatment on any measure, although fewer people dropped out early: 9% vs 34% (Laurenssen et al., 2018).
• Cochrane review: compared with usual treatment, MBT reduced self-harm (risk ratio 0.62) and suicidality (0.10), with low certainty of evidence (Storebø et al., 2020).

TFP: understanding yourself through the relationship with the therapist

TFP (transference-focused psychotherapy) is based on Otto Kernberg's model. In this model, people with BPD hold fragmented pictures of themselves and others ("identity diffusion"), and their main defence is splitting: people, including oneself, are seen as either all good or all bad. The work happens through the way these pictures play out in the relationship with the therapist, within a firm frame and a treatment contract the author's own account; Levy, Draijer & Yeomans, 2017.

The data. In a trial by Clarkin and colleagues (90 people, one year), TFP, DBT and supportive therapy produced broadly comparable results proven by research, low statistical power; Clarkin et al., 2007. Against treatment by experienced community psychotherapists, TFP had less dropout and more people who no longer met BPD criteria, although TFP patients had more sessions proven by research; Doering et al., 2010, as summarised in Evidence-Based Mental Health. Against schema therapy, TFP came out behind (Giesen-Bloo et al., 2006).

GPM: the dark horse with nothing exotic about it

GPM (good psychiatric management) was described by the American psychiatrist John Gunderson. No special techniques: at most one session a week, an honest conversation about the diagnosis and its usual course, attention to work and everyday life, and caution with medication the author's own account; Gunderson & Links, 2014; Choi-Kain et al., 2017. The clinician, in Gunderson's description, should be warm, reliable, interested and not intimidated the author's own account; Gunderson & Links, 2014.

In Toronto, 180 people with BPD and repeated self-harm were randomly assigned to a year of DBT or a year of GPM. Both groups improved significantly, with no difference between them proven by research; McMain et al., 2009. Two years after treatment ended, still no difference, and about two thirds of participants were in diagnostic remission proven by research; McMain et al., 2012.

That is why a review by Harvard researchers calls such models "good enough" and proposes stepped care: start with well-organised general management and move those who need more into an intensive specialist programme (interpretation; Choi-Kain et al., 2017). This is the review authors' position, not a unanimous verdict of the field, but it rests on data.

Seven approaches side by side: a comparison table

Model, mechanism, format and strength of evidence in one place. The "who it may suit" column is a hypothesis, not a rule.

Sources by row: DBT — Storebø et al., 2020; NICE, 2009; Choi-Kain et al., 2017. Schema therapy — Kellogg & Young, 2006; Giesen-Bloo et al., 2006; Arntz et al., 2022; Assmann et al., 2024; Wibbelink et al., 2026. MBT — Bateman & Fonagy, 2009; Laurenssen et al., 2018. TFP — Levy, Draijer & Yeomans, 2017; Clarkin et al., 2007. STEPPS — Blum et al., 2008; Ekiz et al., 2023. CBT-PD — Davidson et al., 2006. GPM — McMain et al., 2009, 2012; Choi-Kain et al., 2017. The "who it may suit" column is our interpretation of these sources; there is not yet any evidence that lets anyone match a method to a person in advance (interpretation; Oud et al., 2018).

What costs less. A review of 30 economic evaluations found that evidence-based psychotherapy for BPD saved on average about US$1,551 per person per year compared with usual treatment, though the data vary widely proven by research; Meuldijk et al., 2017. DBT requires training a whole team; GPM is quicker to learn, and one-day workshops improved clinicians' attitudes to treating BPD (interpretation; Choi-Kain et al., 2017).

So who won? Head-to-head trials and meta-analyses

What happens when the strongest approaches meet face to face, and what you see when all the trials are pooled.

Schema therapy vs DBT: two draws

• Germany, 2024 (PRO*BPD). 164 people, one centre, 18 months of one individual session and one group a week. A year after treatment ended there was no significant difference in BPD severity; DBT did better on anger proven by research; Assmann et al., 2024.
• The Netherlands, 2026 (BOOTS). 204 people (95 in DBT, 109 in schema therapy), nine centres, two years of treatment, main assessment at 36 months. The difference in BPD severity was not significant: d = 0.15 (95% CI −0.17 to 0.47), p = .27; both groups improved substantially. Over two years 52% left DBT and 46% left schema therapy, not a significant difference proven by research; Wibbelink et al., 2026, via the abstract summaries in ReachMD and Medical Dialogues.

Both trials tested whether one method beats the other; neither was designed to prove they are equal. The BOOTS authors state plainly that neither therapy was shown to be superior, and suggest studying next which treatment suits whom the author's own account; Wibbelink et al., 2026, via ReachMD.

What the meta-analyses show

proven by research; Cristea et al., 2017; Oud et al., 2018; Storebø et al., 2020. SMD and g follow the same logic as d: a standardised difference between groups.

What the guidelines say

• American Psychiatric Association (December 2024): recommends treating BPD with a structured psychotherapy (rated 1B); "no therapy emerged as a 'gold standard'"; no evidence was found that any medication treats the disorder's core symptoms; if medication is used, it should be time-limited clinical consensus; APA, 2024, press release.
• NICE (2009): consider a comprehensive DBT programme for women for whom reducing recurrent self-harm is a priority; don't use brief psychological interventions (under three months) outside a service with a clear model and supervision; don't use drug treatment specifically for BPD clinical consensus; NICE, 2009.

Why these data need care

• Allegiance — researchers' commitment to their own method. Most schema therapy trials involved Arntz and fellow developers; MBT trials, Bateman and Fonagy; TFP, the Clarkin and Kernberg group. In an independent MBT trial the difference was smaller (interpretation; Laurenssen et al., 2018).
• "Usual treatment" varies widely, from minimal care to DBT, as in the 2022 trial. The better the control, the smaller the gap (interpretation; Oud et al., 2018).
• Specialist arms often get more therapy hours than control arms (interpretation; Doering et al., 2010; Arntz et al., 2022).
• Dropout is high: in BOOTS about half the participants left within two years proven by research; Wibbelink et al., 2026.

So who won?

The honest answer: no single brand. Specialist psychotherapy for BPD does better than usual treatment, but the margin is modest and there is no winner among the names (interpretation; Cristea et al., 2017; Oud et al., 2018; Storebø et al., 2020). What effective approaches share is not a label but a set of features: a clear model that both the team and the client understand; structure and predictability; psychoeducation; a safety plan; a trained clinician with supervision; and enough time (interpretation; Choi-Kain et al., 2017; Bateman et al., 2018; NICE, 2009). How to read this kind of evidence, and its limits, is the subject of Episode 14.

Life after symptoms

BPD changes over time, but symptoms fade faster than lives recover.

BPD is often presented as a life sentence. The data don't support that. In a 10-year follow-up of 290 former inpatients, 93% had a symptom remission lasting at least two years, and 86% one lasting at least four. Full recovery — remission plus good social and vocational functioning for two years — was reached by 50%, and about a third of them later lost it proven by research, cohort study; Zanarini et al., 2010.

But functioning lags behind symptoms. Two years after a year of DBT or GPM, about two thirds of participants were in diagnostic remission, yet 53% were neither working nor studying and 39% were receiving disability support proven by research; McMain et al., 2012. Six years after CBT-PD, quality of life remained low in both groups proven by research; Davidson et al., 2010.

Hence the takeaway: "which therapy is best?" matters, but so does another question — how to get back to the work, study and relationships a person values (interpretation; McMain et al., 2012).

How to choose, and what good treatment looks like

Questions to ask yourself and a clinician. This is not a self-treatment guide: it's best to choose a format together with a clinician who knows your situation.

1. What matters most right now? If it's recurrent dangerous crises and self-harm, look for an approach with an explicit crisis-support system and a safety plan; for women with this priority, NICE names DBT clinical consensus; NICE, 2009. If there are no acute crises and the main issues are chronic relationship patterns, self-criticism and shame, discuss with a clinician the approaches that work on these in more depth (interpretation).
2. What is actually available near you? Well-delivered accessible care beats ideal care that doesn't exist; structured general management was no worse than DBT in a trial proven by research; McMain et al., 2009.
3. How much time and energy do you have? In studies, STEPPS ran for 20 weeks, CBT-PD up to a year, MBT about 18 months, and schema therapy 18 months to three years documented fact; from the trials above.
4. Group or individual? In schema therapy the combination worked best; group alone did no better than usual treatment proven by research; Arntz et al., 2022.
5. Is the clinician trained in the method, and supervised? How therapists were trained affected outcomes (Bamelis et al., 2014); NICE requires supervision clinical consensus; NICE, 2009.
6. Has the treatment model been explained to you? If you understand how treatment will work and why, that's a sign of a structured approach clinical consensus; NICE, 2009.

Red flags:

• promises of a cure within a month: NICE advises against interventions shorter than three months outside a structured service clinical consensus; NICE, 2009;
• medication as the main treatment for BPD: according to the APA and NICE, medication doesn't treat the core of the disorder clinical consensus; APA, 2024; NICE, 2009;
• an online "schema test" treated as a diagnosis;
• a clinician who can't explain which method they use.

Check yourself

1. How does a schema differ from a mode? A schema is a lasting theme about self and others formed in childhood; a mode is a current state, an emotion plus a response, that can change within minutes.
2. What three coping styles does Young describe? Surrender, avoidance and overcompensation; in the model, all three end up keeping the schema alive.
3. What is limited reparenting? A therapist stance in which, within professional limits, the therapist partly meets the client's core needs: warmth, predictability, protection and boundaries.
4. Which schema therapy format worked best in the 2022 trial? Individual plus group work (d = 1.14 vs usual treatment); the mostly-group format did no better than usual treatment.
5. What did the two head-to-head trials of schema therapy and DBT find? No significant difference on the main outcome. That is not proof that they are equal.
6. Why does the GPM trial matter? Simple structured management, at most once a week, was no worse than a year of DBT, either straight after treatment or two years later.

Key takeaways

• Schema therapy is an integrative branch of CBT: 18 schemas, three coping styles, modes, a warm and active therapist, and experiential techniques.
• The model is useful as a language, but data support its structure only partly, and the contribution of individual techniques hasn't been isolated.
• Schema therapy has good trials, including a large international one; individual plus group work does best.
• Neither DBT nor schema therapy nor any other brand has been shown to beat the others; well-organised general management comes close to them.
• Almost all trials were run by the methods' developers, effects are modest and dropout is high.
• BPD is treatable and, over time, many people reach remission; rebuilding a life takes longer than the symptoms do to fade.
• When choosing, what matters most is access, structure, enough time, a trained and supervised clinician, and a model you understand.

Sources

1. American Psychiatric Association (2024). Updated Borderline Personality Disorder Guideline (press release, 10 December 2024). https://www.psychiatry.org/News-room/News-Releases/Updated-Borderline-Personality-Disorder-Guideline
2. Arntz, A. (2025). Imagery Rescripting: an update of the treatment protocol. Behaviour Research and Therapy, 195, 104913. https://pure.uva.nl/ws/files/302328587/Imagery_Rescripting.pdf
3. Arntz, A., Jacob, G. A., Lee, C. W., et al. (2022). Effectiveness of predominantly group schema therapy and combined individual and group schema therapy for borderline personality disorder: a randomized clinical trial. JAMA Psychiatry, 79(4), 287–299. https://doi.org/10.1001/jamapsychiatry.2022.0010
4. Assmann, N., Schaich, A., Arntz, A., et al. (2024). The effectiveness of dialectical behavior therapy compared to schema therapy for borderline personality disorder: a randomized clinical trial. Psychotherapy and Psychosomatics, 93(4), 249–263. https://doi.org/10.1159/000538404
5. Bach, B., Lockwood, G., & Young, J. E. (2018). A new look at the schema therapy model: organization and role of early maladaptive schemas. Cognitive Behaviour Therapy, 47(4), 328–349. https://doi.org/10.1080/16506073.2017.1410566
6. Bach, B., Simonsen, E., Christoffersen, P., & Kriston, L. (2017). The Young Schema Questionnaire 3 Short Form (YSQ-S3). European Journal of Psychological Assessment, 33(2), 134–143. https://doi.org/10.1027/1015-5759/a000272
7. Bamelis, L. L. M., Evers, S. M. A. A., Spinhoven, P., & Arntz, A. (2014). Results of a multicenter randomized controlled trial of the clinical effectiveness of schema therapy for personality disorders. American Journal of Psychiatry, 171(3), 305–322. https://doi.org/10.1176/appi.ajp.2013.12040518
8. Bateman, A., & Fonagy, P. (1999). Effectiveness of partial hospitalization in the treatment of borderline personality disorder: a randomized controlled trial. American Journal of Psychiatry, 156(10), 1563–1569.
9. Bateman, A., & Fonagy, P. (2008). 8-year follow-up of patients treated for borderline personality disorder: mentalization-based treatment versus treatment as usual. American Journal of Psychiatry, 165(5), 631–638.
10. Bateman, A., & Fonagy, P. (2009). Randomized controlled trial of outpatient mentalization-based treatment versus structured clinical management for borderline personality disorder. American Journal of Psychiatry, 166(12), 1355–1364. https://doi.org/10.1176/appi.ajp.2009.09040539
11. Bateman, A., & Fonagy, P. (2016). Mentalization-Based Treatment for Personality Disorders: A Practical Guide. Oxford University Press.
12. Bateman, A., Campbell, C., Luyten, P., & Fonagy, P. (2018). A mentalization-based approach to common factors in the treatment of borderline personality disorder. Current Opinion in Psychology, 21, 44–49. https://discovery.ucl.ac.uk/id/eprint/1572371/
13. Beck, A. T., Freeman, A., et al. (1990). Cognitive Therapy of Personality Disorders. Guilford Press (3rd ed.: Beck, Davis & Freeman, 2015).
14. Blum, N., St. John, D., Pfohl, B., et al. (2008). Systems Training for Emotional Predictability and Problem Solving (STEPPS) for outpatients with borderline personality disorder: a randomized controlled trial and 1-year follow-up. American Journal of Psychiatry, 165(4), 468–478.
15. Choi-Kain, L. W., Finch, E. F., Masland, S. R., Jenkins, J. A., & Unruh, B. T. (2017). What works in the treatment of borderline personality disorder. Current Behavioral Neuroscience Reports, 4(1), 21–30. https://dash.harvard.edu/server/api/core/bitstreams/7312037e-1c87-6bd4-e053-0100007fdf3b/content
16. Clarkin, J. F., Levy, K. N., Lenzenweger, M. F., & Kernberg, O. F. (2007). Evaluating three treatments for borderline personality disorder: a multiwave study. American Journal of Psychiatry, 164(6), 922–928.
17. Cristea, I. A., Gentili, C., Cotet, C. D., Palomba, D., Barbui, C., & Cuijpers, P. (2017). Efficacy of psychotherapies for borderline personality disorder: a systematic review and meta-analysis. JAMA Psychiatry, 74(4), 319–328. https://doi.org/10.1001/jamapsychiatry.2016.4287
18. Davidson, K., Norrie, J., Tyrer, P., et al. (2006). The effectiveness of cognitive behavior therapy for borderline personality disorder: results from the BOSCOT trial. Journal of Personality Disorders, 20(5), 450–465. https://pmc.ncbi.nlm.nih.gov/articles/PMC1852259
19. Davidson, K. M., Tyrer, P., Norrie, J., Palmer, S. J., & Tyrer, H. (2010). Cognitive therapy v. usual treatment for borderline personality disorder: prospective 6-year follow-up. British Journal of Psychiatry, 197, 456–462. https://doi.org/10.1192/bjp.bp.109.074286
20. Doering, S., Hörz, S., Rentrop, M., et al. (2010). Transference-focused psychotherapy v. treatment by community psychotherapists for borderline personality disorder: randomised controlled trial. British Journal of Psychiatry, 196, 389–395; commentary: Evidence-Based Mental Health, 13(4), 119. https://mentalhealth.bmj.com/content/13/4/119
21. Ekiz, E., van Alphen, S. P. J., Ouwens, M. A., van de Paar, J., & Videler, A. C. (2023). Systems Training for Emotional Predictability and Problem Solving for borderline personality disorder: a systematic review. Personality and Mental Health, 17(1), 20–39. https://doi.org/10.1002/pmh.1558
22. Farrell, J. M., Shaw, I. A., & Webber, M. A. (2009). A schema-focused approach to group psychotherapy for outpatients with borderline personality disorder: a randomized controlled trial. Journal of Behavior Therapy and Experimental Psychiatry, 40(2), 317–328. https://doi.org/10.1016/j.jbtep.2009.01.002 ; ISST press release (2009): https://www.eurekalert.org/news-releases/559343
23. Giesen-Bloo, J., van Dyck, R., Spinhoven, P., et al. (2006). Outpatient psychotherapy for borderline personality disorder: randomized trial of schema-focused therapy vs transference-focused psychotherapy. Archives of General Psychiatry, 63, 649–658. https://scholarlypublications.universiteitleiden.nl/handle/1887/13110
24. Gunderson, J. G., with Links, P. (2014). Handbook of Good Psychiatric Management for Borderline Personality Disorder. American Psychiatric Publishing.
25. International Society of Schema Therapy (ISST). Schema Therapy. https://schematherapysociety.org/Schema-Therapy ; Limited Reparenting. https://schematherapysociety.org/Limited-Reparenting
26. Jacob, G. A., & Arntz, A. (2013). Schema therapy for personality disorders — a review. International Journal of Cognitive Therapy, 6(2), 171–185.
27. Josek, A. K., Schaich, A., Braakmann, D., et al. (2023). Chairwork in schema therapy for patients with borderline personality disorder — a qualitative study of patients' perceptions. Frontiers in Psychiatry, 14, 1180839. https://doi.org/10.3389/fpsyt.2023.1180839
28. Kellogg, S. H., & Young, J. E. (2006). Schema therapy for borderline personality disorder. Journal of Clinical Psychology, 62(4), 445–458.
29. Laurenssen, E. M. P., Luyten, P., Kikkert, M. J., et al. (2018). Day hospital mentalization-based treatment v. specialist treatment as usual in patients with borderline personality disorder: randomized controlled trial. Psychological Medicine. https://repub.eur.nl/pub/104948
30. Levy, K. N., Draijer, N., & Yeomans, F. E. (2017). Transference-focused psychotherapy. In A. Wenzel (Ed.), The SAGE Encyclopedia of Abnormal and Clinical Psychology. SAGE.
31. Lobbestael, J., van Vreeswijk, M., Spinhoven, P., Schouten, E., & Arntz, A. (2010). Reliability and validity of the Short Schema Mode Inventory (SMI). Behavioural and Cognitive Psychotherapy, 38(4), 437–458. https://doi.org/10.1017/S1352465810000226
32. Markowitz, J. C. (2015). Is psychotherapy for personality disorders worth it? Journal of Clinical Psychiatry, 76(11), e1472–e1473. https://www.psychiatrist.com/jcp/is-psychotherapy-personality-disorders-worth/
33. McMain, S. F., Links, P. S., Gnam, W. H., et al. (2009). A randomized trial of dialectical behavior therapy versus general psychiatric management for borderline personality disorder. American Journal of Psychiatry, 166, 1365–1374.
34. McMain, S. F., Guimond, T., Streiner, D. L., Cardish, R. J., & Links, P. S. (2012). Dialectical behavior therapy compared with general psychiatric management for borderline personality disorder: clinical outcomes and functioning over a 2-year follow-up. American Journal of Psychiatry, 169(6), 650–661. https://doi.org/10.1176/appi.ajp.2012.11091416
35. Meuldijk, D., McCarthy, A., Bourke, M. E., & Grenyer, B. F. S. (2017). The value of psychological treatment for borderline personality disorder: systematic review and cost offset analysis of economic evaluations. PLoS ONE, 12(3), e0171592. https://doi.org/10.1371/journal.pone.0171592
36. Nadort, M., Arntz, A., Smit, J. H., et al. (2009). Implementation of outpatient schema therapy for borderline personality disorder with versus without crisis support by the therapist outside office hours: a randomized trial. Behaviour Research and Therapy, 47, 961–973. https://cris.maastrichtuniversity.nl/en/publications/implementation-of-outpatient-schema-therapy-for-borderline-person/
37. Ness, T. M. (2009). STEPPS: viable supplement for treatment of borderline personality disorder. Psychiatric Times. https://www.psychiatrictimes.com/view/stepps-viable-supplement-treatment-borderline-personality-disorder
38. NHS Education for Scotland. The Matrix: STEPPS. https://www.matrix.nhs.scot/explore-the-recommended-interventions-therapies/system-training-for-emotional-predictability-and-problem-solving-for-borderline-personality-disorder-stepps/
39. NICE (2009, updated 2018 and 2022). Borderline personality disorder: recognition and management (CG78). https://www.nice.org.uk/guidance/cg78
40. NIMH (2008). Group therapy program offers meaningful gains for people with borderline personality disorder. https://www.nimh.nih.gov/news/science-news/2008/group-therapy-program-offers-meaningful-gains-for-people-with-borderline-personality-disorder
41. Oud, M., Arntz, A., Hermens, M. L. M., Verhoef, R., & Kendall, T. (2018). Specialized psychotherapies for adults with borderline personality disorder: a systematic review and meta-analysis. Australian & New Zealand Journal of Psychiatry, 52(10), 949–961. https://pure.uva.nl/ws/files/32354883/0004867418791257.pdf
42. Psychiatric News (2021). Dimensional model of personality disorder incorporated into ICD-11. https://psychnews.psychiatryonline.org/doi/10.1176/appi.pn.2021.10.33
43. Schaich, A., Braakmann, D., Richter, A., et al. (2020). Experiences of patients with borderline personality disorder with imagery rescripting in the context of schema therapy — a qualitative study. Frontiers in Psychiatry, 11, 550833. https://pure.uva.nl/ws/files/55180033/fpsyt_11_550833.pdf
44. ScienceDaily (10 February 2014). Schema therapy for personality disorders proven effective (Maastricht University press release). https://www.sciencedaily.com/releases/2014/02/140210101955.htm
45. Storebø, O. J., Stoffers-Winterling, J. M., Völlm, B. A., et al. (2020). Psychological therapies for people with borderline personality disorder. Cochrane Database of Systematic Reviews, CD012955.pub2. https://doi.org/10.1002/14651858.CD012955.pub2
46. University of Iowa, Department of Psychiatry (2017). New book: UI group treatment for borderline personality disorder goes global. https://psychiatry.medicine.uiowa.edu/news/2017/01/new-book-ui-group-treatment-borderline-personality-disorder-goes-global
47. van Asselt, A. D. I., Dirksen, C. D., Arntz, A., et al. (2008). Out-patient psychotherapy for borderline personality disorder: cost-effectiveness of schema-focused therapy v. transference-focused psychotherapy. British Journal of Psychiatry, 192(6), 450–457.
48. Weertman, A., & Arntz, A. (2007). Effectiveness of treatment of childhood memories in cognitive therapy for personality disorders. Behaviour Research and Therapy, 45, 2133–2143.
49. Wibbelink, C. J. M., Kamphuis, J. H., Sinnaeve, R., et al. (2026). Dialectical behavior therapy vs schema therapy for patients with borderline personality disorder. JAMA Psychiatry (online, April 2026). https://doi.org/10.1001/jamapsychiatry.2026.0418 ; abstract summaries: https://reachmd.com/news/dialectical-behavior-therapy-vs-schema-therapy-for-borderline-personality-disorder/2487840/ · https://medicaldialogues.in/psychiatry/news/dialectical-behavior-therapy-and-schema-therapy-show-comparable-effectiveness-in-borderline-personality-disorder-jama-169867
50. Wikipedia. Jeffrey Young (psychologist); Schema therapy. https://en.wikipedia.org/wiki/Jeffrey_Young_(psychologist) · https://en.wikipedia.org/wiki/Schema_therapy (secondary source) · PESI. Jeffrey Young, PhD — speaker biography. https://www.pesi.com/speaker/jeffrey-young-7021
51. Young, J. E. (1999). Young Schema Questionnaire: Long Form (2nd ed.), copyright notice. https://schematherapy.com/id115.htm
52. Young, J. E., Klosko, J. S., & Weishaar, M. E. (2003). Schema Therapy: A Practitioner's Guide. Guilford Press. Russian translation: Moscow, Dialektika, 2020.
53. Zanarini, M. C., Frankenburg, F. R., Reich, D. B., & Fitzmaurice, G. (2010). Time to attainment of recovery from borderline personality disorder and stability of recovery: a 10-year prospective follow-up study. American Journal of Psychiatry, 167, 663–667. https://doi.org/10.1176/appi.ajp.2010.10030308

If you need help now

If your life is in danger, call your local emergency number: 112 in most of Europe, 911 in the US and Canada, 999 in the UK, 000 in Australia. For support lines by country, see Help now.

Image credits

This page uses no third-party images.