Effectiveness of Predominantly Group Schema Therapy and Combined Individual and Group Schema Therapy for Borderline Personality Disorder: A Randomized Clinical Trial
Open accessAbstract
IMPORTANCE: Schema therapy (ST), delivered either in an individual or group format, has been compared with other active treatments for borderline personality disorder (BPD). To our knowledge, the 2 formats have not been compared with treatment as usual (TAU) or with each other. Such comparisons help determine best treatment practices. OBJECTIVE: To evaluate whether ST is more effectively delivered in a predominantly group or combined individual and group format and whether ST is more effective than optimal TAU for BPD. DESIGN, SETTING, AND PARTICIPANTS: In this multicenter, 3-arm randomized clinical trial conducted at 15 sites in 5 countries (Australia, Germany, Greece, the Netherlands, and the UK), outpatients aged 18 to 65 years who had BPD were recruited between June 29, 2010, and May 18, 2016, to receive either predominantly group ST (PGST), combined individual and group ST (IGST), or optimal TAU. Data were analyzed from June 4, 2019, to December 29, 2021. INTERVENTIONS: At each site, cohorts of 16 to 18 participants were randomized 1:1 to PGST vs TAU or IGST vs TAU. Both ST formats were delivered over 2 years, with 2 sessions per week in year 1 and the frequency gradually decreasing during year 2. Assessments were collected by blinded assessors. MAIN OUTCOMES AND MEASURES: The primary outcome was the change in BPD severity over time, assessed with the Borderline Personality Disorder Severity Index (BPDSI) total score. Treatment retention was analyzed as a secondary outcome using generalized linear mixed model survival analysis. RESULTS: Of 495 participants (mean [SD] age, 33.6 [9.4] years; 426 [86.2%] female), 246 (49.7%) received TAU, 125 (25.2%) received PGST, and 124 (25.0%) received IGST (1 of whom later withdrew consent). PGST and IGST combined were superior to TAU with regard to reduced BPD severity (Cohen d, 0.73; 95% CI, 0.29-1.18; P < .001). For this outcome, IGST was superior to TAU (Cohen d, 1.14; 95% CI, 0.57-1.71; P < .001) and PGST (Cohen d, 0.84; 95% CI, 0.09-1.59; P = .03), whereas PGST did not differ significantly from TAU (Cohen d, 0.30; 95% CI, -0.29 to 0.89; P = .32). Treatment retention was greater in the IGST arm than in the PGST (1 year: 0.82 vs 0.72; 2 years: 0.74 vs. 0.62) and TAU (1 year: 0.82 vs 0.73; 2 years: 0.74 vs 0.64) arms, and there was no significant difference between the TAU and PGST arms (1 year: 0.73 vs 0.72; 2 years: 0.64 vs 0.62). CONCLUSIONS AND RELEVANCE: In this randomized clinical trial, IGST was more effective and had greater treatment retention compared with TAU and PGST. These findings suggest that IGST is the preferred ST format, with high retention and continuation of improvement in BPD severity after the completion of treatment. TRIAL REGISTRATION: trialregister.nl Identifier: NTR2392.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
ST and its trauma processing component ImRs are effective for disorders resulting from interpersonal trauma in childhood, including complex PTSD and personality disorders.
“First, ST as a whole, as well as its trauma processing component, Imagery Rescripting (ImRs), are effective for disorders that result from interpersonal trauma in childhood, including complex PTSD and personality disorders [e.g., (13–17)].”
From Schema therapy for Dissociative Identity Disorder: a case report (2023) · cites this paper as [13]
The Conclusions report that combined individual and group schema therapy reduced borderline personality disorder severity more than treatment as usual. The Introduction and Discussion connect schema therapy with processing adverse childhood experiences, but this trial does not establish that Imagery Rescripting is independently effective or that either treatment is effective for complex PTSD or personality disorders generally. Page numbers were not provided in our copy. (LLM, conf=0.88)
Experimental — accuracy not yet measuredThe sample size of N = 73 suffices to reach more than 80% power to detect at a significance level of 0.05 a difference of 30% in dropout rates during the first year between 20% and 50%.
“Moreover, the sample size of N = 73 suffices to reach more than 80% power to detect at a significance level of 0.05, a difference of 30% in dropout rates during the first year between 20% [which is the approximate dropout rate of ST in individual-group format at 1 year; (26)] and 50%, which would be a substantial and alarming increase in the experimental treatment (25, 28).”
From Timing of imagery rescripting during schema therapy for borderline personality disorder: the LUCY trial (2023) · cites this paper as [26]
The planned sample of at least N = 73 participants, with seven repeated measurements and an assumed average correlation of 0.5, provides 80% power to detect a medium effect size of d = 0.5 between treatment conditions.
“At least N = 73 participants were planned to be included. Using the approximation proposed by Twisk (25), and assuming an average correlation of 0.5 between measurements [as in (26)], this suffices, given the seven repeated measurements, to reach 80% power to detect a medium effect size of d = 0.5 between the treatment conditions.”
From Timing of imagery rescripting during schema therapy for borderline personality disorder: the LUCY trial (2023) · cites this paper as [26]
Citation (26) is attached specifically to the assumed average correlation of 0.5 between measurements. The Methods sections (Study Design and Statistical Analysis; page unavailable) describe a different power analysis and repeated-measures model, but do not report that correlation. Searches for the correlation and the claim’s power-analysis figures found no supporting passage in our copy. (LLM, conf=0.72)
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
The citation lists haven’t been fetched yet
References 26 · Cited by 114
- Authors
- Arntz, A., Jacob, G. A., Lee, C. W., Brand-de Wilde, O. M., Fassbinder, E., Harper, R. P., Lavender, A., Lockwood, G., Malogiannis, I. A., Ruths, F. A., Schweiger, U., Shaw, I. A., Zarbock, G., & Farrell, J. M.
- Affiliations
- University of Amsterdam (NL)
- University of Freiburg (DE)
- The University of Western Australia (AU)
- De Viersprong, the Netherlands Institute for Personality Disorders, Halsteren, the Netherlands
- Christian-Albrechts-Universität zu Kiel (DE)
- Bradford District Care NHS Foundation Trust (GB)
- South London and Maudsley NHS Foundation Trust (GB)
- Schema Therapy Institute Midwest, Kalamazoo, Michigan
- National and Kapodistrian University of Athens (GR)
- University of Lübeck (DE)
- Schema Therapy Institute Midwest, Indianapolis, Indiana
- Institut für Verhaltenstherapie-Ausbildung Hamburg (DE)
- Indiana University – Purdue University Indianapolis (US)
- Corresponding author
- Arntz, Arnoud
- ORCID
- Venue
- JAMA Psychiatry
- Publisher
- American Medical Association (AMA)
- Published
- 2022-04-01
- Volume
- 79
- Issue
- 4
- Pages
- 287
- ISSN
- 2168-622X (ISSN-L)
- PMC
- PMC8892362
- OpenAlex
- W4214821242
- Cited by
- 114
- References
- 26
- Citations by year
- 2026: 26 · 2025: 26 · 2024: 30 · 2023: 18 · 2022: 14
- FWCI
- 20.08
- Type
- journal article
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- cc-by
- Locations
- JAMA Psychiatry (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- UvA-DARE (University of Amsterdam) (acceptedVersion, cc-by)
- IUScholarWorks (Indiana University) (submittedVersion, cc-by)
- UWA Profiles and Research Repository (UWA) (submittedVersion, other-oa)
- Data Archiving and Networked Services (DANS) (submittedVersion, other-oa)
- Funders
- South London and Maudsley NHS Foundation Trust
- National University of Ireland
- Institute for the Promotion of Teaching Science and Technology
- Stichting Volksbond Rotterdam
- King's College London
- University of Oxford
- Australian Rotary Health
- Australian Psychological Society
- Deutsche Forschungsgemeinschaft
- Murdoch University
- ZonMw
- Universiteit van Amsterdam
- Universiteit Maastricht
- Else Kröner-Fresenius-Stiftung
- Nederlandse Organisatie voor Wetenschappelijk Onderzoek
- Universität zu Lübeck
- National and Kapodistrian University of Athens
- Achmea
- Stichting tot Steun Vereniging tot Christelijke Verzorging van Geestes- en Zenuwzieken
- Institute of Psychiatry, Psychology and Neuroscience, King’s College London
- Topics
- Social Sciences › Psychology › Clinical Psychology › Personality Disorders and Psychopathology
- Social Sciences › Psychology › Clinical Psychology › Psychotherapy Techniques and Applications
- Social Sciences › Psychology › Experimental and Cognitive Psychology › Anxiety, Depression, Psychometrics, Treatment, Cognitive Processes
- Keywords
- borderline personality disorder, schema therapy, group therapy, individual psychotherapy, treatment effectiveness, treatment retention, usual care, outpatients, psychotherapy
- MeSH terms
- Schema Therapy, Adolescent, Adult, Aged, Borderline Personality Disorder, Female, Germany, Humans, Middle Aged, Outpatients, Psychotherapy, Group, Treatment Outcome, Young Adult
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
