Article Text

Download PDFPDF
Psychological interventions
Dialectical behaviour therapy with skills training seems to be more effective in reducing non-suicidal self-injury
  1. Kate Andreasson
  1. Psychiatric Research Unit, Psychiatric Centre North Zealand, Copenhagen University Hospital, Denmark; Kate.trein.andreasson{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

What is already known on this topic

Patients with borderline personality disorder (BPD) have a 10% lifetime risk of death by suicide.1 Therefore, prevention of self-harm is an important target for treatment in this population. A recent Cochrane review2 concludes that dialectical behaviour therapy (DBT) is the most studied psychotherapeutic treatment for patients with BPD. The authors found that there were indications of beneficial effects for many psychotherapies, including DBT, for the BPD core symptoms and reduction of self-harm. One of the meta-analyses conducted regarding DBT compared with treatment as usual (TAU) showed a moderate statistically significant effect, indicating a beneficial effect in favour of DBT.2 Owing to the multicomponent nature of DBT treatment, Linehan and colleagues made an important study to compare DBT skills training with other DBT components, and tried to identify what is needed to obtain a positive outcome.

Methods of the study

The study was designed as a three arm, single-blinded randomised clinical superiority trial. It was conducted from April 2004 through January 2010. Participants had to have a diagnosis of BPD according to Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) in order to be included. They also had to have made at least two suicide attempts and/or to have had non-suicidal self-injury (NSSI) episodes during the past 5 years, at least one suicide attempt or NSSI act in the 8-week period before entering the study, and at least one suicide attempt in the past year. They were excluded if they had an IQ score of <70 on the Peabody Picture Vocabulary Test—Revised,3 current psychotic or bipolar disorder (DSM-IV axis I), seizure disorder requiring medication, or required primary treatment for another life-threatening condition. The interventions offered were: (1) standard DBT, (2) DBT skills training+case management (DBT-S) and (3) DBT individual therapy+activity-based support group (DBT-I). It was predicted that standard DBT would be significantly better than DBT-S, and that it would be better than DBT-I. The main outcomes were frequency and severity of suicide attempts and NSSI. Other outcomes in the study were suicidal behaviour, reasons for living, use of crisis services, severity of depressive symptoms and anxiety. Ninety-nine women were included and randomised through a computerised adaptive randomisation procedure, which matched different parameters. The participants were treated in the three intervention groups for 1 year, and the follow-up period was 1 year after treatment. In each group, 33 participants were included and assessed quarterly for 2 years by blinded assessors.

What does this paper add?

  • This is the first study investigating the effect of the different components of the DBT. In particular, the comparison of DBT skills training and other components of DBT, such as DBT individual therapy in conformity with the variety of ways the treatment is delivered in ‘real-world’ clinical settings, is very relevant.

  • All three versions of DBT (Standard DBT, DBT-S and DBT-I) were comparably effective at reducing suicidality among those patients with high risk for suicide.

  • DBT interventions including DBT skills training were more effective in reducing NSSI acts than DBT intervention without skills training (post-treatment NSSI episodes: 58.1% in standard DBT group, 55.6% in DBT-S group and 63% in DBT-I group).


  • The power calculation is not mentioned in the article or in the trial protocol (reported as supplement 1 to the published paper). As Linehan and colleagues write in their report, we must assume that the study was underpowered and not powered to assess equivalence between standard DBT and DBT-S.

  • Furthermore, there was a higher dropout rate in the DBT-I and DBT-S groups than in the standard DBT group. The reason for the high dropout rates is not mentioned. When this is combined with the low power limits, the results should be further interpreted with caution.

What next in the research?

Well-powered trials to assess the equivalence between standard DBT, DBT-S and DBT-I, and studies with longer follow-up period than 1 year, should be considered in future research in order to investigate whether the effect standard DBT and DBT-S on NSSI episodes are sustained.

Do these results change your practices and why?

Previous research has shown that patients with BPD and self-harm behaviour benefit from DBT treatment. The findings of the study show that the effect of skills training (standard DBT and DBT-S) has a greater improvement on NSSI and depression than DBT without skills training. This study actually supports the different forms of delivery of DBT interventions, especially those combined with skills training. Standard DBT is considered a treatment of long duration and high cost; therefore, could the use of DBT skills training be a reasonably successful as well as cost-effective treatment alternative?



  • Competing interests None declared.