Dialectical Behavior Therapy for Adolescents With Bipolar Disorder – Review of a recent JAMA Psychiatry article

January 2, 2024

Our Behavioral Tech Institute Board member, Richard B. Stuart, wrote a review of a significant study recently published in JAMA Psychiatry: Dialectical Behavior Therapy for Adolescents With Bipolar Disorder. We share this as an example of important new progress DBT therapists will want to know.

A recent JAMA Psychiatry article by Dr. Tina Goldstein and her University of Pittsburgh collaborators offered an important evaluation of the effectiveness of DBT in reducing suicide attempts by adolescents.

Studies of treatment outcomes with high-risk populations are extremely challenging. Challenges in design, measurement, and data access and interpretation abound in these studies, and this team worked hard to meet them. For example, ethical considerations preclude randomly assigning high risk populations to treatment vs. no-treatment conditions. Accordingly, this study assigned 47 and 53 mostly white, female 14- to 18-year-olds respectively to what became an average of 23 sessions of DBT (14.3 individual and 9.8 family) or an average of 13.1 of sessions of standard of care (SOC) psychotherapy. With unusual precision, the fidelity of the DBT sessions was carefully monitored. Although not monitored for their adherence, the SOC sessions, which included psychoeducation, cognitive behavioral, family, and supportive methods, were assessed to determine that they did not include DBT skill training. Also, because measures may be outwardly similar but have nuanced differences, the researchers wisely used both the Columbia Suicide Severity Rating Scale (Poser et al, 2022); and the ALIFE Self-Injurious/Suicidal Behavior Scale (Goldstein et al, 2012).

The two groups were reasonably well matched at intake, although the DBT group had a slightly higher rate of suicide attempts and the SOC group had slightly higher psychopathology. Due to dropouts and missing data, 12 month follow up data were available for only 32 and 22 DBT and SOC, respectively.

Although the difference in suicide attempts was clinically small, it was, however, statistically significant with the DBT group having fewer suicide attempts than the SOC group. This led the authors to conclude that “DBT may … be considered within a clinical staging approach to treatment for early onset bipolar disorder for youth at increased risk of suicidal behavior.”

Along with expected future publications (personal communication), Dr. Goldstein et al. have offered an excellent model for reporting the outcome of DBT treatment. To strengthen research in this area, future studies should control for any differences in medication that might influence outcomes, the quality of DBT, and any alternative intervention should be determined. Including follow-ups of at least 24 or months is important as a means to evaluate the sustainability of treatment effects. Data on dropouts should be included both to assess any possible adverse effects of intervention and to learn what might be done to better maintain participants’ motivation to complete treatment. In addition, it is essential to discuss the clinical versus statistical significance of outcomes since the former are of greatest interest to the public. These concerns aside, this study is a valuable step toward building empirical support for extending the reach of DBT to highly vulnerable and often treatment resistant populations. It is hoped that other researchers will follow in the footsteps of this ground-breaking study.

Interested to read more about DBT? Read here for this blog about the need to increase access to treatment.

Goldstein TR, Merranko J, Rode N, et al. Dialectical Behavior Therapy for Adolescents With Bipolar Disorder A Randomized Clinical Trial.   JAMA Psychiatry. Published online September 13, 2023. doi:10.1001/jamapsychiatry.2023.3399

Goldstein TR, Ha W, Axelson DW, et al. Predictors of prospectively examined suicide attempts among youth with bipolar disorder.  Arch. Gen. Psychiatry. 2012:69(1);1113-1122.

Posner K, Brown GK, Stanley R, et al. The Columbia Suicide Severity Rating Scale: iinitial validity and internal consistency findings from three multisite studies with adolescents and adults.  Am J Psychiatry, 2011;168(2):1266-1267.

Richard B. Stuart, PhD earned a doctorate in psychology and social work at Columbia University in 1965, Since then, he has continuously taught in departments of psychiatry, family medicine, psychology, and/or social work, at the Universities of Michigan, British Columbia, Utah, and Washington. He is a diplomate in the American Board of Professional Psychology and a fellow in two divisions of the American Psychological Association and the Association for Behavioral and Cognitive Therapy. He was an early board member, and President, of the Association for the Advancement of Behavior Therapy (now ABCT). He has offered behavior therapy to a wide range of patients for more than 50 years. He has authored or co-authored more than 100 publications including several books, e.g. Trick or Treatment, Slim Chance in a Fat World, and Helping Couples Change. For the past 15 years he has been active in promoting patients’ autonomy in end-of-life care by creating advance care planning documents such as Conditional Medical Orders and 6 Steps Living Will. He is devoted to developing innovative applications of the principles of dialectic behavior therapy in services to individuals, organizations, and communities in efforts to meet the needs of diverse populations.


Disclaimer: The Behavioral Tech Institute blog is designed to facilitate the sharing of ideas, experiences, and insights related to Dialectical Behavior Therapy (DBT). The content and views expressed in the articles, comments, and linked resources are those of the individual authors and do not necessarily reflect the views, policies, or positions of Behavioral Tech Institute or staff. Content is provided for information and discussion purposes only and is not intended as professional advice. Contributors to the Behavioral Tech Institute blog are independent, and their participation does not represent an endorsement by Behavioral Tech Institute.