Dialectical Behavioural Therapy for the “Difficult to Treat” Eating Disorder Patients

If there is anything we’ve learned over the  many years of eating disorder research, it is that eating disorders are extremely complex. Often, this complexity is intensified by comorbidities, including post-traumatic stress disorder, depression, and “personality disorders.” Unfortunately, individuals whose disorders are labeled persistent, chronic, or “difficult to treat” may be even less likely to receive the treatment and support they require, deserve, and desire.

“Standard” approaches to eating disorder treatment, such as cognitive behaviour therapy (CBT), may prove ineffectual for these individuals. In a recent article, Federici & Wisniewski (2013) reflected on the difficulty of treating patients whose eating disorders are accompanied by other mental health issues. They noted that focusing on ED symptoms alone generally fails to achieve treatment goals, as behaviours associated with other disorders often decrease ED treatment effectiveness. This situation may leave both patients and clinicians feeling burnt out and unsatisfied (to say the least).

Resultantly, clinicians are turning to alternative therapeutic options for treating more complex cases, including dialectical behaviour therapy (DBT). Federici & Wisniewski explored preliminary data from pre/post case series (in-depth looks at patients’ experiences and outcomes) from a treatment program employing the “Multidiagnostic Eating Disorder-DBT (MED-DBT) Program” specifically designed to address the needs of patients with psychiatric comorbidities.


Dialectical behavior therapy uses elements of cognitive behaviour therapy in combination with a number of techniques designed to enhance psychosocial functioning, including:

  • Distress tolerance
  • Acceptance
  • Mindfulness

Marsha Linehan (1993) developed DBT in as a way to respond to the needs of patients with borderline personality disorder. The overall idea is to move beyond black and white thinking and to support participants in developing strategies to cope with perceived crises. In its’ most basic terms, participants work with a therapist to progress through a series of goals as they reduce black and white thinking.

Specific strategies include, among others:

  • Contingency management (i.e. validation and reinforcement)
  • Behaviour chain analysis (i.e. looking at events leading up to a particular behaviour)
  • Dialectical stance (i.e. examining how behaviours and thoughts are interrelated and taking a “middle view”)
  • Skills training

There is a lot more to DBT than I could capture here, but if you’re interested you can find out more information about DBT here, here and here.


Federici, Wisniewski & Ben-Porath (2012) developed MED-DBT, an intervention specifically designed for ED sufferers with comorbidities. MED-DBT practitioners placce an emphasis on collaborating with patients to address both ED and comorbid condition behaviours.

The intervention was primarily group-therapy-based and delivered in a 6 month, intensive outpatient (e.g., 3 hours per day, 3-5 days a week) or day treatment (e.g., 6 hours per day, 5 days a week) model. Patients could step down from day treatment to intensive outpatient as they progressed through the program. In addition to group sessions, patients worked with individual counselors, nutrition professionals, and psychiatrists to follow Linehan’s DBT model, including after-hours telephone therapy .


Patients were eligible to participate if they:

  • Had previously attempted “standard” treatment for their eating disorder without seeing significant decreases in their symptoms

And at least one of the following:

  • Were diagnosed with comorbid disorders
  • Struggled with emotion regulation
  • Were unable to maintain symptom reduction following “traditional” treatment
  • Displayed behaviours that impacted progress in “traditional therapy” settings (e.g., being absent from treatment on a regular basis)

The sample size was quite low: 7 participants were included in this study. However, for a series of case reports, this is not unexpected. Participants ranged in age from 20-31, with a mean age of 24. Diagnoses varied, and included AN (both subtypes), BN, and eating disorder not otherwised specified. These participants also had a relatively long duration of illness, with a mean age of onset of 12 (range 8-18). All participants had been treated at least 3 times prior to engaging in MED-DBT (average 6.5, range 3-10).

Prior to the intervention, all patients completed up to 4 “commitment and orientation” sessions to determine program fit. Though MED-DBT is often seen as a more flexible mode of treatment, non-negotiable elements include:

  • Willingness to stay alive (e.g., through eliminating self-injurious/suicidal behaviours)
  • Willingness to work on ED symptoms
  • Willingness to remain in the program for 6 months

Patients were required to attend all sessions and were discharged if they missed more than one full week of program.

Outcome Measures

The researchers administered the Eating Disorder Examination Questionnaire (EDE-Q), the Deliberate Self-Harm Inventory (DSHI) and tested for body weight and medical stability (e.g., through blood sodium levels and EKG results) to gauge intervention effectiveness. They also looked at whether participants stayed in treatment, whether (and how often) they were admitted to psychiatric hospitals during the intervention, and whether clients and clinicians were satisfied with the program.


Of the 7 participants, 6 completed the full 6-month intervention. The patient who dropped out was discharged with medical advice and with an invitation to return to treatment following medical stabilization due to heart issues that resulted in repeated hospitalization.

Overall, this very small study provides preliminary supports for the effectiveness and acceptability of MED-DBT. Notably:

  • Weight gain: All underweight patients increased in BMI, though not all reached a BMI of over 18 by the end of the 6 months
  • Symptom reduction: Reductions in binge eating, vomiting and restriction (note that symptoms were decreased but not eliminated in some participants at discharge)
  • Comorbid conditions: Comorbid disorder symptom reduction, including reductions in suicidal/self-injurious behaviours (note that 5 participants reported no symptoms at discharge, 1 reported minimal change and 1 an increase)
  • Medical stability: All patients were medically stable in the month preceding discharge
  • Treatment retention: As noted above, 6 of 7 participants completed the full 6 months

From clinician reports, the authors also observed that participants were willing to engage in treatment and clinicians felt supported and experienced a reduction in burnout.


While these results point to encouraging outcomes for MED-DBT, there are a number of limitations associated with this study, some of which are noted by the authors. As the authors acknowledged, case series analyses offer only preliminary results as the sample size is extremely small and there is no control group. Of note, too, is that the authors themselves developed the intervention.

Additionally, the participants in this study were all Caucasian women, and all were treated at the Cleveland Center for Eating Disorders. It would be interesting to see if the results hold when using a larger sample size, a more diverse group of individuals, and in different locations.

Overall, the idea of MED-DBT is intriguing. While many treatment programs make use of DBT among other treatment modalities, the emphasis placed on DBT in this model is far more central and pronounced.

Given that DBT was initially developed for individuals with borderline personality disorder, it is important to note that 6 of the 7 participants included in this case series analysis were diagnosed with either borderline personality disorder or borderline personality traits.

I would like to see a study that examines the use of MED-DBT with participants who have not been diagnosed with borderline personality disorder to see if similar results emerge. It might be more appropriate, at this point, to suggest that these results bolster effects of treatment among individuals with BPD comorbidity, rather than “multidiagnostic eating disorders” more broadly.

It is particularly interesting that borderline personality disorder is often linked to being labeled as “difficult to treat” or “treatment resistant,” given that somewhere between 26-54% of those diagnosed with eating disorders struggle with borderline personality disorder or borderline personality traits(Sansone & Levitt, 2006).

I’ll try not to get into my feminist-ing about this, but it is surprising to me that this comorbidity has not been addressed in any great detail until relatively recently, particularly because the diagnosis of BPD may result in patients being dismissed as unwilling to engage in treatment (Gremillion, 2003). At any rate, it is encouraging that clinician-researchers such as Federici, Wisniewski & Ben-Porath are exploring ways to improve the experience for both patients and clinicians.


Federici A, & Wisniewski L (2013). An intensive DBT program for patients with multidiagnostic eating disorder presentations: a case series analysis. The International Journal of Eating Disorders, 46 (4), 322-31 PMID: 23381784

Federici A, Wisniewski L & Ben-Porath, D.D. (2012). Development and feasibility of an intensive DBT outpatient program for multi-diagnostic clients with eating disorders. Journal of Counseling and Development, 90, 330–338.

Liked it? Take a second to support Andrea on Patreon!


Andrea is a PhD candidate focusing on individual, familial, and health care definitions and experiences of eating disorder recovery. She has an MSc in Family Relations and Human Development and a BA in Sociology. In her Masters research, she used qualitative and arts-based approaches (digital storytelling) to explore the experiences of young women in recovery from eating disorders. Andrea has recovered from EDNOS. She can be reached at andrea[at]scienceofeds[dot]org.

Leave a Reply

Your email address will not be published.

10 − 5 =