Hostname: page-component-78c5997874-8bhkd Total loading time: 0 Render date: 2024-10-30T05:21:54.188Z Has data issue: false hasContentIssue false

Results of the first randomized controlled trial of integrated cognitive-behavioral therapy for eating disorders and posttraumatic stress disorder

Published online by Cambridge University Press:  07 December 2021

Kathryn Trottier*
Affiliation:
Centre for Mental Health, University Health Network, Toronto, Canada Department of Psychiatry, University of Toronto, Toronto, Canada
Candice M. Monson
Affiliation:
Department of Psychology, Ryerson University, Toronto, Canada
Stephen A. Wonderlich
Affiliation:
Sanford Health, Fargo, North Dakota, USA Sanford Research, Fargo, North Dakota, USA Department of Psychiatry and Behavioral Science, University of North Dakota School of Medicine and Health Sciences,Grand Forks,North Dakota, USA
Ross D. Crosby
Affiliation:
Sanford Research, Fargo, North Dakota, USA Department of Psychiatry and Behavioral Science, University of North Dakota School of Medicine and Health Sciences,Grand Forks,North Dakota, USA
*
Author for correspondence: Kathryn Trottier, E-mail: kathryn.trottier@uhn.ca
Rights & Permissions [Opens in a new window]

Abstract

Background

Eating disorders (EDs) and posttraumatic stress disorder (PTSD) frequently co-occur and can share a functional relationship. The primary aim of this initial randomized controlled trial was to determine whether integrated cognitive-behavioral therapy (CBT) for co-occurring ED-PTSD was superior to standard CBT for ED in improving PTSD symptoms. Intervention safety and desirability, as well as the relative efficacy of the treatments in improving anxiety, depression, and ED symptomatology, were also examined.

Methods

Following a course of intensive ED treatment, individuals with ED-PTSD were recruited to participate and randomized to integrated CBT for ED-PTSD or standard CBT for ED. The sample consisted of 42 individuals with a range of ED diagnoses. Outcomes were assessed at end-of-treatment, 3-, and 6-month follow-up using interview and self-report measures.

Results

Mixed models revealed significant interactions of time and therapy condition on clinician-rated and self-reported PTSD symptom severity favoring Integrated CBT for ED-PTSD. Both treatments were associated with statistically significant improvements in PTSD, anxiety, and depression. Improvements were maintained at 3- and 6-month follow-up. There was good safety with both interventions, and satisfaction with both treatments was high. However, there was a stronger preference for integrated treatment.

Conclusions

Integrating CBTs for PTSD and ED following intensive ED treatment is safe, desirable, and efficacious for improving PTSD symptoms. Future studies with larger sample sizes are needed to determine whether Integrated CBT for ED-PTSD provides benefits over standard CBT for ED with respect to ED outcomes.

Type
Original Article
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution and reproduction, provided the original article is properly cited.
Copyright
Copyright © The Author(s), 2021. Published by Cambridge University Press

Posttraumatic stress disorder (PTSD) frequently co-occurs with eating disorders (EDs) in the general population (Dansky, Brewerton, Kilpatrick, & O'Neil, Reference Dansky, Brewerton, Kilpatrick and O'Neil1997; Hudson, Hiripi, Pope, & Kessler, Reference Hudson, Hiripi, Pope and Kessler2007) and in treatment-seeking samples, with the highest rates observed in intensive ED treatment settings (inpatient, residential, partial hospitalization; e.g. Brewerton et al. Reference Brewerton, Perlman, Gavidia, Suro, Genet and Bunnell2020; Mitchell, Singh, Hardin, & Thompson-Brenner, Reference Mitchell, Singh, Hardin and Thompson-Brenner2021; Vanzhula, Calebs, Fewell, & Levinson, Reference Vanzhula, Calebs, Fewell and Levinson2019). Although psychiatric comorbidity is common in both PTSD and EDs (Hudson et al., Reference Hudson, Hiripi, Pope and Kessler2007; Kessler, Sonnega, Bromet, Hughes, & Nelson, Reference Kessler, Sonnega, Bromet, Hughes and Nelson1995), co-occurring ED and PTSD (i.e. ED-PTSD) is of particular concern because of the potential for a functional relationship between the disorders that may increase the likelihood that the disorders maintain one another (Karr et al., Reference Karr, Crosby, Cao, Engel, Mitchell, Simonich and Wonderlich2013; Liebman et al., Reference Liebman, Becker, Smith, Cao, Keshishian, Crosby and Thomas2020; Vanzhula et al., Reference Vanzhula, Calebs, Fewell and Levinson2019), and result in high rates of relapse following standard ED treatment (Berends, Boonstra, & van Elburg, Reference Berends, Boonstra and van Elburg2018; Yu, Agras, & Bryson, Reference Yu, Agras and Bryson2013). ED-PTSD is also associated with greater ED severity (Rijkers, Maartje, van Hoeken, & Hoek, Reference Rijkers, Schoorl, van Hoeken and Hoek2019), and other psychopathology (Brewerton et al., Reference Brewerton, Perlman, Gavidia, Suro, Genet and Bunnell2020; Scharff, Oritz, Forrest, & Smith, Reference Scharff, Oritz, Forrest and Smith2019; Trottier, Reference Trottier2020). Emerging research suggests that individuals with co-occurring PTSD may be more likely to end ED treatment early (Trottier, Reference Trottier2020), have poorer end-of-treatment (EoT) ED outcomes (Hazzard et al., Reference Hazzard, Crosby, Crow, Engel, Schaefer, Brewerton and Wonderlich2021; Serra et al., Reference Serra, Kiekens, Tarsitani, Vrieze, Bruffaerts, Loriedo and Vanderlinden2020), and relapse following ED treatment (Scharff, Oritz, Forrest, Smith, & Boswell, Reference Scharff, Oritz, Forrest, Smith and Boswell2021). Consistent with these findings, many frontline ED clinicians perceive trauma-related symptoms as a major obstacle to successful ED treatment and believe addressing trauma-related symptoms in individuals with EDs is important (Trottier, Monson, Wonderlich, MacDonald, & Olmsted, Reference Trottier, Monson, Wonderlich, MacDonald and Olmsted2017).

Specific trauma-focused cognitive-behavioral therapies (CBTs) are the first-line treatments for PTSD (e.g. American Psychological Association, 2017). However, individuals with ED-PTSD are not often offered these treatments, in part, due to concerns that they are not ready for trauma-focused therapy, and that it may lead to worsening of ED, trauma-related, and other mental health symptoms (Trottier et al., Reference Trottier, Monson, Wonderlich, MacDonald and Olmsted2017). Despite these perceived challenges, there have been repeated calls for the development and evaluation of integrated CBTs for ED-PTSD (e.g. Brewerton, Reference Brewerton2007; Holzer, Uppala, Wonderlich, Crosby, & Simonich, Reference Holzer, Uppala, Wonderlich, Crosby and Simonich2008; Rijkers et al. Reference Rijkers, Schoorl, van Hoeken and Hoek2019) and many clinicians believe concurrent approaches are most appropriate (Trottier et al., Reference Trottier, Monson, Wonderlich, MacDonald and Olmsted2017).

To date, there is one published study of integrated psychotherapy for ED-PTSD. Trottier and colleagues developed integrated CBT for ED-PTSD by integrating cognitive processing therapy (CPT) for PTSD including a written trauma account (Resick, Monson, & Chard, Reference Resick, Monson and Chard2016) with interventions from enhanced CBT for ED (CBT-E; Fairburn, Cooper, & O'Connor, Reference Fairburn, Cooper, Shafran, Bohn, Hawker, Murphy, Straebler and Fairburn2008) – two leading evidence-based CBTs. In an uncontrolled study, integrated CBT for ED-PTSD was provided to 10 women following a course of intensive hospital-based ED treatment focused on nutritional stabilization and behavioral symptom interruption (Trottier et al., Reference Trottier, Monson, Wonderlich, MacDonald and Olmsted2017). There were statistically significant and large magnitude improvements in PTSD symptoms from baseline to EoT, and 90% remitted from their PTSD diagnosis. There were no statistically significant changes in ED psychopathology over the course of integrated CBT, and 89% of participants who were remitted from their ED at study baseline (end-of-intensive ED treatment) remained remitted from their ED at the end of integrated CBT.

The current study is the first randomized controlled trial (RCT) examining the efficacy of integrated CBT for ED-PTSD. Our primary aim was to determine whether integrated CBT for ED-PTSD was superior to standard CBT for ED in improving PTSD symptoms. We also aimed to investigate the safety and desirability of integrated CBT for ED-PTSD. We hypothesized that integrated CBT would result in greater improvements in clinician-rated PTSD symptoms at EoT and follow-up. Secondary hypotheses were that integrated CBT would also result in greater improvements in self-reported PTSD symptoms, anxiety, and depression at EoT and follow-up. Additionally, we aimed to explore the impact of integrated CBT relative to standard CBT for ED on ED outcomes at EoT and follow-up.

Method

Design

The study was a parallel groups RCT with two active treatment arms. The trial involved a hybrid efficacy–effectiveness design in which the efficacy of integrated CBT for ED-PTSD was tested in a routine ED clinical practice setting with ED program therapists and patients. Following a course of intensive ED treatment, participants were randomly assigned to either (1) standard CBT for ED or (2) integrated CBT for ED-PTSD. The number and length of sessions were standardized across treatments. Participants were assessed at study baseline, end-of-study treatment (EoT), as well as 3 and 6 months after study treatment. Assessors were independent, and blind to treatment conditions. An intent-to-treat approach to data collection and analysis was utilized.

Participants

Forty-three participants with ED-PTSD who received a course of day hospital or inpatient treatment for ED were randomized. However, one individual disclosed after randomization and before starting treatment that she had misrepresented her symptoms during the eligibility assessment and did not actually meet inclusion criteria. Thus, the individual was withdrawn and excluded resulting in a sample size of 42. Inclusion criteria were: Diagnostic and Statistical Manual of Mental Disorders, 5th Edition criteria (DSM-5; American Psychiatric Association, 2013) diagnosis of an ED at admission to intensive treatment, DSM-5 diagnosis of PTSD at study baseline, a body mass index (BMI; weight in kilograms/height in meters2) ⩾ 18.5, and stable or no psychotropic medication at study baseline. Diagnoses were assessed using evidence-based semi-structured interviews (described below). Exclusion criteria included medical conditions known to influence eating and/or weight, current substance dependence, uncontrolled bipolar or psychotic disorder, current participation in another psychosocial intervention or treatment study for ED or PTSD, and previous CPT.

Measures

The Clinician-Administered PTSD Scale-5 (CAPS-5; Weathers et al., Reference Weathers, Blake, Schnurr, Kaloupek, Marx and Keane2013) was used to make a PTSD diagnosis, and to determine PTSD remission. The CAPS-5 is a semi-structured clinician interview that assesses the severity of PTSD symptoms consistent with DSM-5 criteria (DSM-5; American Psychiatric Association, 2013). PTSD diagnostic status was based on meeting the DSM-5 symptom cluster criteria (to be counted a symptom, an item must be rated a 2 or higher). The past month CAPS-5 total severity score was the primary outcome measure. The range of scores on the CAPS-5 is 0–80, with higher scores indicating higher symptom severity. Internal consistency in the current study was adequate (α = 0.78).

The Eating Disorder Examination 17.0 (EDE; Fairburn, Cooper, and O'Connor, Reference Fairburn, Cooper and O'Connor2014) was used to assess ED diagnosis, remission, and severity of ED psychopathology. The EDE is a semi-structured clinician interview that assesses the frequency of ED behaviors and severity of ED psychopathology over the past 3 months. The past month EDE global score (measuring ED psychopathology) is a transdiagnostic measure of ED outcome, and is utilized in studies with mixed ED diagnosis samples (e.g. Fairburn et al., Reference Fairburn, Cooper, Doll, O'Connor, Bohn and Hawker2009). EDE global scores can range from 0 to 6, with higher scores reflecting greater psychopathology. Internal consistency in the current study was good (α = 0.86). ED diagnostic status also served as an ED outcome. There is currently no well-accepted definition of ED remission in the literature. Given the evidence that residual symptoms are common in remitted ED individuals (Tomba, Tecuta, Crocetti, Squarcio, & Tomei, Reference Tomba, Tecuta, Crocetti, Squarcio and Tomei2019), we defined ED remission as no more than two objective binge and/or purge episodes in the past month and a BMI of greater than 18.5 which is in the range of previous definitions used in the literature (Tomba et al., Reference Tomba, Tecuta, Crocetti, Squarcio and Tomei2019).

The Mini-International Neuropsychiatric Interview 7.0 (MINI-7; Sheehan, Reference Sheehan2015) was used to assess exclusion criteria related to other mental health symptoms. The MINI-7 is a structured clinician-administered diagnostic interview assessing the presence of DSM-5 disorders in a number of categories, including mood disorders, anxiety disorders, obsessive-compulsive disorders, and substance use disorders.

The PTSD Checklist-5 (PCL-5; Blevins, Weathers, Davis, Witte, and Domino, Reference Blevins, Weathers, Davis, Witte and Domino2015) served as a secondary measure of PTSD symptom severity. The PCL-5 is a self-report measure of PTSD symptoms over the past month corresponding with DSM-5 criteria. Scores on the PCL-5 range from 0 to 80, with higher scores indicating greater PTSD symptom severity. Participants also completed the Depression Anxiety Stress Scales (DASS; Lovibond and Lovibond, Reference Lovibond and Lovibond1995) which provided measures of depression (score range, 0–42) and anxiety (0–42) over the past week with higher scores reflecting greater severity. All self-report measures had good internal consistency, α = 0.86 for the PCL-5, α = 0.91 for DASS depression, and α = 0.89 for DASS anxiety.

Treatment desirability was assessed at EoT via an investigator-constructed questionnaire. All participants rated their satisfaction with the treatment, and participants who received integrated CBT also rated the extent to which they believed having received integrated treatment would facilitate their ED recovery using a 5-point Likert Scale ranging from not at all to completely. All participants were also asked whether they would prefer integrated treatment for their ED and PTSD or separate treatments addressing one disorder at a time, as well as the extent to which they would recommend the treatment they received to someone else.

Procedure

The first author's institutional review board approved the study protocol and all participants provided their written informed consent before eligibility assessment. Participants were recruited from the Eating Disorders Program at University Health Network, Toronto, Ontario, Canada between October 2015 and November 2018. Potential participants were identified by chart review and were provided with an overview of the study. Interested individuals then met with a member of the study staff to learn more about the study and for the informed consent process. Potential participants then underwent the eligibility/baseline assessment and eligible participants were randomized. Balanced block randomization was stratified according to ED diagnosis (anorexia nervosa v. other ED diagnosis). The study biostatistician generated the randomization, and allocation was concealed in envelopes that were opened by the study research assistant together with the participant when they were deemed eligible. Independent master's or doctoral-level clinicians conducted the CAPS-5, EDE, and MINI-7 interviews.

Treatments and treatment fidelity

Both treatments consisted of 16 individual therapy sessions delivered over 14 weeks. Sessions 1–8 were delivered at a frequency of twice per week, sessions 9–12 were delivered on a weekly basis, and sessions 13–14 were delivered at 2-week intervals. Sessions 1–3 and 15–16 were 50 min and sessions 4–14 were 90 min in length.

CBT for ED was manualized by the first author by adapting interventions from CBT-E (Fairburn et al., Reference Fairburn, Cooper, Shafran, Bohn, Hawker, Murphy, Straebler and Fairburn2008) for delivery following intensive ED treatment. Consistent with the hybrid efficacy-effectiveness trial design, the comparison condition (i.e. CBT for ED) was modeled after the treatment that was offered as part of routine care in the clinical program where the study was conducted. As in the prior pilot study, integrated CBT for ED-PTSD was a manualized treatment consisting of a combination of interventions from CPT (Resick et al., Reference Resick, Monson and Chard2016) and CBT-E. Details on the content of the treatment are described by Trottier and Monson (Reference Trottier and Monson2021). The intensive ED treatment that preceded the study treatments was focused on interrupting binge eating and purging, establishing a pattern of regular, nonrestrictive eating, and weight restoration (as relevant).

Four psychologists from the University Health Network ED program provided both of the study treatments. All therapists had been previously trained in CPT and CBT-Enhanced including attendance at in-person clinical training delivered by developers of the treatments. In order to promote adherence to the treatment protocols, therapists participated in a weekly group case consultation with the first and/or second author focused on the application of the treatments in accordance with study protocols. Therapy sessions were audio-recorded for consultation purposes and for fidelity assessment. Ten percent of therapy sessions in each condition were rated for manual adherence and therapist competence by four independent raters trained to reliability. An established CPT fidelity rating scale (Wiltsey Stirman, Monson, & Resick, Reference Wiltsey Stirman, Monson and Resick2017) was used to rate competence (range 0–6, with 0 indicating no evidence of competence and 6 indicating exceptional competence) and adherence (range 0–1, with 0 indicating a lack of adherence and 1 indicating adherence) with respect to the CPT components of integrated CBT for ED-PTSD. The same rating scale was adapted to rate adherence and competence with respect to the CBT for ED interventions in both treatment conditions. A score of 3 or higher on the competence rating scale indicated adequate competence. Adherence was very strong, with 100% of essential elements delivered in the CBT for ED condition, and 98.6% delivered in the integrated CBT for ED-PTSD condition. One hundred percent of sessions in the CBT for ED condition, and 85% of sessions in the CBT for ED-PTSD condition had a mean competence rating over 3.

Statistical analyses

Power analysis was conducted to ensure adequate power to detect between-groups differences in clinician-rated PTSD at EoT and follow-up. The sample size was calculated assuming the use of multilevel modeling, an effect size of g = 0.80 based on prior studies of CBTs for PTSD (Watts et al., Reference Watts, Schnurr, Mayo, Young-Xu, Weeks and Friedman2013), a conservative estimated correlation between repeated administrations of the CAPS-5 of r = 0.65, estimated measurement attrition of 20%, and a 2-tailed test with alpha equal to 0.05. A sample size of 40 participants was estimated to yield power greater than 90% to find the expected effect.

Analyses were conducted following intent-to-treat principles. All available data at each time-point were used in the multilevel models conducted using SPSS software version 25. Multilevel modeling was conducted on each outcome with a random intercept, and fixed effects for condition, time, and the condition × time interaction, and a first-order autoregressive covariance structure to account for the association between repeated measures. The primary outcomes were mean clinician-rated PTSD symptoms (CAPS-5 scores) at EoT and follow-ups compared between treatment conditions. The secondary and exploratory outcomes (anxiety, depression, and ED psychopathology) were analyzed using the same method. Pairwise comparisons were examined if omnibus tests were significant. To examine change over the course of treatment and maintenance of treatment gains, pairwise comparisons were conducted comparing baseline scores with EoT scores, and EoT scores with scores at 3- and 6-month follow-up. We also examined reliable change (Jacobson & Truax, Reference Jacobson and Truax1991) in CAPS-5 and EDE global scores, as well as loss of diagnostic status (PTSD and ED) at each assessment point.

Results

Participant characteristics and flow

Table 1 shows the characteristics of the sample within each therapy condition at study baseline, and Fig. 1 shows the flow of participants through the study. There was no statistical difference in non-completion rate by condition [integrated CBT for ED-PTSD, n = 5 (26%); CBT for ED only, n = 8 (35%), Fisher's Exact p = 0.74). One participant withdrew before starting the study therapy because they were unwilling to accept randomization to the CBT for ED only condition. One participant was withdrawn by the study team due to repeated treatment interfering behaviors on hospital property which prevented appropriate engagement with treatment. There were no participants withdrawn for safety-related reasons. One participant in the integrated CBT for ED-PTSD condition went to the emergency department due to suicidal ideation in the context of writing her trauma account but was not admitted to the hospital. She subsequently construed her suicidal ideation as a means of avoidance that was maintaining her PTSD and went on to complete treatment with no further safety-related events. There were no suicide attempts or psychiatric hospitalizations reported by participants.

Fig. 1. Consort flow diagram.

Note. ED, eating disorder; PTSD, posttraumatic stress disorder; BMI, body mass index; CBT, cognitive-behavioral therapy. *Participant disclosed after randomization that they misrepresented their symptoms in the eligibility assessment and was subsequently deemed ineligible and excluded from analyses.

Table 1. Participant characteristics

Note. Gender, age, ethnicity, marital status, education, financial situation, ED diagnosis, and duration of ED were assessed at the time of admission to intensive ED treatment, all other variables were assessed at study baseline (i.e. at the end of intensive ED treatment). ED, eating disorder; PTSD, posttraumatic stress disorder; CBT, cognitive behavioral therapy; AN-R, anorexia nervosa-restricting subtype; AN-BP, anorexia nervosa-binge-eating/purging subtype; BN, bulimia nervosa; OSFED, other specified feeding or eating disorder; BMI, body mass index.

Mental health outcomes

There were no significant differences between conditions on the primary and secondary outcome measures at baseline. Table 2 shows estimated marginal mean scores on the outcome measures by therapy condition at each time point, as well as within-group effect sizes for baseline to EoT for each condition. Tests of fixed effects revealed significant interactions of condition and time on both clinician-rated and self-reported PTSD symptoms (CAPS-5 and PCL-5, respectively), F (3, 45.26) = 4.06, p = 0.01, and F (3, 41.65) = 7.13, p = 0.001, respectively. CAPS-5 scores decreased with both treatments, with a significantly greater decrease in the CBT for ED-PTSD condition than the CBT for ED only condition at EoT, 3-month, and 6-month follow-up. Pairwise comparisons indicated that improvements were maintained over follow-ups. The same pattern was found for PCL-5 scores. The condition by time interactions were not significant for anxiety, F (3, 45.84) = 0.64, p = 0.60, depression, F (3, 38.08) = 2.44, p = 0.08, or ED psychopathology, F (3, 34.04) = 1.55, p = 0.22. However, estimates of fixed effects indicated that improvements in anxiety from baseline to 6-month follow-up were significantly greater in the CBT for ED-PTSD than the CBT for ED condition, F (1, 19) = 5.48, p = 0.03. Between-group differences in anxiety were not significant for EoT and 3-month follow-up.

Table 2. Continuous outcomes at all-time points

Note. CAPS-5, Clinician-Administered PTSD Scale-5; PCL-5, PTSD Checklist-5; DASS, Depression Anxiety Stress Scales; EDE Global, Eating Disorder Examination 17.0 Global Score; ED, eating disorder; PTSD, posttraumatic stress disorder; EoT, end of treatment; mo, month; FU, follow up. *Denotes an effect size corresponding to a statistically significant pairwise comparison at the p < 0.05 level.

There were significant main effects of time on anxiety, depression, and ED psychopathology, F (3, 45.84) = 12.36, p < 0.001; F (3, 38.08) = 6.27, p = 0.001; and F (3, 34.04) = 3.88, p = 0.02, respectively. Pairwise comparisons revealed DASS anxiety and depression scores improved significantly with both treatments, and that changes were maintained over follow-ups. There was also a significant main effect of condition on anxiety, F (1, 37.44) = 8.11, p = 0.01. Across the two therapy conditions, pairwise comparisons revealed no significant changes in Global EDE scores over the course of treatment or follow-up.

Table 3 shows PTSD and ED diagnostic status at EoT and follow-ups. A greater proportion of participants no longer met the criteria for PTSD in the integrated CBT for ED-PTSD condition than in the CBT for ED condition at EoT and 3-month follow-up, Fisher's Exact, p = 0.002 and p = 0.03, respectively. The difference was not statistically significant at the 6-month follow-up. There were no significant differences between conditions in proportions of individuals who experienced reliable improvements and deteriorations in CAPS-5 scores over the course of treatment and at follow-ups. Overall, there were similar proportions of participants who experienced reliable improvements, deteriorations, and no change in EDE global scores between therapy conditions. There was no statistically significant difference in percentages of participants considered to no longer meet the diagnostic threshold for an ED.Footnote Footnote 1

Table 3. Reliable change and diagnostic status by treatment condition

Note. PTSD, posttraumatic stress disorder; ED, eating disorder; CAPS-5, Clinician-Administered PTSD Scale-5; EDE, Eating Disorder Examination 17.0. Loss of PTSD diagnosis was defined as no longer meeting Diagnostic and Statistical Manual of Mental Disorders (Fifth Edition) symptom criteria as assessed with the CAPS-5; a reliable change represented a change from baseline of 11.64 points. Loss of ED diagnosis was defined as no more than 2 objective binge and/or purge episodes in the past month and a body mass index (BMI; weight in kilograms/height in meters2) of greater than 18.5; a reliable change represented a change in EDE global score of 0.91 points.

Treatment satisfaction and preferences

Both treatments were rated highly with 96.8% at least somewhat satisfied, and 90.4% very satisfied to completely satisfied. There were no differences in treatment satisfaction between conditions, Mann–Whitney U = 0.40. All participants in both treatments who provided data said they would recommend the treatment they received to someone else. There was a strong preference for integrated treatment for both ED and PTSD (87.1%) over separate treatments for ED or PTSD. Finally, 93.3% of those who received integrated CBT thought having received PTSD treatment would facilitate their ED recovery.

Discussion

This study represents the first controlled evaluation of integrated psychotherapy for ED-PTSD, which has been identified as a research priority in EDs (Obeid, McVey, Seale, Preskow, & Norris, Reference Obeid, McVey, Seale, Preskow and Norris2020; Rijkers et al., Reference Rijkers, Schoorl, van Hoeken and Hoek2019; Van Furth, van der Meer, & Cowan, Reference Van Furth, van der Meer and Cowan2016). We found strong evidence for the efficacy of integrated CBT for ED-PTSD in improving PTSD, in that the treatment led to large magnitude improvements in PTSD, and a higher rate of PTSD remission, relative to standard CBT for ED at EoT, with improvements maintained over 6-month follow-up. The lack of further improvement in ED psychopathology after intensive treatment that was found across both study conditions was likely related to the large magnitude improvements participants had already made during the intensive ED treatment that preceded the study treatments [pre-intensive treatment M = 4.40, s.d. = 1.12 and end-of-intensive treatment M = 2.85, s.d. = 1.09, t (41) = 9.95, p < 0.001, g = 1.01], and the likelihood of worsening ED symptomatology following intensive ED treatment even with additional treatment (Carter et al., Reference Carter, McFarlane, Bewell, Olmsted, Woodside, Kaplan and Crosby2008; MacDonald, McFarlane, Trottier, Mahan, & Olmsted, Reference MacDonald, McFarlane, Trottier, Mahan and Olmsted2021). Ultimately, our hope is that integrated CBT for ED-PTSD will result in better ED outcomes than standard CBT for ED. However, this initial RCT was not powered to detect such differences, which would presumably be of a smaller magnitude in light of participants having already participated in intensive ED treatment. Indeed, the between-group effect size for ED psychopathology indicated a small to moderate magnitude benefit of integrated CBT over standard CBT for ED. Importantly, and in contrast to the expectations of some clinicians (Trottier et al., Reference Trottier, Monson, Wonderlich, MacDonald and Olmsted2017), mixed models analyses and analyses of clinical status at EoT and follow-ups did not suggest worsening of ED or concomitant mental health symptoms in the integrated CBT for ED-PTSD condition relative to the standard CBT for ED condition. This is a vital first step toward demonstrating the safety and efficacy of integrating trauma-focused and ED psychotherapies. Results from this randomized efficacy-effectiveness study indicate that individuals with ED-PTSD can not only tolerate, but also greatly benefit from, manualized trauma-focused psychotherapy. This new information is critical to overcoming barriers to accessing evidence-based PTSD treatments related to clinician perceptions that trauma processing will lead to an increase in ED and other mental health symptoms, including PTSD symptoms (Trottier et al., Reference Trottier, Monson, Wonderlich, MacDonald and Olmsted2017).

Although integrated CBT resulted in greater PTSD symptom reduction, higher PTSD remission rates, and larger magnitude improvements in anxiety, CBT for ED also led to improvements in PTSD, anxiety, and depression, as well as similar ED outcomes, and high treatment satisfaction. These findings raise questions about to whom integrated CBT should be provided. Given that integrated CBT for ED-PTSD led to superior PTSD outcomes, and that integrated treatment was highly preferred by participants and perceived by participants to facilitate their ED recovery, we suggest that integrated psychotherapy for ED-PTSD should be offered whenever possible. Further, given the high prevalence of co-occurring PTSD in those presenting for intensive ED treatment, the likelihood that PTSD and ED maintain one another when they co-occur, and high rates of relapse in EDs (Berends et al., Reference Berends, Boonstra and van Elburg2018; Yu et al., Reference Yu, Agras and Bryson2013), we hope that these findings will spur a shift in practice such that evidence-based psychotherapy for PTSD is made available in ED treatment programs.

At the same time, we would be remiss to not highlight the good outcomes experienced in this study with standard CBT for ED. Psychiatric comorbidity is one of several reasons that clinicians tend to stray from providing manualized psychotherapy for EDs (Waller & Turner, Reference Waller and Turner2016), and our findings indicate that in the case of ED-PTSD, manualized CBT for ED is beneficial. That said, the improvements in PTSD with standard CBT for ED may have been, in part, due to the assessment and treatment process being more trauma-informed than is typical of such treatments in research studies and clinical practice. The recruitment process, multiple assessments of PTSD symptoms, and the fact that the study therapists were experienced working with individuals who have PTSD, may have contributed to improvements in PTSD.

Strengths and limitations

There are several important strengths of this initial trial, including that it was conducted in a clinical practice setting with program therapists, and that it utilized an active treatment comparison group, which was consistent with routine care. Several limitations should also be considered in interpreting and generalizing the results of this trial. Although adequately powered for our primary hypothesis, the study's ED diagnostic heterogeneity and relatively small sample size do not allow us to draw conclusions about the impact of integrated CBT for ED-PTSD on ED outcomes relative to standard CBT for ED. The sample size also limits examination of some factors that might moderate treatment outcome, such as type of trauma or duration of illness with PTSD or ED, as well as our ability to examine potential mediators of treatment response. Larger trials that can examine these questions, as well as whether similar results can be achieved at other sites are needed. Another potential limitation is that participants had to have completed a course of intensive ED treatment focused on nutritional stabilization and ED symptom interruption in order to be eligible to participate. Individuals who did not complete the course of intensive treatment and/or did not reach a BMI of 18.5 were selected out; thus, our sample may have been made up of individuals who are more likely to complete treatment. It is possible that non-completion rates may have been higher, and outcomes poorer, if participants had not already completed this course of treatment. That said, these inclusion criteria are consistent with recommendations in the literature regarding when to initiate trauma-focused treatment with individuals with EDs (e.g. Brewerton, Reference Brewerton and Brewerton2004). The majority of participants were also taking psychotropic medication. It is possible that these medications may interact with the psychotherapies that were provided and outcomes may not generalize to those who are not on psychotropic medication.

Another important next step in developing and testing integrated psychotherapies for ED-PTSD is to adapt the protocol to be a stand-alone individual therapy (i.e. not requiring a course of intensive ED treatment beforehand). There are also many other important empirical questions to be answered such as at what minimum BMI trauma-focused psychotherapy can be effectively initiated in individuals with ED, whether engaging in ED behaviors such as binge eating and purging negatively affects outcomes of integrated treatment, and whether integration of other evidence-based CBTs for PTSD (e.g. prolonged exposure therapy; Foa, Hembree, Rothbaum, and Rauch, Reference Foa, Hembree, Rothbaum and Rauch2019) and EDs (e.g. integrative cognitive-affective therapy; Wonderlich et al., Reference Wonderlich, Peterson, Smith, Klein, Mitchell and Crow2015) is also efficacious. We hope that these first RCT findings will catalyze change in the treatment of individuals with ED-PTSD and encourage further research on integrated psychotherapies with the goal of enabling more fulsome mental health recoveries for these individuals.

Acknowledgements

We wish to express our gratitude to Dr Rachel Liebman for her work leading the fidelity ratings, and to Isabel Shapiro for her assistance with manuscript preparation.

Financial support

This research was supported by the Ontario Mental Health Foundation and a University of Toronto Academic Scholar Award awarded to the first author.

Conflict of interest

Dr Crosby is a paid Statistical Consultant for Health Outcomes Solutions, Winter Park, Florida, USA. Dr Monson receives royalties for the manual: Cognitive Processing Therapy for PTSD: A Comprehensive Manual, by Resick PA, Monson CM, Chard KM. Guilford Press, ©2017. Dr Wonderlich receives royalties for the book: Integrative Cognitive Affective Therapy for Bulimia Nervosa, by Wonderlich SA, Peterson C, Smith T, Klein M, Mitchell JE, Crow S. Guilford Press, ©2015.

Ethical standards

The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional committees on human research.

Footnotes

The online version of this article has been updated since original publication. A notice detailing the changes has also been published at https://doi.org/10.1017/S0033291721005365

The notes appear after the main text.

1 Given the lack of a consensus definition, the impact of more and less stringent thresholds for loss of ED diagnosis were examined. As expected, the number of participants meeting the threshold increased with less stringent and decreased with more stringent definitions. The proportion of individuals meeting the threshold was similar in each treatment condition regardless of the threshold used.

References

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). Washington, DC: Author.Google Scholar
American Psychological Association. (2017). PTSD Treatments. https://www.apa.org/ptsd-guideline/treatments.Google Scholar
Berends, T., Boonstra, N., & van Elburg, A. (2018). Relapse in anorexia nervosa: A systematic review and meta-analysis. Current Opinion in Psychiatry, 31(6), 445455. doi: 10.1097/YCO.0000000000000453CrossRefGoogle ScholarPubMed
Blevins, C. A., Weathers, F. W., Davis, M. T., Witte, T. K., & Domino, J. L. (2015). The posttraumatic stress disorder checklist for DSM-5 (PCL-5): Development and initial psychometric evaluation. Journal of Traumatic Stress, 28(6), 489498. doi: 10.1002/jts.22059CrossRefGoogle ScholarPubMed
Brewerton, T. D. (2004). Eating disorders, victimization, and comorbidity: Principles of treatment. In Brewerton, T. D. (Ed.), Clinical handbook of eating disorders: An integrated approach (pp. 509545). New York: Marcel Dekker.CrossRefGoogle Scholar
Brewerton, T. D. (2007). Eating disorders, trauma, and comorbidity: Focus on PTSD. Eating Disorders: The Journal of Treatment and Prevention, 15, 285304. doi: 10.1080/10640260701454311CrossRefGoogle ScholarPubMed
Brewerton, T. D., Perlman, M. M., Gavidia, I., Suro, G., Genet, J., & Bunnell, D. W. (2020). The association of traumatic events and posttraumatic stress disorder with greater eating disorder and comorbid symptom severity in residential eating disorder treatment centers. International Journal of Eating Disorders, 53(12), 20612066. doi: 10.1002/eat.23401CrossRefGoogle ScholarPubMed
Carter, J. C., McFarlane, T. L., Bewell, C., Olmsted, M. P., Woodside, D. B., Kaplan, A. S., & Crosby, R. D. (2008). Maintenance treatment for anorexia nervosa: A comparison of cognitive behavior therapy and treatment as usual. International Journal of Eating Disorders, 42(3), 202207. doi: 10.1002/eat.20591CrossRefGoogle Scholar
Dansky, B. S., Brewerton, T. D., Kilpatrick, D. G., & O'Neil, P. M. (1997). The national women's study: Relationship of victimization and posttraumatic stress disorder to bulimia nervosa. International Journal of Eating Disorders, 21, 213228. doi: 10.1002/(sici)1098-108x(199704)21:3 < 213::aid-eat2>3.0.co;2-n3.0.CO;2-N>CrossRefGoogle ScholarPubMed
Fairburn, C. G., Cooper, Z., Doll, H. A., O'Connor, M., Bohn, K., & Hawker, D. M. (2009). Transdiagnostic cognitive-behavioral therapy for patients with eating disorders: A two-site trial with 60-week follow-up. American Journal of Psychiatry, 166, 311319. doi: 10.1176/appi.ajp.2008.08040608CrossRefGoogle ScholarPubMed
Fairburn, C. G., Cooper, Z., & O'Connor, M. (2014). The Eating Disorder Examination (EDE 17). Retrieved from http://www.credo-oxford.com/pdfs/EDE_17.0D.pdf.Google Scholar
Fairburn, C. G., Cooper, Z., Shafran, R., Bohn, K., Hawker, D., Murphy, R. & Straebler, S. (2008). Enhanced cognitive behavior therapy for eating disorders: The core protocol. In Fairburn, C. G. (ed.), Cognitive behavior therapy and eating disorders (pp. 47183). New York: Guilford Press.Google ScholarPubMed
Foa, E., Hembree, E. A., Rothbaum, B. O., & Rauch, S. (2019). Prolonged exposure therapy for PTSD: Emotional processing of traumatic experiences – therapist guide, 2nd ed.). Oxford: Oxford University Press. doi: 10.1093/med-psych/9780190926939.001.0001.CrossRefGoogle Scholar
Hazzard, V., Crosby, R., Crow, S., Engel, S., Schaefer, L., Brewerton, T., … Wonderlich, S. (2021). Treatment outcomes of psychotherapy for binge-eating disorder in a randomized controlled trial: Examining the roles of childhood abuse and post-traumatic stress disorder. European Eating Disorders Review, 29(4), 611621. doi: 10.1002/erv.2823CrossRefGoogle Scholar
Holzer, S., Uppala, S., Wonderlich, S., Crosby, R., & Simonich, H. (2008). Mediational significance of PTSD in the relationship of sexual trauma and eating disorders. Child Abuse & Neglect, 32, 561566. doi: 10.1016/j.chiabu.2007.07.011CrossRefGoogle ScholarPubMed
Hudson, J. I., Hiripi, E., Pope, H. G., & Kessler, R. C. (2007). The prevalence and correlates of eating disorders in the national comorbidity survey replication. Biological Psychiatry, 61(3), 348358. doi: 10.1016/j.biopsych.2006.03.040CrossRefGoogle ScholarPubMed
Jacobson, N. S., & Truax, P. (1991). Clinical significance: A statistical approach to defining meaningful change in psychotherapy research. Journal of Consulting and Clinical Psychology, 59(1), 1219. doi: 10.1037//0022-006x.59.1.12CrossRefGoogle ScholarPubMed
Karr, T. M., Crosby, R. D., Cao, L., Engel, S. G., Mitchell, J. E., Simonich, H., & Wonderlich, S. (2013). Posttraumatic stress disorder as a moderator of the association between negative affect and bulimic symptoms: An ecological momentary assessment study. Comprehensive Psychiatry, 54(1), 6169. doi: 10.1016/j.comppsych.2012.05.011CrossRefGoogle Scholar
Kessler, R. C., Sonnega, A., Bromet, E., Hughes, M., & Nelson, C. B. (1995). Posttraumatic stress disorder in the national comorbidity survey. Archives of General Psychiatry, 52(12), 10481060. doi: 10.1001/archpsyc.1995.03950240066012CrossRefGoogle ScholarPubMed
Liebman, R. E., Becker, K. R., Smith, K. E., Cao, L., Keshishian, A. C., Crosby, R. D., … Thomas, J. J. (2020). Network analysis of posttraumatic stress and eating disorder symptoms in a community sample of adults exposed to childhood abuse. Journal of Traumatic Stress, 34(3), 665674. doi: 10.1002/jts.22644CrossRefGoogle Scholar
Lovibond, P. F., & Lovibond, S. H. (1995). The structure of negative emotional states: Comparison of the Depression Anxiety Stress Scales (DASS) with the beck depression and anxiety inventories. Behaviour Research and Therapy, 33(3), 335343. doi: 10.1016/0005-7967(94)00075-uCrossRefGoogle ScholarPubMed
MacDonald, D. E., McFarlane, T., Trottier, K., Mahan, M., & Olmsted, M. P. (2021). Maintenance treatment for eating disorders following inpatient or day treatment: Outcomes of intensive outpatient group and individual CBT treatments. Eating Disorders: The Journal of Treatment and Prevention. doi: 10.1080/10640266.2021.1969787CrossRefGoogle ScholarPubMed
Mitchell, K. S., Singh, S., Hardin, S., & Thompson-Brenner, H. (2021). The impact of comorbid posttraumatic stress disorder on eating disorder treatment outcomes: Investigating the unified treatment model. International Journal of Eating Disorders, 54(7), 12601269. doi: 10.1002/eat.23515CrossRefGoogle ScholarPubMed
Obeid, N., McVey, G., Seale, E., Preskow, W., & Norris, M. L. (2020). Cocreating research priorities for anorexia nervosa: The Canadian eating disorder priority setting partnership. International Journal of Eating Disorders, 53(5), 662672. doi: 10.1002/eat.23234CrossRefGoogle ScholarPubMed
Resick, P. A., Monson, C. M., & Chard, K. M. (2016). Cognitive processing therapy for PTSD: A comprehensive manual. New York, New York: Guilford Publications.Google Scholar
Rijkers, C., Schoorl, M., van Hoeken, D., & Hoek, H. W. (2019). Eating disorders and posttraumatic stress disorder. Current Opinion in Psychiatry, 32(6), 510517. doi: 10.1097/YCO.0000000000000545CrossRefGoogle ScholarPubMed
Scharff, A., Oritz, S. N., Forrest, L. N., & Smith, A. R. (2019). Comparing the clinical presentation of eating disorder patients with and without trauma history and/or comorbid PTSD. Eating Disorders: The Journal of Treatment and Prevention, 29(1), 88102. doi: 10.1080/10640266.2019.1642035CrossRefGoogle ScholarPubMed
Scharff, A., Oritz, S. N., Forrest, L. N., Smith, A. R., & Boswell, J. F. (2021). Post-traumatic stress disorder as a moderator of transdiagnostic, residential eating disorder treatment outcome trajectory. Journal of Clinical Psychology, 77(4), 9861003. doi: 10.1002/jclp.23106CrossRefGoogle ScholarPubMed
Serra, R., Kiekens, G., Tarsitani, L., Vrieze, E., Bruffaerts, R., Loriedo, C., … Vanderlinden, J. (2020). The effect of trauma and dissociation on the outcome of cognitive behavioural therapy for binge eating disorder: A 6-month prospective study. European Eating Disorders Review, 28(3), 309317. doi: 10.1002/erv.2722CrossRefGoogle ScholarPubMed
Sheehan, D. V. (2015). MINI International neuropsychiatric interview 7.0. Jacksonville, FL: Medical Outcomes Systems Inc.Google Scholar
Tomba, E., Tecuta, L., Crocetti, E., Squarcio, F., & Tomei, G. (2019). Residual eating disorder symptoms and clinical features in remitted and recovered eating disorder patients: A systematic review with meta-analysis. International Journal of Eating Disorders, 52(7), 759776. doi: 10.1002/eat.23095CrossRefGoogle ScholarPubMed
Trottier, K. (2020). Posttraumatic stress disorder predicts non-completion of day hospital treatment for bulimia nervosa and other specified feeding/eating disorder. European Eating Disorders Review, 28(3), 343350. doi: 10.1002/erv.2723CrossRefGoogle ScholarPubMed
Trottier, K., & Monson, C. (2021). Integrating cognitive processing therapy for posttraumatic stress disorder with cognitive behavioral therapy for eating disorders in project recover. Eating Disorders: The Journal of Treatment and Prevention, 29(3), 119. doi: 10.1080/10640266.2021.1891372.CrossRefGoogle ScholarPubMed
Trottier, K., Monson, C., Wonderlich, S., MacDonald, D. E., & Olmsted, M. P. (2017). Frontline clinicians’ perspectives on and utilization of trauma-focused therapy with individuals with eating disorders. Eating Disorders: Journal of Treatment and Prevention, 25(1), 2236. doi: 10.1080/10640266.2016.1207456CrossRefGoogle ScholarPubMed
Van Furth, E. F., van der Meer, A., & Cowan, K. (2016). Top 10 research priorities for eating disorders. The Lancet. Psychiatry, 3(8), 706707. doi: 10.1016/S2215-0366(16)30147-XCrossRefGoogle ScholarPubMed
Vanzhula, I. A., Calebs, B., Fewell, L., & Levinson, C. A. (2019). Illness pathways between eating disorder and post-traumatic stress disorder symptoms: Understanding comorbidity with network analysis. European Eating Disorders Review: The Journal of the Eating Disorders Association, 27(2), 147160. doi: 10.1002/erv.2634CrossRefGoogle ScholarPubMed
Waller, G., & Turner, H. (2016). Therapist drift redux: Why well-meaning clinicians fail to deliver evidence-based therapy, and how to get back on track. Behaviour Research and Therapy, 77, 129137. doi: 10.1016/j.brat.2015.12.005CrossRefGoogle ScholarPubMed
Watts, B. V., Schnurr, P. P., Mayo, L., Young-Xu, Y., Weeks, W. B., & Friedman, M. J. (2013). Meta-analysis of the efficacy of treatments for posttraumatic stress disorder. The Journal of Clinical Psychiatry, 74(6), e541e550. doi: 10.4088/jcp.12r08225CrossRefGoogle ScholarPubMed
Weathers, F. W., Blake, D. D., Schnurr, P. P., Kaloupek, D. G., Marx, B. P., & Keane, T. M. (2013). The Clinician Administered PTSD Scale for DSM-5 (CAPS-5).Google Scholar
Wiltsey Stirman, S., Monson, C. M., & Resick, P. A. (2017). Cognitive Processing Therapy: Therapist Adherence and Competence Protocol. Unpublished rating form available from the authors.Google Scholar
Wonderlich, S. A., Peterson, C. B., Smith, T. L., & Klein, M. H., Mitchell, J. E., & Crow, S. J. (Collaborators). (2015). Integrative cognitive-affective therapy for bulimia nervosa: A treatment manual. New York: The Guilford Press.Google Scholar
Yu, J., Agras, W. S., & Bryson, S. (2013). Defining recovery in adult bulimia nervosa. Eating Disorders, 21, 379394. doi: 10.1080/10640266.2013.827536CrossRefGoogle ScholarPubMed
Figure 0

Fig. 1. Consort flow diagram.Note. ED, eating disorder; PTSD, posttraumatic stress disorder; BMI, body mass index; CBT, cognitive-behavioral therapy. *Participant disclosed after randomization that they misrepresented their symptoms in the eligibility assessment and was subsequently deemed ineligible and excluded from analyses.

Figure 1

Table 1. Participant characteristics

Figure 2

Table 2. Continuous outcomes at all-time points

Figure 3

Table 3. Reliable change and diagnostic status by treatment condition