Abstract
Background:
Cognitive-behavioral therapy for avoidant/restrictive food intake disorder (ARFID; CBT-AR) theoretically targets three prototypic motivations (sensory sensitivity, lack of interest/low appetite, fear of aversive consequences), aligned with three modularized interventions. As an exploratory investigation, we: (1) evaluated change in candidate mechanisms in relationship to change in ARFID severity, and (2) tested if assignment (vs. not) to a module resulted in larger improvements in the corresponding mechanism.
Method:
Males and females (N=42; 10–55 years) participated in an open trial of CBT-AR.
Results:
Decreases in candidate mechanisms had medium to large correlations with decreases in ARFID severity—sensory sensitivity: −0.7 decrease (r=0.42, p=.01); lack of interest/low appetite: −0.3 decrease (r=0.60, p<.0001); and fear of aversive consequences: −1.1 decrease (r=0.33, p=.05). Linear mixed models revealed significant weekly improvements for each candidate mechanism across the full sample (ps<.0001) with significant interactions for the sensory and fear of aversive consequences modules.
Discussion:
Sensory sensitivity and fear of aversive consequences improved more if the CBT-AR module was received, but lack of interest/low appetite may improve regardless of receipt of the corresponding module. Future research is needed to test target engagement in CBT-AR with adaptive treatment designs, and to identify valid and sensitive measures of candidate mechanisms.
Keywords: Avoidant/restrictive food intake disorder, feeding and eating disorders, picky eating, cognitive-behavioral therapy, psychotherapy trial, mechanisms, experimental therapeutics
INTRODUCTION
Outpatient behavioral therapies for avoidant/restrictive food intake disorder (ARFID) have been shown to be feasible and acceptable, with promising clinical outcomes (Burton Murray et al., 2023; Dumont et al., 2019; Lock et al., 2019; Shimshoni et al., 2020; Thomas et al., 2021; Thomas et al., 2020). Identification of mechanisms through which behavioral therapies work is key to inform treatment parsimony (Guastaferro & Collins, 2019). Specific treatment components may lead to changes in certain characteristics (i.e., mechanisms of action) that then produce symptom improvement (Kazdin, 2007; Kraemer et al., 2002), but there is little research on which components and which target mechanisms of action are involved in behavioral treatments for ARFID.
Cognitive-behavioral therapy for ARFID (CBT-AR) uses an exposure-based approach to target the three prototypical motivations for ARFID currently described as candidate mechanisms (i.e., sensory sensitivity, lack of interest/low appetite, fear of aversive consequences; APA 2021), which in turn decrease food avoidance. Proof-of-concept studies of CBT-AR showed large pre- to post-treatment decreases in overall ARFID severity and decreases in candidate mechanisms (Thomas et al., 2021; Thomas et al., 2020). However, evaluation of ARFID candidate mechanisms throughout the course of treatment (e.g., session-by-session) has not yet been studied, which is important for empirical testing of candidate mechanisms of action (Kazdin, 2007). Further, it is unknown if receipt of a module hypothesized to target a candidate mechanism is associated with changes in the mechanism that then relate to ARFID severity improvement.
Among children, adolescents, and adults who were offered 20–30 sessions of CBT-AR in an open trial, we evaluated changes in measures of candidate mechanisms collected at each treatment session in relation to pre- and post-treatment ARFID severity. We aligned this exploratory study with the refinement phase (Phase 1B) of the Obesity-Related Behavioral Intervention Trials (ORBIT) model for behavioral treatment development and testing, which encourages evaluating the necessity of specific components to enhance parsimony (Czajkowski et al., 2015). We: (1) evaluated change in candidate mechanisms in relationship to change in ARFID severity, and (2) tested if participants who were exposed to a CBT-AR module had larger improvements in the corresponding candidate mechanism than participants not exposed to the module.
METHOD
We used data collected as part of an open trial of CBT-AR, on which pre- and post-treatment outcomes have already been published on a subset of participants from both child/adolescent (10–17 years old; Thomas et al. 2020) and adult (age ≥18; Thomas et al. 2021) samples. See Supplemental Content for more detail, including Supplemental Figure 1 for patient flow.
Treatment
CBT-AR (Thomas & Eddy, 2019) is a flexible treatment modularized to address each of the three prototypic ARFID motivations, and modules were selected and implemented based on the participant’s ARFID diagnosis and the treating clinician’s judgment. Participants were offered 20–30 sessions, allowing low-weight participants an additional 10 sessions to reach individualized (via growth charts) medically determined target weight ranges. Participants who were under the age 16 years and/or were underweight received the family-supported treatment format while participants who were over the age of 16 years and were not underweight received the individual treatment format. Per the CBT-AR manual (Thomas & Eddy, 2019), each session lasted 50 minutes. Treatment was provided by a clinical psychologist or clinical psychology trainee (intern, post-doctoral fellow) at the MGH EDCRP.
In the first stage of CBT-AR, patients receive psychoeducation about ARFID and start to observe their eating patterns and early changes such as gaining weight for those who are underweight. In addition, patients work to create and maintain a regular eating schedule as well as increase the volume and/or variety of their food depending on their ARFID prototypic motivation. The second stage involves learning about the dangers of nutritional deficiencies and treatment planning. During treatment planning, patients choose foods that they will use for exposure in the third stage of treatment. The goal is to widen diet across the five food groups, gain weight and/or reduce nutritional deficiencies, and improve psychosocial impairment.
Stage three addresses maintaining mechanisms through exposure to avoided foods and eating-related situations. For individuals with multiple ARFID motivations, the CBT-AR manual recommends that the clinician order the implementation of modules by starting with the module corresponding to the patient’s most impairing motivation. Exposure is completed during CBT-AR sessions and at home. To address sensory sensitivity, the corresponding module involves tasting new (usually at least five) foods each session using the 5 Steps approach to objectively describe food characteristics (by sight, touch, smell, taste, texture). To address a lack of interest/low appetite, the corresponding module involves interoceptive exposures to fullness (e.g., water chugging), increasing attention to the rewarding properties of favorite foods, and self-monitoring of thoughts/feelings and physical sensations including hunger/fullness. To address feared outcomes from eating/food, the corresponding module involves creating a fear/avoidance hierarchy and systematic exposure to feared food and eating-related situations to test negative predictions and promote new learning. The fourth and final stage is creating a plan for relapse prevention.
Session Measures
We collected the self-report measures described below at each CBT-AR session. The clinical trial began in 2016, prior to the availability of validated measures for ARFID prototypical motivations. Thus, measures used were considered proxy measures since they were selected because they were the best approximation corresponding to ARFID motivations available at the time.
Sensory Sensitivity.
The Adult Eating Behavior Questionnaire (AEBQ) is a 34-item self-report questionnaire of avoidance and approach behaviors towards eating and food (Hunot et al., 2016). We used the five-item AEBQ-food fussiness (AEBQ-FF) subscale as a measure of sensitivity to the sensory characteristics of novel foods and food variety (e.g., “I refuse new foods at first”). Items are rated on a 5-point Likert scale from 1 (“strongly disagree”) to 5 (“strongly agree”) and averaged. Higher AEBQ-FF scores indicate greater sensory sensitivity severity.
Lack of Interest in Eating/Low Appetite.
We used the four-item AEBQ Satiety Responsiveness (AEBQ-SR) subscale as a measure of low appetite (e.g., “I get full easily”), which has been shown to moderately correlate with an interview measure of lack of interest in eating/food in children, adolescents, and young adults with ARFID (Cooper-Vince et al., 2022). Items are rated on a 5-point Likert scale from 1 (“strongly disagree”) to 5 (“strongly agree”) and averaged. Higher AEBQ-SR subscale scores indicate greater early satiation and food avoidance due to fullness.
Fear of Aversive Consequences.
We used a sum of three items adapted from the Anxiety Sensitivity Index (ASI-adapted) to assess fear of aversive consequences around eating. The original ASI has 16 items assessing sensitivity and distress around anxiety-related symptoms, including physiologic symptoms (Taylor et al., 2007). Items are rated on a 5-point Likert scale from 0 (“very little”) to 4 (“very much”). In the current study, we used three items assessing physiologic symptoms with the structure of “It scares me when…” We used the following three items: “It scares me when I am nauseous;” “It scares me when I think I might choke;” and “It scares me when I feel full.”
Pre- and Post-Treatment Measures
Pica, ARFID, and Rumination Disorder Interview (PARDI).
At pre- and post-treatment assessment visits, the PARDI was administered by an independent, trained evaluator who did not serve as a study therapist. The PARDI is a semi-structured interview that evaluates ARFID psychopathology (Bryant-Waugh et al., 2019; Cooper-Vince et al., 2022). The PARDI severity scale assesses severity of psychosocial impairment, health consequences, and nutritional compromise due to avoidant/restrictive eating. For Aim 1, we evaluated the association between pre- to post-treatment changes in the PARDI ARFID severity score which ranges from 0 (indicating a lack of symptoms) to 6 (extreme symptom severity) with changes in the AEBQ-FF, AEBQ-SR, and ASI-adapted.
Data Analysis
We performed all data analyses in R (Team, 2014). A participant was considered a dropout if they dropped out of treatment prior to completing all four CBT-AR stages as mutually agreed upon with the therapist. For Aim 1 (mechanisms of change), for each candidate mechanism (AEBQ-FF, AEBQ-SR, ASI-Adapted), we evaluated changes between the first and last sessions. We calculated the average, standard deviation (SD), and range of candidate mechanism scores at the start and end of CBT-AR sessions and of ARFID severity (PARDI severity score) at pre- and post-treatment. We calculated Pearson or Spearman correlations between each candidate mechanism change score and ARFID severity change score; for participants who dropped out or did not have scores, we carried forward their pre-treatment ARFID severity scores.
For Aim 2 (target engagement), we conducted a multilevel model for weekly changes in each candidate mechanism with an interaction term for the aligning module received (i.e., AEBQ-FF with sensory sensitivity yes/no; AEBQ-SR with lack of interest/low appetite yes/no; ASI-Adapted with fear of aversive consequences yes/no). We used restricted maximum likelihood estimating, which allowed us to utilize all available data (including participants who dropped out) and yields conservative standard error estimates. Each session was planned to occur weekly, but were not scheduled exactly seven days apart; thus, we used number of weeks since Session 1 for our time variable. Each model included the candidate mechanism as the dependent variable and time (Weeks), module received (yes/no), and non-time-varying covariates (age at start of treatment, sex) as independent variables. We controlled for age at start of treatment and sex in each model. All models included random intercepts. We added random slopes if log-likelihood ratio tests indicated model improvement. CBT-AR is offered for 20–30 sessions, but one participant received 40 sessions due to clinical need; for the Aim 2 analysis, we removed sessions beyond 30. We also conducted a sensitivity analysis of outliers by participant using Cook’s distance plots and re-ran each model with outliers removed (see Supplemental Content).
RESULTS
Data Completion
Of 42 participants included, five (12%) dropped out; among those who dropped out the range of sessions completed was 1 to 12. For participants who completed treatment (i.e., completed all four CBT-AR stages), number of sessions completed was M(SD)=22.3 (5.6), range=13–40. Module completion was as follows: n=27 participants for sensory sensitivity [M(SD)=13.6 (4.9), range=6–24 sessions], n=13 lack of interest/low appetite [M(SD)=4.5 (3.6), range=1–13 sessions], n=10 fear of aversive consequences [M(SD)=11.4 (6.1), range=4–19 sessions]. Ten participants completed two modules (n=1 sensory sensitivity + fear of aversive consequences; n=1 lack of interest/low appetite + fear of aversive consequences; n=8 sensory sensitivity + lack of interest/low appetite). One participant completed three modules.
Aim 1: Mechanisms of Change
Among treatment completers, we evaluated differences in each candidate mechanism from the beginning to the end of CBT-AR, and if differences in each mechanism related to change in ARFID severity (Table 1; Supplemental Figure 2).
Table 1.
Clinical characteristics of participants who received CBT-AR (N=42).
| M (SD; range) or n (%) | ||
|---|---|---|
|
Age (years) Child/adolescent (age <18 years) Adult (age > 18 years) |
17.4 (8.4; 10–55) 27 (64%) 15 (36%) |
|
| Sex- Female | 23 (55%) | |
|
Race Asian More than 1 race white |
1 (2%) 2 (5%) 39 (93%) |
|
|
Ethnicity Hispanic/Latino(a) Non- Hispanic/Latino(a) |
3 (7%) 39 (93%) |
|
|
BMI Percentile |
25.8 (6.7; 16–37) 18.8 (24.6; 0–96) |
|
|
BMI categories for ≥ 20 years old (kg/m2) Underweight (<18.5) Normal weight (18.5–24.9) Overweight (25.0–29.9) Obesity (≥30.0) |
3 (21%) 4 (29%) 4 (29%) 3(21%) |
|
|
BMI categories for < 20 years old (percentiles) Underweight (<5th) Normal weight (5th to <85th) Overweight (85th to <95th) Obesity (≥95th) |
10 (36%) 17 (61%) 0 (0%) 1 (4%) |
|
|
Treatment Format Individual Family-supported |
14 (33%) 28 (67%)2 |
|
|
Module received1 Sensory Sensitivity Lack of Interest Fear of Aversive Consequences |
26 (62%) 13 (31%) 10 (24%) |
|
| Outcome | Pre-Treatment 3 | Post-Treatment 4 |
| ARFID Severity (PARDI; range=0–6) | 2.4 (0.7; 1.0–3.8) | 1.5 (1.0; 0.0–3.6) |
| Candidate Mechanisms | Session 1 | Last Session |
| Sensory Sensitivity (AEBQ-FF; range=1–5) | 4.2 (0.7; 2.4–5.0) | 3.3 (0.9; 1.6–5.0) |
| Lack of Interest (AEBQ-SR; range=1–5) | 3.1 (0.8; 1.0–4.3) | 2.8 (0.9; 1.0–5.0) |
| Fear of Aversive Consequences (ASI-adapted5; range=0–12) | 3.0 (2.9; 0.0–10.0) | 2.1 (2.8; 0.0–10.0) |
Note. CBT-AR=cognitive-behavioral therapy for avoidant/restrictive food intake disorder; M=mean; SD-standard deviation; PARDI=Pica, ARFID, Rumination Disorder Interview; AEBQ-FF=Adult Eating Behavior Questionnaire-food fussiness; AEBQ-SR=AEBQ-satiety responsiveness; ASI=Anxiety Sensitivity Index. Demographics were self-reported.
CBT-AR is modularized to address each of the three prototypic ARFID motivations, and modules were selected and implemented based on the participants’ ARFID diagnosis and the treating clinician’s judgment.
One adult participant received the family-supported format.
n=1 participant did not have pre-treatment PARDI data so was not included in analysis.
n=6 participants did not have post-treatment PARDI data—we carried forward their last ARFID severity score (from pre-treatment).
We used a sum of three items adapted from the Anxiety Sensitivity Index (ASI-adapted) to assess fear of aversive consequences around eating.
Sensory Sensitivity.
Between first and last sessions, there was a 0.7 average decrease on the AEBQ-FF (lower food fussiness) (−2.9 to +0.7 range), which had a medium and significant correlation with change in ARFID severity (r=0.42, p=.01).
Lack of Interest in Eating/Low Appetite.
Between first and last sessions, there was a 0.3 average decrease on the AEBQ-SR (lower sensitivity to fullness) (−2.5 to +2.1 range), which had a large, significant correlation with change in ARFID severity (r=0.60, p<.0001).
Fear of Aversive Consequences.
Between first and last sessions, there was a 1.1 average decrease on the ASI-adapted (lower fear of eating) (−6.0 to +4.0 range), which had a medium albeit non-significant correlation with change in ARFID severity (r=0.33, p=.05).
Aim 2: Target Engagement
We conducted a series of multilevel models to test if there were session-by-session reductions in each candidate mechanism and if this changed with CBT-AR module received (Figure 1; Supplemental Tables 1 and 2). Missing datapoints for session-level measures was 0.9%.
Figure 1.
Change in each candidate mechanism across treatment by module received with confidence intervals.
Note. ARFID=avoidant/restrictive food intake disorder; AEBQ-FF=Adult Eating Behavior Questionnaire-food fussiness; AEBQ-SR=AEBQ-satiety responsiveness; ASI=Anxiety Sensitivity Index.
Sensory Sensitivity.
There was a significant decrease in the AEBQ-FF (candidate mechanism) over time as a function of module received (sensory sensitivity yes/no) [estimate=−0.01, p=.007]—that is, participants who had the sensory sensitivity motivation (and received the corresponding module) decreased 0.01 points more per week than participants who did not have sensory sensitivity. However, with outliers removed, there was no longer a significant effect for module received [estimate=−0.01, p=.064].
Lack of Interest in Eating/Low Appetite.
There was a significant decrease in the AEBQ-SR (candidate mechanism) over time [estimate=−0.01, p<.0001]—that is, participants on average decreased 0.01 points per week, regardless of module received. There was no significant interaction with module received.
Fear of Aversive Consequences.
There was a significant decrease in the ASI-Adapted (candidate mechanism) over time as a function of module received (fear of aversive consequences yes/no) [estimate = −0.1, p<.0001]—that is, participants who had the fear of aversive consequences motivation (and received the corresponding module) decreased 0.1 points more per week on the ASI-Adapted than participants who did not have fear of aversive consequences. With outliers removed, results remained similar [estimate=−0.04, p=.001].
DISCUSSION
Taken together, our findings suggest that there were significant weekly improvements in all candidate measures corresponding to the three prototypical ARFID motivations (i.e., sensory sensitivity, lack of interest/low appetite, fear of aversive consequences). Pre- to post-treatment improvements in measures corresponding to ARFID motivations also all had medium to large correlations with decreases in ARFID severity. Weekly improvements in sensory sensitivity and fear of aversive consequences were larger for those who received the corresponding module compared to those who did not.
While we found overall large decreases in each candidate measure across treatment, there were only significant interactions for the sensory sensitivity and fear of aversive consequences modules. It is possible that interventions received by all participants (e.g., regular eating) could have affected overall changes in lack of interest/low appetite. The lack of interest/low appetite measure items capture features specific to early satiation (i.e., getting full quickly) rather than low appetite overall. The interaction for sensory sensitivity was also smaller than that for fear of aversive consequences and was no longer significant after outliers were removed—this could be because the sensory sensitivity measure captures characteristics present across all ARFID motivations (e.g., low enjoyment of avoided foods). The ASI-adapted may also have greater specificity compared to other measures used, specific to fear of aversive consequences that are actually non-specific to food intake. However, there was also wide variability in whether the lack of interest/low appetite and fear of aversive consequences measures changed compared to the sensory sensitivity measure, possibly reflective of the predominance of sensory sensitivity (62%) in our sample. Future research is needed to identify which measures adequately capture the three ARFID prototypical motivations (e.g., the PARDI-ARQ, Nine Item ARFID Screen) and what other behavioral and biologic factors may mediate outcomes.
While a strength of our paper is the multiple datapoints with session-level data for the candidate mechanisms, limitations include the open trial design, small sample size, use of proxy measures, and lack of more timepoints for the outcome variable (i.e., ARFID severity). While appropriate for this stage of exploratory investigation, the limited sample size also precluded the ability to evaluate heterogenous sample characteristics (e.g., age group, treatment length, treatment format), which should be evaluated in future research as treatment moderators. We were only able to examine change in each mechanism when an individual had the presentation for a module (e.g., all individuals with sensory sensitivity received the sensory sensitivity module—those who did not receive it did not have the sensory sensitivity presentation). We also had inconsistent timing of when each participant received different CBT-AR modules—delivery was flexible either based on clinician judgment or other factors (e.g., lack of interest module provided first to those who needed to gain weight). Further research is needed with randomized designs aimed at parsimony (e.g., multiphase optimization strategy; MOST) and identifying the best timing of interventions for optimal target engagement (e.g., sequential multiple assignment randomized trial; SMART) (Powell et al., 2021).
Supplementary Material
Public significance statement:
The mechanisms through which components of cognitive-behavioral therapy for ARFID (CBT-AR) work have yet to be elucidated. We conducted an exploratory investigation to test if assignment (vs. not) to a CBT-AR module resulted in larger improvements in the corresponding prototypic ARFID motivation that the module intended to target. Measures of the sensory sensitivity and the fear of aversive consequences motivations improved more in those who received the corresponding treatment module, whereas the lack of interest/low appetite measure improved regardless of if the corresponding module was received.
Funding:
This study was funded by the Hilda & Preston Davis Foundation, American Psychological Foundation, Global Foundation for Eating Disorders, and R01 MH108595 (PIs Thomas/Lawson/Micali). This manuscript was also supported by the National Institute of Diabetes and Digestive and Kidney Diseases (HBM, K23 DK131334; Nutrition Obesity Research Center at Harvard, P30 DK040561) and the National Institute of Mental Health (KRB, K23 MH125143; LB K23 MH127465; EAL K24 MH120568). Our funding sources played no role in study design, data collection, data analysis and interpretation, writing of the report, or in the decision to submit the article for publication.
Conflict of interest:
Drs. Thomas and Eddy receive royalties from Cambridge University Press for the sale of their books, Cognitive-Behavioral Therapy for Avoidant/Restrictive Food Intake Disorder: Children, Adolescents, and Adults and The Picky Eater’s Recovery Book. Dr. Becker receives royalties from Cambridge University Press for the sale of The Picky Eater’s Recovery Book. Drs. Misra and Lawson receives royalties from UpToDate.
Data availability statement:
The data that support the findings of this study are available from the corresponding author upon reasonable request.
REFERENCES
- APA. (2021). Diagnostic and Statistical Manual of Mental Disorders (DSM-5-TR) (5th- Text Revision ed.). American Psychiatric Publishing, Arlington, VA. [Google Scholar]
- Bryant-Waugh R, Micali N, Cooke L, Lawson EA, Eddy KT, & Thomas JJ (2019). The Pica, ARFID, and Rumination Disorder Interview, a multi-informant, semi-structured interview of feeding disorders across the lifespan: A pilot study for ages 10–22. International Journal of Eating Disorders, 52, 378–387. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Burton Murray H,I,W, Becker KR Ljótsson B, Madva E, Staller K, Kuo B, & Thomas JJ (2023). Development of a brief cognitive-behavioral treatment for avoidant/restrictive food intake disorder in the context of disorders of gut-brain interaction: Initial feasibility, acceptability, and clinical outcomes. International Journal of Eating Disorders, 56(3), 616–627. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Cooper-Vince CE, Nwaka C, Eddy KT, Misra M, Hadaway NA, Becker KR, Lawson EA, Cooke L, Bryant-Waugh R, Thomas JJ, & Micali N (2022). The factor structure and validity of a diagnostic interview for avoidant/restrictive food intake disorder in a sample of children, adolescents, and young adults. Int J Eat Disord, 55(11), 1575–1588. 10.1002/eat.23792 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Czajkowski SM, Powell LH, Adler N, Naar-King S, Reynolds KD, Hunter CM, Laraia B, Olster DH, Perna FM, Peterson JC, Epel E, Boyington JE, & Charlson ME (2015). From ideas to efficacy: The ORBIT model for developing behavioral treatments for chronic diseases. Health Psychol, 34(10), 971–982. 10.1037/hea0000161 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Dumont E, Jansen A, Kroes D, de Haan E, & Mulkens S (2019). A new cognitive behavior therapy for adolescents with avoidant/restrictive food intake disorder in a day treatment setting: A clinical case series. Int J Eat Disord, 52(4), 447–458. 10.1002/eat.23053 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Guastaferro K, & Collins LM (2019). Achieving the Goals of Translational Science in Public Health Intervention Research: The Multiphase Optimization Strategy (MOST). Am J Public Health, 109(S2), S128–s129. 10.2105/ajph.2018.304874 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hunot C, Fildes A, Croker H, Llewellyn CH, Wardle J, & Beeken RJ (2016). Appetitive traits and relationships with BMI in adults: Development of the Adult Eating Behaviour Questionnaire. Appetite, 105, 356–363. 10.1016/j.appet.2016.05.024 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kazdin AE (2007). Mediators and mechanisms of change in psychotherapy research. Annu. Rev. Clin. Psychol, 3, 1–27. [DOI] [PubMed] [Google Scholar]
- Kraemer HC, Wilson GT, Fairburn CG, & Agras WS (2002). Mediator of moderators of treatment effects in randomized clinical trials. Archives of General Psychiatry, 59, 877–883. [DOI] [PubMed] [Google Scholar]
- Lock J, Sadeh-Sharvit S, & L’Insalata A (2019). Feasibility of conducting a randomized clinical trial using family-based treatment for avoidant/restrictive food intake disorder. Int J Eat Disord, 52(6), 746–751. 10.1002/eat.23077 [DOI] [PubMed] [Google Scholar]
- Powell L, Freedland KE, & Kauffman P (2021). Behavioral Clinical Trials for Chronic Diseases. Springer. [Google Scholar]
- Shimshoni Y, Silverman WK, & Lebowitz ER (2020). SPACE-ARFID: A pilot trial of a novel parent-based treatment for avoidant/restrictive food intake disorder. Int J Eat Disord, 53(10), 1623–1635. 10.1002/eat.23341 [DOI] [PubMed] [Google Scholar]
- Taylor S, Zvolensky MJ, Cox BJ, Deacon B, Heimberg RG, Ledley DR, Abramowitz JS, Holaway RM, Sandin B, & Stewart SH (2007). Robust dimensions of anxiety sensitivity: development and initial validation of the Anxiety Sensitivity Index-3. Psychological assessment, 19(2), 176. [DOI] [PubMed] [Google Scholar]
- Team, R. C. (2014). R: A language and environment for statistical computing. In http://www.R-project.org/
- Thomas JJ, Becker KR, Breithaupt L, Murray HB, Jo JH, Kuhnle MC, Dreier MJ, Harshman SG, Kahn DL, Hauser K, Slattery M, Misra M, Lawson EA, & Eddy KT (2021). Cognitive-behavioral therapy for adults with avoidant/restrictive food intake disorder. Journal of Behavioral and Cognitive Therapy, 31, 47–55. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Thomas JJ, Becker KR, Kuhnle MC, Jo JH, Harshman SG, Wons OB, Keshishian AC, Hauser K, Breithaupt L, Liebman RE, Misra M, Wilhelm S, Lawson EA, & Eddy KT (2020). Cognitive-behavioral therapy for avoidant/restrictive food intake disorder: Feasibility, acceptability, and proof-of-concept for children and adolescents. Int J Eat Disord, 53(10), 1636–1646. 10.1002/eat.23355 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Thomas JJ, & Eddy K (2019). Cognitive-behavioral therapy for avoidant/restrictive food intake disorder: children, adolescents, and adults. Cambridge University Press. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.