Abstract
Objective.
This study examined baseline predictors of rapid response and its prognostic significance in a clinical trial of behaviorally-based weight-loss treatment (BBWLT) for BED in patients with obesity.
Methods.
191 participants receiving BBWLT were assessed at baseline, throughout treatment, and posttreatment (6 months) by independent assessors. Rapid response was defined as ≥65% reduction in binge eating by the fourth treatment week. Patients with versus without rapid response were compared on demographic features, a broad range of current/past clinical and psychiatric variables, and treatment attitudes. Rapid response was used to prospectively predict posttreatment outcomes.
Results.
Rapid response, which characterized 63% (N=120) of participants, was not associated significantly with any demographic features or with any current/past clinical and psychiatric variables. Higher ratings (at week one) regarding the logic of BBWLT and greater confidence treatment would help with binge eating and weight loss were associated significantly with rapid response (at week four). Rapid response was prospectively associated with significantly better binge-eating and weight-loss outcomes.
Conclusions.
Our findings indicate that rapid response to BBWLT for BED prospectively predicts superior clinical outcomes in both binge eating and weight loss. Treatment attitudes, rather than patient demographic or clinical severity variables, are prospectively associated with rapid response.
Keywords: obesity, eating disorders, treatment, binge eating, weight loss
Introduction
Binge-eating disorder (BED) is defined by recurrent binge eating, marked distress about binge eating, and the absence of extreme weight-compensatory behaviors (1). BED is a serious public health problem with high social and economic costs (2). BED is prevalent, associated strongly with obesity, with increased risk for psychiatric and medical comorbidities, and with serious psychosocial impairments (3,4). Despite the public health significance of BED, most people with BED go untreated and those who seek help infrequently receive evidence-based interventions (5).
Treatment research has produced empirical support for specific psychological treatments and certain pharmacological agents (6,7,8). Even with the “best-available” treatments, many patients with BED do not benefit sufficiently; a substantial minority do not achieve abstinence from binge eating (9) and many do not lose weight (6). Research combining treatments has generally not resulted in enhanced outcomes for BED (10). In the hopes of improving outcomes, research has aimed at identifying predictors, moderators, and mediators of treatment response as one avenue to refine treatment approaches (11). To date, few reliable predictors and nearly no significant moderators have been found (12,13).
Although most patient factors have not reliably predicted/moderated treatment outcomes for BED or for other eating disorders (12), a treatment process referred to as early or “rapid response” (RR) to treatment has emerged as a consistent predictor of positive therapeutic benefits (14,15,16). As detailed by Grilo and colleagues (17), following initial observations of clinically meaningful early improvements in depression, research with bulimia nervosa (e.g., 18) applied variants of signal detection methods and studies reported evidence for the specific predictive utility of early rapid response (generally 50–60% reductions by the 4th week of treatment) for favorable treatment outcomes. In a series of studies with BED, Grilo and colleagues found that RR defined as 65%−70% reductions in binge-eating frequency by the 4th week of treatments for BED – identified using receiver operating characteristic (ROC) curves, prospectively predicted important clinical outcomes with a variety of distinct interventions (17,19,20,21,22). Initial studies of RR specifically to behavioral weight loss treatments for BED found RR prospectively predicted good outcomes in both binge eating and weight loss (20,22). Findings that RR is predictive of significant weight losses in BED is particularly important given the strong association between BED and obesity (3) and especially noteworthy given that weight losses with most treatments for BED tend to be minimal (6) and are typically less than those reported in treatment studies for obesity without BED (23).
While RR has predicted positive outcomes to treatments reliably across eating disorders, including BED (14,15,16), very little is known about who will respond rapidly to treatment (15). Contrary to clinical lore that “less complicated patients respond better,” studies to date have failed to identify differences in sociodemographic or clinical severity variables associated with RR to treatments for BED (17,19,21,22). Previous studies of BED and critical reviews of RR across the different eating disorders have highlighted the need for further research on RR to specific treatments with a greater focus on identifying within-treatment mechanisms associated with a RR (15). An improved understanding of RR is needed as that might prove informative for understanding either predictors of RR or provide clues about the processes or mechanisms involved in RR’s associations with treatment responses which, in turn, could refine treatments further.
The current study aimed to examine predictors of RR and the prognostic significance of RRs for predicting outcomes specifically to behaviorally-based weight loss treatment (BBWLT) using data from a randomized controlled trial (RCT; 24). As context for the present study of RR to BBWLT, we briefly summarize the design and the primary treatment outcomes previously reported (24). The primary treatment trial evaluated two 6-month BBWLTs: standard behavioral weight-loss versus Stepped-Care treatment beginning with standard behavioral weight-loss. Both BBWLT conditions began with identical BWL delivered for four weeks, at which time RR was asessed; that is, RR to BWL was assessed following the same exact four BWL sessions prior to any additional interventions (which followed only in the Stepped Care condition). As context for the present study and analyses, we highlight that the two BBWLT conditions did not differ significantly either in the proportion categorized with RR following one month of treatment (24) or on any primary or secondary outcomes (24), and therefore are combined here in our analyses of RR. Both BBWLTs were associated with significant improvements in binge eating (74.4% and 66.5% abstinence rates, respectively) and weight loss (5.1% and 5.8%, respectively).
The primary aim of the current study (an a priori secondary aim) was to examine the significance of RR. The first aim was to explore sociodemographic and clinical variables as predictors of RR specifically for BBWLT for BED by replicating past findings plus testing additional clinical variables. We hypothesized that patient sociodemographic, baseline clinical characteristics, and psychiatric comorbidity would be unrelated to RR. The second aim was to examine whether patients’ treatment expectations would predict RR. In the sole study to examine this question to date, RR to dialectical behavior therapy (DBT) for BED was associated with significantly higher treatment expectations for success (25). Accordingly, we hypothesized that patients’ positive expectations would be associated with RR. Should this hypothesis be confirmed, we planned to explore for a possible sequential relationship between treatment expectations, RR, and clinical outcomes at post-treatment. The final aim was to test the prognostic significance of RR to BBWLT for BED. We hypothesized that RR would be associated with significantly greater clinical improvements in primary (i.e., reductions in binge eating and in weight) and secondary (i.e., associated eating-disorder psychopathology and depression levels) outcomes.
Methods
Participants
Participants were a consecutive series of 191 patients with BED with co-existing obesity recruited via media advertisements who were randomized to a single-site treatment study testing BBWLT at a medical school program (24). A detailed description of the treatment study and primary outcomes is published (24); the key components are summarized here. Eligibility criteria included meeting DSM-IV-TR criteria for BED, having a BMI ≥ 30, and being between 18 and 60 years of age. Exclusionary criteria included serious mental illness (e.g., psychosis or bipolar disorder), pregnancy or breastfeeding, medical conditions that impact eating and weight (e.g., uncontrolled diabetes or thyroid problems), and current treatment for eating/weight. The study was approved by Yale Institutional Review Board, and all participants provided written informed consent.
Assessments
Assessment and diagnostic procedures were performed by trained and monitored doctoral clinicians (psychologists) at baseline and post-treatment. BED diagnosis and psychiatric diagnoses (current or lifetime mood, anxiety, substance use, anorexia nervosa, and bulimia nervosa) were assessed using the Structured Clinical Interview for DSM-IV Axis I Disorders (SCID-I/P; 26). In addition to the SCID-I/P, a structured clinical interview assessed the age at participants were first overweight, experienced their first binge eating episode, and onset of first meeting BED diagnosis.
Eating Disorder Examination (EDE), a semi-structured interview that assesses current eating-disorder psychopathology (27), was administered at baseline and at post-treatment. The EDE is a well-established assessment with good reliability (28). The EDE assesses binge eating frequency (i.e., an objectively unusually large quantity of food eaten while experiencing a subjective loss of control during a discrete time period). The EDE has four subscales (dietary restraint, eating concern, weight concern, and shape concern) and generates a global total score reflecting eating-disorder psychopathology. Items are rated on a scale of 0 to 6, with higher scores representing greater frequency/severity.
Both height and weight were measured at baseline and post-treatment. Body mass index (BMI) and percent weight decrease were calculated from these measurements.
A battery of self-report measures was completed by participants which included the measures listed below. Emotional Overeating Questionnaire (EOQ) is a 6-item questionnaire that assesses overeating frequency in response to different emotional triggers (29). Fat Phobia Scale (FPS) is a 14-item scale that measures opinions towards people with obesity, with higher scores reflecting more negative views (30). Yale Food Addiction Scale (YFAS) is a 16-item scale that measures the construct of “food addiction” by assessing craving, tolerance, and withdrawal with respect to specific foods (31). Weight Bias Internalization Scale (WBIS) is a 11-item scale developed as a measure to assess self-directed stigma related to beliefs about the implications of weight status, with higher scores reflecting greater internalized bias (32). Beck Depression Inventory (BDI) is a 21-item self-report measure of symptoms of depression, with higher scores representing higher levels (33). Rosenberg Self-Esteem Scale (RSES) is a well-established measure of self-esteem (34). Difficulties in Emotion Regulation Scale (DERS) is a 36-item measure that assesses six dimensions of emotional regulation and generates a global score, with higher scores reflecting greater difficulties in emotional regulation (35). Self-Control Scale – Brief (SCS-B) is a 13-item questionnaire measuring perceived self-control in general, with higher scores indicative of greater perceived self-control (36).
Treatment attitudes were assessed using specific questions rated on Likert-type scales administered after the first treatment session during which the BBWLT was described in detail to participants. Participants provided their responses to the questions which were collected in sealed envelopes by independent assessors (i.e., kept separate from the BWL clinicians). Questions included: “How logical does this treatment program seem to you?” “How confident are you that this treatment program will help you manage your binge eating?” and “How confident are you that this treatment program will help you manage your weight?” Scores were recorded on a 7-point Likert scale ranging from “not at all” to “extremely.”
Treatment
Treatments were delivered by trained research-clinicians (doctoral students) monitored for adherence and supervised in the delivery of the manualized treatment protocols. The BBWLTs (standard care and Stepped-Care) followed manualized protocols of empirically supported interventions (6,7) matched in total sessions and time to minimize differences in attention. The standard care comprised 16 BBWLT sessions and the Stepped-Care comprised 15 sessions (either BBWLT or 4 BBWLT sessions followed by 11 guided-self-help-CBT sessions) plus pharmacotherapy with FDA-approved weight-loss agents.
Statistical Analysis
Individuals were considered “rapid responders” (RRs) if they experienced a 65% or greater decrease in binge eating episodes by week 4 of treatment (17). Percent decrease in binge eating episodes was calculated by assessing total episodes at baseline and at week 4. A total of 10 participants dropped out of treatment prior to being assessed for RR, and 61 participants experienced less than a 65% reduction, all of whom were considered non-rapid responders (non-RRs).
Statistical analyses were conducted using SPSS version 26.0. We examined differences between RRs and non-RRs in a series of univariate and multivariate analyses of variance and chi-squared tests. Mediational analyses were completed in a series of logistic regressions: first, we tested the relationship on the predictor on the outcome variable. Then, we tested the relationship of the mediator to the outcome variable. Finally, we included both the predictor and the mediator variable on the outcome variable. If the mediator was significant and the predictor variable was no longer significant, the outcome was interpreted to be fully mediated by RR.
Results
Baseline Participant Characteristics
Table 1 shows the demographic information of rapid (n=120; 62.8%) and non-RRs (n=71; 37.2%). No significant differences between RRs and non-RRs were observed on any sociodemographic variables, including age, race/ethnicity, sex, sexual orientation, and highest level of education (all ps > .12). No significant differences between RRs and non-RRs were observed on any of the clinical features, current/lifetime psychiatric comorbidities, or associated psychological variables (all ps > .05; Tables 2–3).
Table 1.
Demographic characteristics of participants with a rapid response versus without a rapid response
| Rapid Response n = 120 (62.8%) | No Rapid Response n = 71 (37.2%) | Statistics | |
|---|---|---|---|
| Age, mean (S.D.) | 48.53 (9.16) | 48.21 (10.07) | F(1,189) = 0.05; p = .82; ηp2 < .001 |
| Race/Ethnicity, n (%) | χ2(4) = 3.84; p = .43; Φ = .14 | ||
| Black | 21 (17.5%) | 7 (9.9%) | |
| Hispanic/Latino | 6 (5.0%) | 2 (2.8%) | |
| White | 91 (75.8%) | 59 (83.1%) | |
| Asian | 1 (.8%) | 1 (1.4%) | |
| Other | 1 (.8%) | 2 (2.8%) | |
| Sex, n (%) | χ2(1) = 2.27; p = .13; Φ = .11 | ||
| Female | 90 (75.0%) | 46 (64.8%) | |
| Male | 30 (25.0%) | 25 (35.2%) | |
| Sexual Orientation, n (%) | χ2(2) = 0.73; p = .69; Φ = .06 | ||
| Gay | 1 (.8%) | 1 (1.4%) | |
| Bisexual | 1 (.8%) | 0 (0%) | |
| Heterosexual | 116 (98.3%) | 68 (98.6%) | |
| Highest level of education, n (%) | χ2(5) = 6.75; p = .24; Φ = .19 | ||
| did not complete HS | 3 (2.5%) | 0 (0%) | |
| HS diploma | 33 (27.7%) | 23 (32.9%) | |
| Associate Degree | 19 (16.0%) | 13 (18.6%) | |
| Bachelor’s Degree | 33 (27.7%) | 21 (30.0%) | |
| Master’s Degree | 27 (22.7%) | 8 (11.4%) | |
| Doctoral Degree | 4 (3.4%) | 5 (7.1%) |
Table 2.
Clinical characteristics and psychiatric comorbidities of participants with a rapid response versus without a rapid response
| Rapid Response n = 120 (62.8%) | No Rapid Response n = 71 (37.2%) | Statistics | |
|---|---|---|---|
| Clinical characteristics, mean (S.D.) | |||
| Age first overweight, m (sd) | 19.78 (11.34) | 17.34 (9.72) | F(1,169) = 1.99; p = .16; ηp2 = .01 |
| Age first binge m (sd) | 26.00 (14.33) | 25.14 (13.21) | F(1,180) = .17; p = .69; ηp2 < .01 |
| Age first BED m (sd) | 29.71 (15.00) | 27.00 (13.93) | F(1,189) = 1.53; p = .22; ηp2 < .01 |
| Psychiatric comorbidities, n (%) | |||
| Mood disorder current | 24 (20.0) | 17 (23.9) | χ2(1) = .41; p = .52; Φ = .05 |
| Lifetime mood disorder | 65 (54.2) | 34 (47.9) | χ2(1) = .70; p = .40; Φ = .06 |
| Anxiety disorder current | 31 (25.8) | 21 (29.6) | χ2(1) = 0.32; p = .57; Φ = .04 |
| Lifetime anxiety disorder | 41 (34.2) | 32 (45.1) | χ2(1) = 2.25; p = .13; Φ = .11 |
| Lifetime substance use disorder | 36 (30.0) | 21 (31.0) | χ2(1) = .02; p = .89; Φ = .01 |
| Lifetime AN/BN diagnosis | 8 (6.7) | 6 (8.5) | χ2(1) = .21; p = .65; Φ = .03 |
Note. BED = binge-eating disorder; AN = anorexia nervosa; BN = bulimia nervosa
Table 3.
Baseline eating disorder and associated psychiatric characteristics of participants with a rapid response versus without a rapid response
| Rapid Response n = 120 (62.8%) | No Rapid Response n = 71 (37.2%) | Statistics | |
|---|---|---|---|
| ED characteristics, mean (S.D.) | |||
| OBE frequency | 21.26 (15.62) | 17.11 (12.64) | F(1,189) = 3.60; p = .06; ηp2 = .02 |
| BMI | 38.52 (5.67) | 39.77 (6.42) | F(1,189) = 1.99; p = .16; ηp2 = .01 |
| EDE Global | 2.68 (.91) | 2.68 (.92) | F(1,189) < .001; p = .99; ηp2 < .001 |
| EDE Subscales | F(4,186) = 1.59; p = .18; ηp2 = .03 | ||
| Restraint | 1.77 (1.21) | 1.61 (1.29) | NS |
| Eating concern | 2.20 (1.28) | 2.30 (1.35) | NS |
| Shape concern | 3.55 (1.25) | 3.69 (1.11) | NS |
| Weight concern | 3.22 (1.11) | 3.13 (1.11) | NS |
| EOQ mean | 1.78 (1.18) | 1.83 (1.24) | F 1,188) = .07; p = .79; ηp2 < .001 |
| FPS | 3.65 (.71) | 3.74 (.73) | F(1,187) = .67; p = .42; ηp2 < .01 |
| YFAS total | 4.76 (1.80) | 4.82 (1.80) | F(1,185) = .05; p = .83; ηp2 < .001 |
| WBIS total | 4.46 (1.23) | 4.79 (1.31) | F(1,188) = 2.98; p = .09; ηp2 = .02 |
| Psychiatric characteristics, mean (S.D.) | |||
| BDI total | 14.53 (8.46) | 15.93 (9.05) | F(1,188) = 1.16; p = .28; ηp2 < .01 |
| RSES total | 30.20 (6.35) | 28.89 (7.11) | F(1,188) = 1.72; p = .19; ηp2 < .01 |
| DERS overall | 77.28 (22.73) | 82.61 (24.73) | F(1,187) = 2.25; p = .14; ηp2 = .01 |
| SCS-B | 39.64 (7.83) | 38.30 (8.61) | F(1,185) = 1.18; p = .28; ηp2 < .01 |
Note. ED = eating disorder; OBE = objective bulimic episode; BMI = body mass index; EDE = eating disorder examination interview; EOQ = Emotional Overeating questionnaire; FPS = Fat Phobia Scale; YFAS = Yale Food Addiction Scale; WBIS = Weight Bias Internalization Scale; BDI = Beck Depression Inventory; RSES = Rosenberg’s Self-Esteem Scale; DERS = Difficulty in Emotion Regulation Scale; SCS-B = Brief Self-Control Scale
Treatment Attitudes
Table 4 summarizes differences between RRs and non-RRs on treatment attitudes. Participants categorized with RR following one month of treatment rated treatment (after one session) as significantly more logical and more likely to help them lose weight and help with binge eating (all ps < .03) than participants categorized as non-RRs.
Table 4.
Treatment attitudes of participants with a rapid response versus without a rapid response
| Rapid Response n = 120 (62.8%) | No Rapid Response n = 71 (37.2%) | Statistics | |
|---|---|---|---|
| Logical treatment | 5.12 (1.12) | 4.65 (1.15) | F(1,152) = 5.88; p = .02; ηp2 = .04 |
| Confidence in help BED | 4.59 (1.39) | 4.00 (1.34) | F(1,152) = 6.34; p = .01; ηp2 = .04 |
| Confidence in decrease weight | 4.50 (1.42) | 3.92 (1.45) | F(1,152) = 5.30; p = .02; ηp2 = .03 |
Note. Results presented in mean (S.D.) format. BED = binge-eating disorder
Post-Treatment Outcomes
Table 5 summarizes analyses comparing RR and non-RR groups on clinical outcomes, including weight status, changes in eating disorder symptomology, and other related psychological outcomes. Participants categorized with RR had significantly lower binge eating frequency (p = .01) and significantly greater percent decrease in binge eating frequency (p < .01) than those without RR. Participants categorized with RR had significantly greater weight loss (p < .001) and were significantly more likely to achieve 5% weight loss (p < .01) than participants without RR. RR was not associated significantly with eating-disorder psychopathology (EDE global) or depression (BDI-II) scores at post-treatment. Parallel analyses adjusting for baseline levels did not result in changed findings. Furthermore, parallel analyses restricted to the Step Care (N=152) condition revealed nearly identical findings.
Table 5.
Clinical outcomes at post-treatment of participants with a rapid response versus without a rapid response
| Rapid Response n = 120 (62.8%) | No Rapid Response n = 71 (37.2%) | Statistics | |
|---|---|---|---|
| % weight loss, mean (S.D.) | − 6.93 (6.18) | − 3.01 (7.11) | F(1,161) = 13.21; p < .001; ηp2 = .08 |
| 5% weight loss n (%) | 64 (56.6) | 18 (31.0) | χ2(1) = 10.07; p < .01; Φ = .24 |
| OBE frequency, mean (S.D.) | 1.43 (2.96) | 4.42 (11.60) | F(1,168) = 6.67; p = .01; ηp2 = .04 |
| %decrease OBE, mean (S.D.) | 91.11 (22.64) | 74.49 (57.14) | F(1,168) = 7.32 p < .01; ηp2 = .04 |
| EDE Global, mean (S.D.) | 1.75 (.83) | 1.81(.95) | F(1,170) = .17; p = .68; ηp2 < .01 |
| EDE Subscales, mean (S.D.) | F(4,165) = 2.22; p = .07; ηp2 = .05 | ||
| Restraint | 2.01 (1.16) | 1.68 (1.19) | NS |
| Eating concern | .76 (.99) | .81 (1.02) | NS |
| Shape concern | 2.06 (1.26) | 2.39 (1.37) | NS |
| Weight concern | 2.17 (.99) | 2.36 (1.13) | NS |
| BDI total, mean (S.D.) | 8.35 (7.16) | 10.71 (9.64) | F(1,168) = 3.26; p = .07; ηp2 = .02 |
Note. OBE = objective bulimic episode; EDE = Eating Disorder Examination interview, BDI = Beck Depression Inventory
Exploratory Analysis: Mediational Model
We conducted an exploratory analysis to assess for a temporal relationship between averaged treatment attitudes (if treatment is logical, will help with binge eating, and will help with weight loss), RR to treatment (yes/no), and whether patients achieved 5% weight loss at post-treatment. RR fully mediated the relationship between treatment attitudes and 5% weight loss at post-treatment. Logistic regression was first used to analyze the relationship between treatment attitudes and 5% weight loss at post. We found that as treatment attitudes increased, the likelihood of losing 5% of weight also increased. (χ2(1)=7.46, B=0.40, SE=0.15, Wald=6.74, p<0.01). Next, we found that as treatment attitudes increased, the likelihood of being a rapid responder also increased (χ2(1)=7.13, B=−0.39, SE=0.15, Wald=6.75, p< 0.01). Finally, when including both treatment attitudes and RR as predictors, treatment attitudes were no longer a significant predictor (χ2(2)=18.40, B=0.30, SE=0.15, Wald=3.84, p=0.50), while RR was (χ2(2)=18.40, B = −1.24, SE = .39, Wald = 10.23, p < .01), indicating full mediation.
Discussion
This study compared rapid responders to non-rapid responders to behaviorally-based weight loss treatment for BED on a variety of sociodemographic and clinical features at both baseline and post-treatment. The first major finding was that RR was not associated with any patient characteristics, including psychiatric history, examined prior to treatment. These findings replicate and extend previous similar reports (17,19,21,22). We emphasize, however, that this finding was despite our “shotgun” strategy of considering many measures in an attempt to explore broader clinical characteristics than examined in previous studies. Collectively, this finding suggests that RR to BBWLT occurs regardless of the complexity of patients’ sociodemographic characteristics, clinical presentations, and BED/weight and psychiatric comorbidity histories. Put differently, RR does not occur just in patients with fewer sociodemographic disadvantages or those with lower levels of psychopathology.
The second main finding was that RR was prospectively associated with significantly better outcomes in binge eating and weight loss at post-treatment. This replicates previous findings for BWL (20,21) and provides further confidence in the specificity of the predictive utility of RR to BBWLT; this is important as the literature shows that RR differs across different pharmacological versus psychological (17) and between different psychological treatments for BED (21,25).
The third main and novel finding was that RR (determined after 4 weeks of treatment) was associated significantly with treatment attitudes (assessed after the first treatment session). Specifically, participants who had a RR had significantly higher ratings of how logical and how confident they were that the BBWLT would be for their BED and associated excess weight than those who did not attain RR. These findings are consistent with those reported in one previous study of DBT for BED that patients’ expectations for success were associated with RR (25). The BWL manualized protocols emphasize the treatment as an active collaboration between the clinician and patient and therefore begin the treatment with a detailed overview of the therapeutic model and the patient’s active involvement in the process. This highlights the importance of clinicians fully explaining the treatment rationale and helping the patients understand how their efforts within the treatment model will play a role in helping them achieve their goals. Indeed, research on different psychoeducation approaches and models can result in different perceptions of credibility and expectations for effectiveness, with some evidence favoring certain psychological (cognitive and behavioral) approaches (37). More broadly, research has highlighted the importance of attitudes towards treatment and therapist for enhancing general psychotherapy outcomes (38), and associations with less frequent premature termination from psychotherapy for anorexia nervosa (39) and with higher remission rates for bulimia nervosa (40).
Finally, we found a full mediational model with RR mediating the relationship between treatment attitudes at the start of treatment and 5% weight loss at post-treatment. This finding further emphasizes the importance of providing clear rationale for treatment and explaining the possible benefits (such as decreased binge eating and decreased weight). More favorable attitudes towards treatment (greater logic and confidence) were related to a greater likelihood of being a rapid responder, and being a rapid responder increased the likelihood of achieving 5% weight loss at posttreatment. This finding suggests that that patient’s perceptions and confidence in a specific treatment from the outset (in the first week) impacts weight outcomes; this novel finding warrants future research and empirical replication, despite its “clinical commonsense.” More broadly, patient’s treatment “preferences” have been reported to be associated with treatment outcomes for depression (41), but treatment preferences had yet to be tested in eating disorder trials (42).
We note several strengths and potential limitations of the study to consider as context for the findings. The manualized BBWLT treatments were delivered by highly trained and closely monitoring research clinicians to maintain fidelity and changes throughout the course of treatment and at posttreatment were independently and reliably assessed with psychometrically established measures. RR was assessed using an empirically developed definition supported in previous studies of RR in BED. Our study focused on RR specifically to BBWLT delivered in a carefully monitored research setting; generalizability to other treatments, to different clinical settings and practitioners, and to persons with different sociodemographic and clinical characteristics or to those who do not participate in research studies in uncertain. These findings should be further replicated in patients who seek alternative treatments in different clinical settings and practitioners. For example, studies of treatment delivery in the community suggests considerable departures from evidence-based and manualized treatments used in this and other studies (43). Relatedly, the careful “informed consent” (IRB protocols) and close adherence to the manualized BBWLT protocol may depart somewhat from how BWL treatments are initiated and explained to patients in some busy clinical settings. Patients’ ratings of “logic” and “confidence” in the treatments were assessed using single-item questions, rather than validated questionnaires. While future research should examine treatment attitudes more comprehensively, we note that research has documented advantages to using clear and concrete single-item questions versus multiple measures for complex constructs (44). We also note that, while 5% weight losses appear clinically meaningful (i.e., related to important positive health outcomes, 45), it is possible that some patients with excess weight might nonetheless feel dissatisfied with such weight loss, and this suggests the importance of refining interventions to both promote further weight losses to improve physical health while and enhancing body image and self-evaluation to improve psychological health
Our findings add further to the literature indicating the predictive utility of RR for attaining positive outcomes in both binge eating and weight loss specifically with BBWLT (20,21). Our study yielded new findings suggesting the potentially important role of enhancing patient’s beliefs about the “logic” of BBWLT for BED and their “confidence” that this evidence-based behavioral treatment can help them with both their binge eating and weight loss. This suggests the importance of providing patients with a clear rationale for the BBWLT, the patient’s active role in the treatment collaboration, and reasonable expectations for what can be achieved. The findings regarding patient’s confidence in the treatment suggest that clinicians should be poised to provide additional support and guidance to patients who do not show a rapid response to treatment, and this might be informed by re-assessing their beliefs about the treatments and concerns that it might not help. A related potential related avenue for research to guide interventions concerns patient’s confidence (i.e., “self-efficacy”) in themselves for achieving lifestyle behavioral changes.
In summary, RR was not predicted by sociodemographic variables, past or present comorbid psychiatric disorders, or current severity of illness or associated clinical variables. In contrast, positive treatment attitudes (“logic” of treatment and “confidence” that the treatment would help binge eating and weight loss) predicted RR – and in turn – was associated with the likelihood of achieving 5% weight loss at post-treatment. These new findings emphasize the importance of beginning treatment with a clear and logical rationale for the treatment and what it can reasonably help to accomplish.
Study Importance.
What is already known about this subject?
Rapid response (RR) is a reliable predictor of treatment response to certain interventions for binge-eating disorder (BED).
It remains unclear who responds rapidly to behaviorally-based weight loss treatments (BBWLT) for BED and its prognostic significance for broad outcomes is uncertain.
What are the new findings in your manuscript?
Patient factors, severity of psychopathology, and psychiatric comorbidity are unrelated to RR to BBWLT for BED.
Patients with RR rated the treatments after the first week as significantly more logical and more confidently that they would help them.
Rapid response was prospectively associated with significantly better binge-eating and weight-loss outcomes; mediational model suggested positive treatment attitudes lead to rapid response, which in turn, leads to achieving 5% weight loss.
How might your results change the direction of research or the focus of clinical practice?
Clinicians should provide patients with a clear rationale for the treatment and reasonable expectations for what can be achieved.
Clinicians should be poised to provide additional support and guidance to patients who do not show an early rapid response to treatment.
Funding:
This research was supported by National Institutes of Health (NIH) grant R01 DK49587 (Dr. Grilo). The authors were also supported, in part, by NIH grants R01 DK114075 and R01 DK112771. Funding agency played no role in the content of this paper.
Footnotes
Potential Conflicts of interest: The authors declare no conflicts of interest. Dr. Grilo reports broader interests, which did not influence this research, including Honoraria for lectures, CME activities, and lectures at scientific conferences, and Royalties from Guilford Press and Taylor & Francis Publishers for academic books.
Data Sharing:
De-identified data will be provided in response to reasonable written request to achieve goals in an approved written proposal.
References
- 1.American Psychiatric Association. Diagnostic and statistical manual of mental disorders, Fifth edition (DSM-5). Washington, DC: American Psychiatric Association, 2013. [Google Scholar]
- 2.Streatfeild J, Hickson J, Austin SB, Hutcheson R, Kandel JS, Lampert JG… Pezullo L. Social and economic cost of eating disorders in the United States: evidence to inform policy action. Int J Eat Disord 2021;54:851–868. [DOI] [PubMed] [Google Scholar]
- 3.Udo T, Grilo CM. Prevalence and correlates of DSM-5-defined eating disorders in a nationally representative sample of US adults. Biol Psychiatry 2018;84:345–354. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Udo T, Grilo CM. Psychiatric and medical correlates of DSM-5 eating disorders in a nationally representative sample of adults in the United States. Int J Eat Disord 2019;52:42–50. [DOI] [PubMed] [Google Scholar]
- 5.Coffino JA, Udo T, Grilo CM. Rates of help-seeking in U.S. adults with lifetime DSM-5 eating disorders: Prevalence across diagnoses and sex and ethnic/racial differences. Mayo Clinic Proceedings 2019;94:1415–1426. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Hilbert A, Petroff D, Herpertz S, Pietrowsky R, Tuschen-Caffier B, Vocks S, Schmidt R. Meta-analysis of the efficacy of psychological and medical treatments for binge-eating disorder. J Consult Clin Psychol 2019;87:91–105. [DOI] [PubMed] [Google Scholar]
- 7.Grilo CM. Psychological and behavioral treatments for binge-eating disorder. J Clin Psychiatry 2018;78(S1): 20–24. [DOI] [PubMed] [Google Scholar]
- 8.McElroy SL. Pharmacologic treatments for binge-eating disorder. J Clin Psychiatry 2017;78(S1):14–19. [DOI] [PubMed] [Google Scholar]
- 9.Linardon J Rates of abstinence following psychological or behavioral treatments for binge-eating disorder: Meta-analysis. Int J Eat Disord 2018;51:785–797. [DOI] [PubMed] [Google Scholar]
- 10.Reas DL, Grilo CM. Psychotherapy and medications for eating disorders: better together?. Clin Ther 2021;43:17–39. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Kraemer HC, Wilson GT, Fairburn CG, Agras WS. Mediators and moderators of treatment effects in randomized clinical trials. Arch Gen Psychiatry 2002;59:877–883. [DOI] [PubMed] [Google Scholar]
- 12.Linardon J, de la Piedad Garcia X, Brennan L. Predictors, moderators, and mediators of treatment outcome following manualised cognitive-behavioural therapy for eating disorders: A systematic review. Eur Eat Disord Rev 2017;25:3–12. [DOI] [PubMed] [Google Scholar]
- 13.Vall E, Wade TD. Predictors of treatment outcome in individuals with eating disorders: a systematic review and meta-analysis. Int J Eat Disord 2015;48:946–971. [DOI] [PubMed] [Google Scholar]
- 14.Chang PG, Delgadillo J, Waller G. Early response to psychological treatment for eating disorders: a systematic review and meta-analysis. Clin Psychol Review 2021;86:102032. [DOI] [PubMed] [Google Scholar]
- 15.Linardon J, Brennan L, de la Piedad Garcia X. Rapid response to eating disorder treatment: a systematic review and meta-analysis. Int J Eat Disord 2016;49:905–919. [DOI] [PubMed] [Google Scholar]
- 16.Nazar BP, Gregor LK, Albano G et al. Early response to treatment in eating disorder: a systematic review and a diagnostic test accuracy meta-analysis. Eur Eat Disord Rev 2017;25:67–79 [DOI] [PubMed] [Google Scholar]
- 17.Grilo CM, Masheb RM, Wilson GT. Rapid response to treatment for binge eating disorder. J Consult Clin Psychol 2006;74:602–613. [DOI] [PubMed] [Google Scholar]
- 18.Agras WS, Crow SJ, Halmi KA, Mitchell JE, Wilson GT, & Kraemer HC. Outcome predictors for the cognitive-behavior treatment of bulimia nervosa: data from a multisite study. Am J Psychiatry 2000;157:1302–1308. [DOI] [PubMed] [Google Scholar]
- 19.Grilo CM, Masheb RM. Rapid response predicts binge eating and weight loss in binge eating disorder: Findings from a controlled trial of orlistat with guided self-help cognitive behavioral therapy. Behav Res Ther 2007;45:2537–2550. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Masheb RM, Grilo CM. Rapid response predicts treatment outcomes in binge-eating disorder: implications for stepped care. J Consult Clin Psychol 2007;75:39–644. [DOI] [PubMed] [Google Scholar]
- 21.Grilo CM, White MA, Wilson GT, Gueorguieva R, Masheb RM. Rapid response predicts 12-month post-treatment outcomes in binge eating disorder: theoretical and clinical implications. Psychol Med 2012;42:807–817. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Grilo CM, White MA, Masheb RM, Gueorguieva R. Predicting meaningful outcomes to medication and self-help treatments for binge-eating disorder in primary care: the significance of early rapid response. J Consult Clin Psychol 2015; 83:387. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Blaine B, Rodman J. Responses to weight loss treatment among obese individuals with and without BED: matched-study meta-analysis. Eat Weight Disord 2007;12:54–60. [DOI] [PubMed] [Google Scholar]
- 24.Grilo CM, White MA, Masheb RM, Ivezaj V, Morgan PT, Gueorguieva R. Randomized controlled trial testing the effectiveness of adaptive “SMART” stepped-care treatment for adults with binge-eating disorder comorbid with obesity. Am Psychol 2020:75:204–218. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Safer DL, Joyce EE. Does rapid response to two group psychotherapies for binge eating disorder predict abstinence?. Behav Res Ther 2011;49:339–345. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.First MB, Spitzer RL, Gibbon M, Williams JBW. Structured clinical interview for DSM-IV Axis I disorders-patient Edition (SCID-I/P, Ver 2.0). NY: NYSPI, 1996. [Google Scholar]
- 27.Fairburn CG, Cooper Z, O’Connor M. (2008). The Eating Disorder Examination (16.0D). In Fairburn CG. Cognitive Behavior Therapy and Eating Disorders. NY: Guilford Press, 2008. [Google Scholar]
- 28.Grilo CM, Masheb RM, Lozano-Blanco C, Barry DT. Reliability of the Eating Disorder Examination in patients with binge eating disorder. Int J Eat Disord 2004;35:80–85. [DOI] [PubMed] [Google Scholar]
- 29.Masheb RM, Grilo CM. Emotional overeating and its associations with eating disorder psychopathology among overweight patients with binge eating disorder. Int J Eat Disord 2006; 39: 141–146. [DOI] [PubMed] [Google Scholar]
- 30.Bacon JG, Scheltema KE, Robinson BE. Fat phobia scale revisited: The short form. Int J Obes 2001; 25:252–257. [DOI] [PubMed] [Google Scholar]
- 31.Gearhardt AN, Corbin WR, Brownell KD. Preliminary validation of the Yale Food Addiction Scale. Appetite 2009; 52:430–436. [DOI] [PubMed] [Google Scholar]
- 32.Durso LE, Latner JD. Understanding self-directed stigma: development of the weight bias internalization scale. Obesity 2008;16:S80–S86. [DOI] [PubMed] [Google Scholar]
- 33.Beck AT, Steer R, Garbin M. Psychometric properties of the Beck Depression Inventory: 25 years of evaluation. Clin Psychol Rev 1998;8:77–100. [Google Scholar]
- 34.Rosenberg M Rosenberg self-esteem scale (RSE). Acceptance and commitment therapy. Measures Package 1965;61:18. [Google Scholar]
- 35.Gratz KL, Roemer L. Multidimensional assessment of emotion regulation and dysregulation: Development, factor structure, and initial validation of the difficulties in emotion regulation scale. J Psychopathol Behav Assess 2004; 26:41–54. [Google Scholar]
- 36.Tagney JP, Baumeister RF, Boone AL. High self-control predects good adjustment, less pathology, better grades, and interpersonal success. J Pers 2004; 72:271–324. [DOI] [PubMed] [Google Scholar]
- 37.Farrell NR, Lee AA, Deacon BJ. Biological or psychological? Effects of eating disorder psychoeducation on self-blame and recovery expectations among symptomatic individuals. Behav Res Ther 2015;74:32–37. [DOI] [PubMed] [Google Scholar]
- 38.Constantino MJ, Coyne AE, Boswell JF, Iles BR, Vîslă A. A meta-analysis of the association between patients’ early perception of treatment credibility and their posttreatment outcomes. Psychotherapy 2018;55:486. [DOI] [PubMed] [Google Scholar]
- 39.Jordan J, McIntosh VV, Carter FA, Joyce PR, Frampton CM, Luty SE, … Bulik CM. Predictors of premature termination from psychotherapy for anorexia nervosa: Low treatment credibility, early therapy alliance, and self-transcendence. Int J Eat Disord 2017;50:979–983. [DOI] [PubMed] [Google Scholar]
- 40.Wilson GT, Loeb KL, Walsh BT, Labouvie E, Petkova E, Liu X, Waternaux C. Psychological versus pharmacological treatments of bulimia nervosa: predictors and processes of change. J Consult Clin Psychol 1999;67:451. [DOI] [PubMed] [Google Scholar]
- 41.Kocsis JH, Leon AC, Markowitz JC, Manber R, Arnow B, Klein DN, Thase ME. Patient preference as a moderator of outcome for chronic depression treated with nefazodone, cognitive behavioral analysis system of psychotherapy, or their combination. J Clin Psychiatry 2009;70:354–361 [DOI] [PubMed] [Google Scholar]
- 42.McHugh RK, Whitton SW, Peckham AD, Welge JA, Otto MW. Patient preference for psychological vs pharmacologic treatment of psychiatric disorders: a meta-analytic review. J Clin Psychiatry 2013;74:595–602 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Shafran R, Clark DM, Fairburn CG, Arntz A., Barlow DH, Ehlers A, … Wilson G. Mind the gap: Improving the dissemination of CBT. Behav Res Ther 2009;47:902–909. [DOI] [PubMed] [Google Scholar]
- 44.Bergkvist L Appropriate use of single-item measures is here to stay. Market Lett 2015;26:245–255. [Google Scholar]
- 45.Brown JD, Buscemi J, Milson V, Malcolm R, O’Neil PM Effects on cardiovascular risk factors of weight losses limited to 5–10%. Transl Behav Med 2016;6:339–346. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
De-identified data will be provided in response to reasonable written request to achieve goals in an approved written proposal.