Abstract
Objective:
Study aims were to: (1) document the relationship between a history of childhood abuse and weight change during behavioral obesity treatment and (2) estimate the indirect effect of childhood abuse on weight change through binge eating severity.
Methods:
Participants (N=431) were enrolled in a behavioral weight loss intervention. Childhood physical and emotional abuse history and current binge eating severity were self-reported. Percent weight loss at 6 months was calculated using measured weight. Adjusted mediation models examined whether there was an indirect effect of childhood physical and emotional abuse on 6-month percent weight loss that operated through binge eating severity.
Results:
After covariate adjustment, childhood physical abuse, but not emotional abuse, predicted a lower percent weight loss (B=−1.78%; 95% CI: [−3.10, −0.47]). Although childhood physical and emotional abuse were positively related to baseline binge eating severity, binge eating severity did not mediate the associations between either childhood abuse type and percent weight loss.
Conclusions:
Individuals with a childhood physical abuse history had a lower weight loss percentage than those without such histories during behavioral obesity treatment. This effect was not explained by binge eating severity. Individuals with a history of childhood abuse may benefit from trauma-informed obesity care.
Keywords: Physical Abuse, Emotional Abuse, Childhood Abuse, Obesity, Binge Eating
Introduction
Exposure to adverse childhood experiences, including childhood abuse, is prevalent in the United States1 and has lasting effects on adult morbidity and mortality.2 An extensive body of literature has identified childhood abuse as an indicator of adult obesity risk,3-7 with longitudinal evidence showing that childhood abuse predicts elevated weight status across the lifecourse3,5 and is associated with risk for incident severe obesity in adulthood.4 Meta-analytic findings further document that adults with a childhood abuse history have between 1.23-1.43 times the odds of having obesity as those without such a history.6,7 Although numerous mechanisms have been proposed to explain how childhood abuse contributes to the development and progression of obesity, including disruptions to inflammatory and metabolic functioning,8 high rates of psychopathology,5 and greater engagement in binge and emotional eating,9 understanding of these mechanisms remains limited. It also remains unknown whether a childhood abuse history impacts weight loss during behavioral obesity treatment. As such, there remains a need to clarify whether and how childhood abuse relates to behavioral weight loss outcomes.
Most research examining the relationship between childhood abuse and obesity treatment has focused on bariatric surgery. Despite some evidence indicating that exposure to adverse childhood experiences, including abuse, is associated with less weight loss following bariatric surgery,10 these findings have been mixed, with additional research showing no relationship between adverse childhood experiences and post-operative weight loss or regain.11
Few efforts have extended this work to examine the association between childhood abuse and behavioral obesity treatment outcomes. Research has broadly linked trauma exposure and post-traumatic stress symptoms to greater weight loss barriers12 and less weight loss during behavioral obesity treatment.13 However, previous research has not focused on the specific impact of childhood abuse on behavioral obesity outcomes. Given that childhood abuse is a highly prevalent type of childhood trauma,1 with many survivors experiencing post-traumatic stress symptoms, it is possible that individuals with a childhood abuse history may respond suboptimally to behavioral obesity interventions.
One factor that may impact behavioral weight loss outcomes among individuals with a childhood abuse history is binge eating. Binge eating is characterized by the consumption of a large amount of food accompanied by a perceived sense of loss of control over eating14 and occurs with increased prevalence among individuals with a childhood abuse history.15 Binge eating is theorized to develop among those with a childhood abuse history due to poor emotion regulation abilities stemming from the negative impact of abuse on childhood development.16 For example, children exposed to abuse may not have the opportunity to develop adaptive coping strategies, and may instead rely on behaviors, like binge eating, that aid in avoidance or emotional dampening to manage the psychological distress that accompanies exposure to traumatic events.17 Without intervention, avoidant coping behaviors may persist into adulthood, and in the case of binge eating, contribute to weight gain over time.18 Binge eating has also been associated with less weight loss during behavioral obesity treatment.19 For example, recent findings from the Look AHEAD Trial showed that participants in an intensive lifestyle intervention who engaged in binge eating lost half as much weight compared to those who did not binge eat.20 Because individuals with a childhood abuse history are at elevated risk of binge eating, they may have greater difficulties losing weight during behavioral obesity treatment as a result.
This study examined the interrelations among childhood abuse, binge eating, and weight loss outcomes in the BestFIT trial, a behavioral obesity intervention focused on identifying optimal treatment sequences. This study aimed to: (1) document the relationship between a childhood abuse history and weight change during behavioral obesity treatment and (2) estimate the indirect effect of childhood abuse on weight change through binge eating severity. These aims were investigated for childhood emotional abuse, childhood physical abuse, and any experience of childhood abuse (i.e., childhood emotional and/or physical abuse). It was hypothesized that individuals with any history of childhood abuse would have a lower percent weight loss during treatment compared to those without a childhood abuse history, and that this relationship would be mediated by greater binge eating severity among those with childhood abuse.
Methods
Participants and procedures
Data for this secondary analysis were from the BestFIT behavioral obesity trial, the protocol for which has been previously described.21 Briefly, BestFIT was a two-stage sequential multiple assignment randomized trial (SMART) designed to ascertain the optimal time after initiation of standard behavioral weight loss treatment (SBT) to identify suboptimal responders, and to examine whether it is better for suboptimal responders to switch to portion-controlled meals (PCM) or acceptance-based treatment (ABT). Following the baseline assessment, all participants were enrolled in SBT and randomized to complete a response assessment at either Week 3 or Week 7 of treatment. Participants randomized to complete a response assessment at Week 3 were considered suboptimal responders if they had lost less than 2.5% of their starting body weight. Participants randomized to complete a response assessment at Week 7 were considered suboptimal responders if they had lost less than 5.0% of their starting body weight. Participants identified as suboptimal responders at either time point were re-randomized with equal probability to one of two second-stage treatments: (1) augmentation of SBT with PCM or (2) an enhanced acceptance-based behavioral version of SBT (ABT). Participants identified as responders continued with SBT without re-randomization. Regardless of treatment arm, all participants were offered a total of 20 treatment sessions delivered in-person over an approximately 20-week period. Participants completed outcome measures at baseline and at 6-, 8-, and 18-months following the baseline assessment. The present study includes data from the baseline and 6-month assessments.
Participants included 468 community adults from the Minneapolis-St. Paul, Minnesota metropolitan and were recruited through radio, print, web-based advertisements, and direct mailings. Participants were eligible if they: (1) were 21-70 years old; (2) had a body mass index (BMI) ≥30.0 and ≤45 kg/m2; and (3) indicated being willing and able to participate in the study for 18 months. Participants were ineligible if they: (1) were unable to engage in physical activity; (2) were pregnant, breastfeeding, or planning a pregnancy within the next 18 months; (3) were involved in another diet intervention study or organized weight loss program; (4) had dietary restrictions (e.g., gluten-free); (5) had diabetes and were insulin-dependent; or (6) reported serious mental illness (e.g., schizophrenia, psychosis, or bipolar disorder) or had experienced a “nervous breakdown” that could interfere with participation. Written informed consent was obtained for each enrolled participant, and a baseline assessment was completed at HealthPartners Institute. Both HealthPartners and the University of Minnesota Institutional Review Boards approved the research protocol. Primary findings from the BestFIT trial showed that 44% of participants had a 10% weight loss or more, though no significant differences in total weight loss between 3-week and 7-week responders or PCM and ABT were found at 6 months.22
Measures
Childhood abuse.
Childhood abuse was measured by four questions adapted from the Revised Conflict Tactics Scale (CTS2), a widely-used, valid and reliable measure of family violence.23 All items referred to the participant’s first 18 years of life and were rated using a 5-point Likert scale, with response options being never, rarely, sometimes, often, or very often. Decision rules for categorizing the childhood abuse items were in line with previous research.24,25 Participants were coded as having experienced childhood emotional abuse if they responded often or very often to either of the following two questions: “How often did an adult in your family yell and scream at you?” and “How often did an adult in your family say hurtful things to you?” Participants were coded as having experienced childhood physical abuse if they had been “pushed, grabbed, or shoved” by an adult in their family often or very often or were “hit so hard it left bruises or marks” by an adult in their family rarely or more. An overall childhood abuse variable was created by dichotomizing participants who experienced “any” (≥1 experience of childhood emotional or physical abuse) versus “no” childhood abuse. Sexual abuse was not assessed due to concerns that it would be perceived as overly intrusive to participants.
Binge eating severity.
The Binge Eating Scale (BES) is a 16-item, self-report measure of binge eating severity26 that was administered at the baseline and 6-month assessments. The BES measures key behavioral (e.g., eating rapidly), affective (e.g., guilt about overeating), and cognitive (e.g., preoccupation with food) symptoms related to binge eating. For each item, participants selected from a statement best describing their experiences. Items were summed, creating a total score ranging from 0-46, with higher scores indicating greater binge eating severity (Cronbach’s α = 0.87).
Percent weight loss.
Weight change from baseline to 6 months was calculated by subtracting staff-measured weight at the baseline assessment from staff-measured weight at the 6-month assessment. This value was divided by baseline weight and multiplied by 100 to calculate percent of total body weight lost at 6 months.
Because some participants were missing the 6-month assessment (n=53), missing 6-month weights were imputed to retain the full sample. A multiply imputed dataset (m=20) was constructed to carry out sensitivity analyses and assess model performance.27 The imputation model assumed missing at random conditional on covariates, which included all analysis covariates and the auxiliary variables education, coach and intervention completion.28 A Markov chain Monto Carlo imputation model was used to achieve monotone missingness followed by fully conditional specification to impute the remaining missing values. A more detailed description of participant flow through the 6-month assessment has been described elsewhere.29
Covariates.
Covariates were informed by the Accumulating Data to Optimally Predict Obesity Treatment (ADOPT) framework.30 Sociodemographic covariates included self-reported age, gender (male versus female), race (non-Hispanic white versus other racial groups), and annual household income (<$75,000 versus ≥$75,000). Baseline BMI (kg/m2) was calculated using staff-measured height and weight collected at the baseline assessment. To covary for treatment effects, dummy codes were created for each randomization schedule and included: (1) 3-week non-responders randomized to ABT; (2) 3-week non-responders randomized to PCM; (3) 3-week responders; (4) 7-week non-responders randomized to ABT; (5) 7-week non-responders randomized to PCM; and (6) 7-week responders.
Statistical Analysis
Analyses were performed using SPSS Version 26.0 (IBM, Armonk, NY). Participants with missing childhood abuse data (n=11) and those who were missing a treatment assignment because they dropped out of the trial prior to their treatment response assessment (n=26) were excluded from analysis. The final analytic sample included 431 participants. Sample characteristics were calculated, and differences were compared between participants who reported any versus no childhood abuse using independent samples t-tests for continuous variables and chi-square tests for categorical variables.
Multiple linear regression models were estimated to examine the total effect of childhood abuse on percent weight loss at 6 months. Crude and adjusted regression models were run for each childhood abuse predictor (i.e., any childhood abuse, childhood physical abuse, and childhood emotional abuse).
Mediation models testing the indirect effect of childhood abuse on percent weight loss at 6 months through binge eating severity were computed using the PROCESS macro for SPSS developed by Hayes.31 For each model, the bootstrap sample was set to 5,000 and the confidence interval to 95%. Crude and adjusted mediation models were calculated for each childhood abuse predictor (i.e., any childhood abuse, childhood physical abuse, and childhood emotional abuse). Models initially included the mediating effect of baseline binge eating severity. Because participation in behavioral obesity interventions can reduce binge eating in the short term,32 additional analyses were conducted exploring the mediating effect of change in binge eating severity from baseline to 6 months in the sample of participants who completed the BES at 6 months (n=411). For this analysis, change in binge eating was calculated by subtracting BES scores at 6 months from those collected at baseline.
All adjusted models included the effects of age, gender, income, baseline BMI, and treatment assignment. Continuous predictors were grand-mean centered to facilitate interpretation.
Results
Prevalence of childhood abuse
Nearly one-third of the sample reported any history of childhood abuse (31.3%, n=135). Of those indicating a childhood abuse history, 80.0% (n=108) reported experiencing childhood emotional abuse and 64.4% (n=87) reported experiencing childhood physical abuse. As shown in Table 1, participants with any history of childhood abuse were more likely to be younger, female, belong to a racial group other than non-Hispanic white, and have an annual household income less than $75,000 per year compared to those without such a history. Any history of childhood abuse was also associated with greater baseline binge eating severity as well as lower total weight loss and a lower percent weight loss by 6-month follow-up compared to no such history. A childhood abuse history was not related to change in binge eating severity during treatment.
Table 1.
Sample Characteristics (n = 431)
Total | No Childhood Abuse |
Any Childhood Abuse |
|||||
---|---|---|---|---|---|---|---|
n = 431 | n = 296, 68.7% | n = 135, 31.3% | |||||
Variable | Mean | SD | Mean | SD | Mean | SD | p |
Age | 48.87 | 10.35 | 49.69 | 9.98 | 47.06 | 10.97 | 0.01 |
BMI | 35.94 | 3.89 | 35.82 | 3.84 | 36.21 | 3.96 | 0.33 |
BES | 14.29 | 7.56 | 13.56 | 7.15 | 15.90 | 8.18 | 0.003 |
6-mos BES change | −7.66 | 6.18 | −7.33 | 5.84 | −8.30 | 6.96 | 0.15 |
6-mos total kg lost | −9.62 | 6.69 | −10.24 | 6.79 | −8.26 | 6.29 | 0.004 |
6-mos percent total body weight lost | 9.43 | 6.20 | 9.99 | 6.17 | 8.20 | 6.12 | 0.005 |
% | n | % | n | % | n | p | |
Women | 76.1% | 328 | 72.0% | 213 | 85.2% | 115 | 0.003 |
Non-Hispanic white | 80.5% | 347 | 85.8% | 254 | 68.9% | 93 | <0.001 |
Annual income ≥$75,000 | 63.1% | 272 | 67.6% | 200 | 53.3% | 72 | 0.005 |
Note:All variables were assessed at baseline accept when noted otherwise. Boldfaced data indicate significance (p < 0.05). BES = binge eating severity; BMI = body mass index.
Regression analyses examining the total effect of childhood abuse on 6-month weight change
As shown in Table 2, crude analyses demonstrated that participants with any childhood abuse history lost 1.8% less of their total body weight over 6 months than did those without a childhood abuse history. Additional crude models of specific abuse types showed that participants with a childhood physical abuse history lost 2.6% less of their total body weight than those without a childhood physical abuse history, and those with a childhood emotional abuse history lost 1.6% less of their total body weight than those without a childhood emotional abuse history over 6 months (see Table 3).
Table 2.
Multiple linear regression models examining the total effect of any childhood abuse history on percent total body weight lost at 6 months
Crude Analysis | Adjusted Analysis | |||
---|---|---|---|---|
B | 95% CI | B | 95% CI | |
Any Childhood Abuse | ||||
No (ref) | -- | -- | -- | -- |
Yes | −1.79 | −3.04, −0.53 | −0.66 | −1.81, 0.48 |
Gender | ||||
Men (ref) | -- | -- | -- | -- |
Women | -- | -- | −1.02 | −2.29, 0.24 |
Race | ||||
Other racial group (ref) | -- | -- | -- | -- |
Non-Hispanic white | -- | -- | 0.39 | −1.0, 7.75 |
Annual income | ||||
<$75,000 (ref) | -- | -- | -- | -- |
≥$75,000 | -- | -- | 0.24 | −0.88, 1.36 |
Age | -- | -- | 0.09 | 0.04, 0.14 |
Baseline BMI | -- | -- | −0.08 | −0.21, 0.06 |
Treatment Arms | ||||
3-week ABT | -- | -- | −7.00 | −8.75, −5.25 |
3-week PCM | -- | -- | −1.63 | −2.21, −1.06 |
3-week responders | -- | -- | −0.39 | −0.70, −0.07 |
7-week ABT | -- | -- | −3.30 | −4.20, −2.39 |
7-week PCM | -- | -- | −1.60 | −2.05, −1.16 |
7-week responders (ref) | -- | -- | -- | -- |
Note: All models controlled for the effects of treatment assignment. Boldfaced data indicate significance (p < 0.05). ABT = acceptance-based treatment; BMI = body mass index; CI = confidence interval; PCM = portion-controlled meals
Table 3.
Multiple linear regression models examining the total effect of childhood physical and emotional abuse histories on percent total body weight lost at 6 months
Crude Analysis | Adjusted Analysis | |||
---|---|---|---|---|
Childhood Physical Abuse | B | 95% CI | B | 95% CI |
Childhood Physical Abuse | ||||
No (ref) | -- | -- | -- | -- |
Yes | −2.56 | −4.00, −1.12 | −1.79 | −3.10, −0.47 |
Gender | ||||
Men (ref) | -- | -- | -- | -- |
Women | -- | -- | −1.09 | −2.33, 0.15 |
Race | ||||
Other racial group (ref) | -- | -- | -- | -- |
Non-Hispanic white | -- | -- | 0.13 | −1.24, 1.49 |
Annual income | ||||
<$75,000 (ref) | -- | -- | -- | -- |
≥$75,000 | -- | -- | 0.11 | −1.00, 1.22 |
Age | -- | -- | 0.09 | 0.04, 0.14 |
Baseline BMI | -- | -- | −0.08 | −0.21, 0.06 |
Treatment Arms | ||||
3-week ABT | -- | -- | −6.94 | −8.67, −5.22 |
3-week PCM | -- | -- | −1.54 | −2.11, −0.96 |
3-week responders | -- | -- | −0.36 | −0.68, −0.05 |
7-week ABT | -- | -- | −3.29 | −4.18, −2.39 |
7-week PCM | -- | -- | −1.55 | −1.99, −1.11 |
7-week responders (ref) | -- | -- | -- | -- |
Childhood Emotional Abuse | B | 95% CI | B | 95% CI |
Childhood Emotional Abuse | ||||
No (ref) | -- | -- | -- | -- |
Yes | −1.58 | −2.91, −0.25 | −0.36 | −1.56, 0.84 |
Gender | ||||
Men (ref) | -- | -- | -- | -- |
Women | -- | -- | −1.00 | −2.27, 0.26 |
Race | ||||
Other racial group (ref) | -- | -- | -- | -- |
Non-Hispanic white | -- | -- | 0.40 | −0.93, 1.73 |
Annual income | ||||
<$75,000 (ref) | -- | -- | -- | -- |
≥$75,000 | -- | -- | 0.35 | −0.93, 1.73 |
Age | -- | -- | 0.09 | 0.04, 0.14 |
Baseline BMI | -- | -- | −0.08 | −0.22, 0.05 |
Treatment Arms | ||||
3-week ABT | -- | -- | −7.00 | −8.72, −5.27 |
3-week PCM | -- | -- | −1.63 | −2.20, −1.06 |
3-week responders | -- | -- | −0.41 | −0.73, −0.10 |
7-week ABT | -- | -- | −3.35 | −4.25, −2.44 |
7-week PCM | -- | -- | −1.62 | −2.07, −1.18 |
7-week responders (ref) | -- | -- | -- | -- |
Note: All models controlled for the effects of treatment assignment. Boldfaced data indicate significance (p < 0.05). ABT = acceptance-based treatment; BMI = body mass index; CI = confidence interval; PCM = portion-controlled meals
After covariate adjustment, participants with a childhood physical abuse history lost 1.8% less of their total body weight over 6 months than did those without such a history. Examination of adjusted mean differences showed that participants with a childhood physical abuse history had an average percent weight loss of 8.1% compared to 9.8% among those without a childhood physical abuse history. Estimated effects for childhood emotional abuse and any childhood abuse were attenuated and close to the null after adjustment.
Age was positively associated with percent weight loss in all models. Compared to 7-week responders, participants in all other treatment arms had a lower percent weight loss. Estimated effects for all other covariates included the null.
Mediation analyses testing the indirect effect of childhood abuse on 6-month weight change through binge eating severity
As shown in Table 4, all childhood abuse variables were positively and significantly associated with baseline binge eating severity in both crude and adjusted models. However, baseline binge eating severity was unrelated to 6-month percent weight change in all crude and adjusted models. In addition, no significant indirect effects of any childhood abuse variable on 6-month percent weight change through baseline binge eating severity were found in either crude or adjusted models.
Table 4.
Mediation models testing the indirect effect of childhood abuse on percent total body weight lost at 6 months through binge eating severity
Crude Analysis | Adjusted Analysis | |||
---|---|---|---|---|
Any Childhood Abuse | B | 95% CI | B | 95% CI |
Component Effects | ||||
Any Abuse → BES (path a) | 2.34 | 0.82, 3.87 | 2.09 | 0.54, 3.64 |
BES → 6 mos weight change (path b) | −0.05 | −0.13, 0.02 | −0.63 | −0.09, 0.05 |
Direct Effect | ||||
Any Abuse → 6 mos weight change (path c') | −1.66 | −2.93, −0.39 | −0.63 | −1.78, 0.53 |
Indirect Effect | ||||
Any Abuse → BES → 6 mos weight change (path ab) | −0.13 | −0.34, 0.04 | −0.04 | −0.19, 0.10 |
Childhood Physical Abuse | B | 95% CI | B | 95% CI |
Component Effects | ||||
Physical Abuse → BES (path a) | 1.82 | 0.04, 3.59 | 2.13 | 0.33, 3.93 |
BES → 6 mos weight change (path b) | −0.06 | −0.13, .0.02 | −0.01 | −0.08, 0.06 |
Direct Effect | ||||
Physical Abuse → 6 mos weight change (path c') | −2.47 | −3.92, −1.02 | −1.76 | −3.08, −0.43 |
Indirect Effect | ||||
Physical Abuse → BES → 6 mos weight change (path ab) | −0.10 | −0.33, 0.04 | −0.02 | −0.18, 0.14 |
Childhood Emotional Abuse | B | 95% CI | B | 95% CI |
Component Effects | ||||
Emotional Abuse → BES (path a) | 2.39 | 0.75, 4.02 | 1.90 | 0.25, 3.54 |
BES → 6 mos weight change (path b) | −0.06 | −0.14, 0.02 | −0.02 | −0.09, 0.05 |
Direct Effect | ||||
Emotional Abuse → 6 mos weight change (path c') | −1.44 | −2.80, −0.08 | −0.34 | −1.57, 0.88 |
Indirect Effect | ||||
Emotional Abuse → BES → 6 mos weight change (path ab) | −0.14 | −0.36, 0.04 | −0.04 | −0.19, 0.09 |
Note: All models controlled for effects of gender, race, annual income, age, baseline body mass index, and treatment assignment. Boldfaced data indicate significance (p < 0.05). BES = binge eating severity; CI = confidence interval
Analyses detailing the results of 6-month change in binge eating severity as a mediator between childhood abuse and 6-month percent weight change are shown in Supplemental Table 1. Unlike the previous mediation models, none of the childhood abuse variables related to 6-month change in binge eating severity in either crude or adjusted models. Moreover, 6-month change in binge eating severity negatively related to 6-month percent weight change in both crude and adjusted analyses, indicating that individuals with a greater reduction in binge eating severity had a higher percent weight loss during treatment. As with the previous mediation models, no significant indirect effects of any childhood abuse variable on 6-month percent weight change through 6-month change in binge eating severity were found in either crude or adjusted models.
Discussion
To our knowledge, this study is the first to examine whether childhood abuse was associated with weight change during a behavioral weight loss intervention, and if binge eating, a specific behavior known to be associated with both childhood abuse and obesity, mediated this relationship. Similar to community samples of adults,1 nearly 1 in 3 participants in this study reported a childhood abuse history, with 25% reporting childhood emotional abuse and 21% reporting childhood physical abuse. Results indicated that childhood physical, but not emotional, abuse was a meaningful predictor of a lower weight loss percentage during intervention, even after adjustment for relevant covariates. Although both childhood physical and emotional abuse were related to higher binge eating severity at baseline, neither of these abuse experiences related to change in binge eating severity during treatment. In addition, neither baseline nor change in binge eating severity mediated the association between childhood physical or emotional abuse and percent weight loss during treatment. Taken together, these findings indicate that individuals with a childhood physical abuse history may benefit from additional support during behavioral weight loss intervention.
In this study, participants with a childhood physical abuse history achieved a clinically meaningful 8% reduction in total body weight. However, this outcome remained significantly lower than the nearly 10% reduction in total body weight observed among participants without a childhood physical abuse history. Interestingly, this effect was not observed among participants exposed to childhood emotional abuse. Although crude analyses showed that exposure to childhood emotional abuse predicted a lower weight loss percentage during intervention, this effect attenuated and was no longer significant in adjusted models. This finding suggests that childhood emotional abuse may be related to behavioral weight loss outcomes but does not account for unique variance after taking covariate effects into account. In line with these results, previous studies have found that a physical abuse history is a stronger predictor of obesity risk3 and poorer weight loss outcomes following bariatric surgery33 when compared to emotional abuse, indicating that the experience of physical abuse may uniquely impact mechanisms related to the development and persistence of obesity, such as through dysregulations to the hypothalamic-pituitary-adrenal axis34 or greater symptoms of depression, anxiety, and posttraumatic stress.35
Contrary to expectations, the relationship between childhood abuse and weight loss was not explained by greater baseline binge eating severity or change in binge eating severity across treatment. Both childhood physical and emotional abuse positively related to baseline binge eating severity, as was expected based on prior research.15 However, neither of these abuse experiences related to change in binge eating severity during treatment, suggesting that individuals exposed to childhood physical or emotional abuse had similar reductions in binge eating severity as did those without such experiences. Furthermore, baseline binge eating severity did not predict weight loss outcomes during treatment, though additional analyses showed that a greater reduction in binge eating severity over the course of treatment was linked to a higher weight loss percentage. This latter finding is consistent with previous reports demonstrating that participation in behavioral obesity treatment can reduce binge eating in the short term to improve weight loss outcomes.32 Taken together, these findings suggest that, although heightened binge eating severity remains a serious public health concern among individuals with a childhood abuse history, it does not appear to impact their attempts at behavioral weight loss. Moreover, behavioral obesity treatment may help to reduce binge eating severity among individuals with and without a childhood abuse history, leading to a higher percent weight loss.
These findings have important clinical implications. The clinically significant weight loss observed among individuals exposed to childhood abuse suggests that the behavioral weight loss approach taken in this trial may be useful for such individuals. Although this study was not specifically designed to identify efficacious behavioral weight loss approaches for individuals with childhood abuse histories, it provides potential directions for future studies. In particular, individuals exposed to childhood abuse may benefit from frequent monitoring during behavioral weight loss treatment and, if they are found to be suboptimally responding, being offered PCM or ABT to further assist in their weight loss efforts. Given previous work showing that integrating behavioral obesity and psychological treatments, as was done for the ABT treatment arm, can improve weight loss outcomes among those with psychiatric symptoms,36 future work is needed to systematically compare the potential benefits of using each of these strategies for individuals with childhood abuse histories in real-world settings.
Although individuals exposed to childhood abuse appear to achieve clinically meaningful weight loss during behavioral obesity treatment, it is worth noting that such individuals, particularly those exposed to childhood physical abuse, had a lower weight loss percentage than did those without childhood abuse histories. Although trauma-related symptoms decrease during standard behavioral obesity treatments,37 individuals with childhood abuse histories may benefit from trauma-informed obesity treatment approaches to further improve their outcomes.38 Trauma-informed care is focused on understanding, recognizing, and responding to the effects of trauma to improve patient outcomes and has been widely implemented in a variety of treatment settings.39,40 However, limited efforts have attempted to extend trauma-informed approaches to behavioral weight loss interventions. A recent proof-of-concept study by Hoerster and colleagues41 supported the feasibility and potential effectiveness of delivering trauma-informed obesity care. The authors adapted an established behavioral weight loss program to include cognitive behavioral therapy skills relevant for trauma- and stressor-related symptoms and found that these adaptations led to significant weight loss among a sample of Veterans with posttraumatic stress disorder. This approach is currently being compared to standard treatment.42 Given the longstanding mental, physical, and behavioral health implications of childhood abuse, particularly in relation to obesity, there remains a need to develop tailored behavioral weight loss approaches that address the downstream consequences of childhood abuse exposure to improve weight loss outcomes.
There were some limitations in this study. First, childhood abuse was measured retrospectively, which may have contributed to recall bias. Childhood abuse measures were also limited to physical and emotional abuse. Although childhood sexual abuse was not assessed due to concerns that it would be perceived as overly intrusive to participants, future work is needed to expand these findings by examining the relationship between childhood sexual abuse and behavioral weight loss outcomes given its established link to adult obesity.4,6,7 Second, the measure of binge eating severity included in this study was assessed through self-report and did not measure clinically significant binge eating episodes. Third, individuals who self-reported serious mental illness (n=15) were excluded and psychiatric diagnoses were not clinically assessed. Given that childhood abuse is associated with increased risk for mental illness, which itself is associated with difficulty losing weight,43 future work is needed to explore the interrelations among childhood abuse, mental illness, and behavioral weight loss outcomes. Fourth, this study used a SMART design to identify suboptimal responders and prescribe additional treatment to standard behavioral weight loss, which may have contributed to the high percent weight loss observed in this study. It will be important for future work to explore how individuals with childhood abuse histories respond to more standard behavioral weight loss approaches. Finally, participants were predominantly comprised of non-Hispanic white women. Because exposure to childhood abuse is disproportionately high in racial groups other than white,44 for whom standard behavioral obesity treatment may be less effective,45 and may also influence eating and weight outcomes differently in men versus women,15 it will be important for future work to extend these findings to more racially and gender diverse samples. Despite these limitations, this study also had several notable strengths, including a large sample size and longitudinal assessment of weight change and binge eating severity during intervention.
Conclusion
In conclusion, this is the first study to assess whether childhood abuse differentially relates to treatment response during behavioral weight loss intervention. This study identified childhood physical, but not emotional, abuse as a meaningful predictor of weight loss outcomes during behavioral obesity treatment. Although both childhood physical and emotional abuse were related to greater binge eating severity at baseline, they were unrelated to changes in binge eating severity during treatment. In addition, neither baseline binge eating severity nor changes in binge eating severity during treatment indirectly related to weight loss outcomes. Therefore, further research is needed to explore the specific mechanisms that limit weight loss among individuals with a childhood abuse history, particularly those exposed to childhood physical abuse. Nevertheless, as this work continues, it would seem prudent now for a childhood abuse history to be added to the core measures included in the ADOPT framework30 to inform research on obesity treatment and for all weight loss programs to first assess for childhood abuse and then, if appropriate, consider incorporating strategies, such as trauma-informed obesity treatment, into their programs.
Supplementary Material
What is already known about this subject?
Childhood abuse is associated with increased risk for obesity and binge eating in adulthood
Binge eating may relate to reduced weight loss during behavioral obesity treatment
What are the new findings in your manuscript?
Childhood physical abuse, but not emotional abuse, related to a lower percent weight loss during behavioral obesity treatment
Binge eating severity did not mediate the associations of either childhood physical or emotional abuse on percent weight loss
How might your results change the direction of research or the focus of clinical practice?
Childhood physical abuse is a meaningful predictor of weight loss outcomes during behavioral obesity treatment
Weight loss programs may benefit from screening for childhood abuse and consider incorporating strategies, such as trauma-informed obesity treatment, into their programs
Acknowledgments
Participant data from this study will not be made available for sharing.
FUNDING:
This work is supported by a grant from the National Cancer Institute (1R01CA188892, PI: Sherwood). Rebecca Emery Tavernier’s time was supported by the National Center for Advancing Translational Sciences under TL1 TR002493 (PI: Fulkerson) and UL1 TR002494 (PI: Blazar).
Footnotes
DISCLOSURE: The authors declared no conflict of interest
CLINICAL TRIAL REGISTRATION: http://www.clinicaltrials.gov, NCT02368002: BestFIT: A Personalized Weight Loss Program
References
- 1.Merrick MT, Ford DC, Ports KA, Guinn AS. Prevalence of Adverse Childhood Experiences From the 2011-2014 Behavioral Risk Factor Surveillance System in 23 States. JAMA Pediatr. 2018;172(11):1038–1044. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Merrick MT, Ford DC, Ports KA, et al. Vital Signs: Estimated Proportion of Adult Health Problems Attributable to Adverse Childhood Experiences and Implications for Prevention—25 States, 2015-2017. MMWR Morb Mortal Wkly Rep. 2019;68(44):999. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Bentley T, Widom CS. A 30-year follow-up of the effects of child abuse and neglect on obesity in adulthood. Obesity (Silver Spring.). 2009;17(10):1900–1905. [DOI] [PubMed] [Google Scholar]
- 4.Richardson AS, Dietz WH, Gordon-Larsen P. The association between childhood sexual and physical abuse with incident adult severe obesity across 13 years of the National Longitudinal Study of Adolescent Health. Pediatr Obesity. 2014;9(5):351–361. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Francis MM, Nikulina V, Widom CS. A Prospective Examination of the Mechanisms Linking Childhood Physical Abuse to Body Mass Index in Adulthood. Child Maltreat. 2015;20(3):203–213. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Wang Y, Wu B, Yang H, Song X. The effect of childhood abuse on the risk of adult obesity. Ann Clin Psychiatry. 2015;27(3):175–184. [PubMed] [Google Scholar]
- 7.Danese A, Tan M. Childhood maltreatment and obesity: systematic review and meta-analysis. Mol Psychiatry. 2014;19(5):544–554. [DOI] [PubMed] [Google Scholar]
- 8.Li L, Chassan RA, Bruer EH, Gower BA, Shelton RC. Childhood maltreatment increases the risk for visceral obesity. Obesity. 2015;23(8):1625–1632. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Greenfield EA, Marks NF. Violence from parents in childhood and obesity in adulthood: Using food in response to stress as a mediator of risk. Soc Sci Med. 2009;68(5):791–798. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Lodhia NA, Rosas US, Moore M, et al. Do adverse childhood experiences affect surgical weight loss outcomes? J Gastrointest Surg. 2015;19(6):993–998. [DOI] [PubMed] [Google Scholar]
- 11.King WC, Hinerman A, Kalarchian MA, Devlin MJ, Marcus MD, Mitchell JE. The impact of childhood trauma on change in depressive symptoms, eating pathology, and weight after Roux-en-Y gastric bypass. Surg Obes Relat Dis. 2019;15(7):1080–1088. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Klingaman EA, Hoerster KD, Aakre JM, Viverito KM, Medoff DR, Goldberg RW. Veterans with PTSD report more weight loss barriers than Veterans with no mental health disorders. Gen Hosp Psychiatry. 2016;39:1–7. [DOI] [PubMed] [Google Scholar]
- 13.Hoerster KD, Lai Z, Goodrich DE, et al. Weight loss after participation in a national VA weight management program among veterans with or without PTSD. Psychiatr Serv. 2014;65(11):1385–1388. [DOI] [PubMed] [Google Scholar]
- 14.Association AP. Feeding and Eating Disorders: DSM-5® Selections. American Psychiatric Pub; 2015. [Google Scholar]
- 15.Emery RL, Yoon C, Mason SM, Neumark-Sztainer D. Childhood maltreatment and disordered eating attitudes and behaviors in adult men and women:Findings from project EAT. Appetite. 23 2021;163:105224. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Burns EE, Fischer S, Jackson JL, Harding HG. Deficits in emotion regulation mediate the relationship between childhood abuse and later eating disorder symptoms. Child Abuse Negl. 2012;36(1):32–39. [DOI] [PubMed] [Google Scholar]
- 17.Cloitre M, Cohen LR, Koenen KC. Treating survivors of childhood abuse: Psychotherapy for the interrupted life. Guilford Press; 2011. [Google Scholar]
- 18.Sonneville KR, Horton NJ, Micali N, et al. Longitudinal associations between binge eating and overeating and adverse outcomes among adolescents and young adults: does loss of control matter? JAMA Pediatr. 2013;167(2):149–155. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Masheb RM, Lutes LD, Kim HM, et al. High-frequency binge eating predicts weight gain among veterans receiving behavioral weight loss treatments. Obesity (Silver Spring). 2015;23(1):54–61. [DOI] [PubMed] [Google Scholar]
- 20.Chao AM, Wadden TA, Gorin AA, et al. Binge Eating and Weight Loss Outcomes in Individuals with Type 2 Diabetes: 4-Year Results from the Look AHEAD Study. Obesity (Silver Spring.). 2017;25(11):1830–1837. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Sherwood NE, Butryn ML, Forman EM, et al. The BestFIT trial: A SMART approach to developing individualized weight loss treatments. Contemp Clin Trials. 2016;47:209–216. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Sherwood NE, Crain AL, Seburg EM, et al. BestFIT Sequential Multiple Assignment Randomized Trial Results: A SMART Approach to Developing Individualized Weight Loss Treatment Sequences. Annals Behav Med. 2021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Straus MA, Hamby SL, Boney-McCoy S, Sugarman DB. The revised Conflict Tactics Scales (CTS2): Development and preliminary psychometric data. J Fam Issues. 1996;17(3):283–316. [Google Scholar]
- 24.Chapman DP, Whitfield CL, Felitti VJ, Dube SR, Edwards VJ, Anda RF. Adverse childhood experiences and the risk of depressive disorders in adulthood. J Affect Dis. 2004;82(2):217–225. [DOI] [PubMed] [Google Scholar]
- 25.Felitti VJ, Anda RF, Nordenberg D, et al. Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults. The Adverse Childhood Experiences (ACE) Study. Am J Prev Med. 1998;14(4):245–258. [DOI] [PubMed] [Google Scholar]
- 26.Gormally J, Black S, Daston S, Rardin D. The assessment of binge eating severity among obese persons. Addict Behav. 1982;7(1):47–55. [DOI] [PubMed] [Google Scholar]
- 27.Graham-Bermann SA, Lynch S, Banyard V, Devoe ER, Halabu H. Community-based intervention for children exposed to intimate partner violence: an efficacy trial. J Consult Clin Psychol. 2007;75(2):199–209. [DOI] [PubMed] [Google Scholar]
- 28.Hardt J, Herke M, Leonhart R. Auxiliary variables in multiple imputation in regression with missing X: a warning against including too many in small sample research. BMC Med Res Methodol. 2012;12:184. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Crane MM, Seburg EM, Levy RL, Jeffery RW, Sherwood NE. Using targeting to recruit men and women of color into a behavioral weight loss trial. Trials. 2020;21(1):537. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.MacLean PS, Rothman AJ, Nicastro HL, et al. The Accumulating Data to Optimally Predict Obesity Treatment (ADOPT) Core Measures Project: Rationale and Approach. Obesity (Silver Spring). 2018;26 Suppl 2(Suppl 2):S6–S15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Hayes AF. PROCESS [Macro]. The Ohio State University. 2012. [Google Scholar]
- 32.Munsch S, Biedert E, Meyer A, et al. A randomized comparison of cognitive behavioral therapy and behavioral weight loss treatment for overweight individuals with binge eating disorder. Int J Eat Dis. 2007;40(2):102–113. [DOI] [PubMed] [Google Scholar]
- 33.Wedin S, Madan A, Correll J, et al. Emotional eating, marital status and history of physical abuse predict 2-year weight loss in weight loss surgery patients. Eating Behav. 2014;15(4):619–624. [DOI] [PubMed] [Google Scholar]
- 34.Carpenter LL, Shattuck TT, Tyrka AR, Geracioti TD, Price LH. Effect of childhood physical abuse on cortisol stress response. Psychopharmacology. 2011;214(1):367–375. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Lindert J, von Ehrenstein OS, Grashow R, Gal G, Braehler E, Weisskopf MG. Sexual and physical abuse in childhood is associated with depression and anxiety over the life course: systematic review and meta-analysis. Int J Public Health. 2014;59(2):359–372. [DOI] [PubMed] [Google Scholar]
- 36.Ma J, Rosas LG, Lv N, et al. Effect of Integrated Behavioral Weight Loss Treatment and Problem-Solving Therapy on Body Mass Index and Depressive Symptoms Among Patients With Obesity and Depression: The RAINBOW Randomized Clinical Trial. JAMA. 2019;321(9):869–879. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Johannessen KB, Berntsen D. Losing the symptoms: weight loss and decrease in posttraumatic stress disorder symptoms. J Clin Psychol. 2013;69(6):655–660. [DOI] [PubMed] [Google Scholar]
- 38.Mason SM, Bryn Austin S, Bakalar JL, et al. Child Maltreatment's Heavy Toll: The Need for Trauma-Informed Obesity Prevention. Am J Prev Med. 2016;50(5):646–649. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Roberts SJ, Chandler GE, Kalmakis K. A model for trauma-informed primary care. J Am Assoc Nurse Pract. 2019;31(2). [DOI] [PubMed] [Google Scholar]
- 40.Beckett P, Holmes D, Phipps M, Patton D, Molloy L. Trauma-informed care and practice: Practice improvement strategies in an inpatient mental health ward. J Psychos Nurs Mental Health Serv. 2017;55(10):34–38. [DOI] [PubMed] [Google Scholar]
- 41.Hoerster KD, Tanksley L, Simpson T, et al. Development of a Tailored Behavioral Weight Loss Program for Veterans With PTSD (MOVE!+UP): A Mixed-Methods Uncontrolled Iterative Pilot Study. Am J Health Promot. 2020;34(6):587–598. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Hoerster KD, Tanksley L, Sulayman N, et al. Testing a tailored weight management program for veterans with PTSD: The MOVE! + UP randomized controlled trial. Contemp Clin Trials. 2021;107:106487. [DOI] [PubMed] [Google Scholar]
- 43.Aschbrenner K, Carpenter-Song E, Mueser K, Kinney A, Pratt S, Bartels S. A qualitative study of social facilitators and barriers to health behavior change among persons with serious mental illness. Comm Mental Health J. 2013;49(2):207–212. [DOI] [PubMed] [Google Scholar]
- 44.Maguire-Jack K, Lanier P, Lombardi B. Investigating racial differences in clusters of adverse childhood experiences. Am J Orthopsychiatry. 2020;90(1):106. [DOI] [PubMed] [Google Scholar]
- 45.Fitzgibbon ML, Tussing-Humphreys LM, Porter JS, Martin IK, Odoms-Young A, Sharp LK. Weight loss and African-American women: a systematic review of the behavioural weight loss intervention literature. Obes Rev. 2012;13(3):193–213. [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.