Abstract
Background
PTSD is a risk factor for obesity, but the range of behaviors that contribute to this association are not known.
Purpose
To examine associations between self-reported PTSD symptoms in 2007, with and without comorbid depression symptoms, and three problematic overeating behaviors in 2010; to estimate the associations of PTSD-related overeating behaviors with obesity.
Methods
Cross-sectional and longitudinal analyses included 7438 male (n=2478) and female (n=4960) participants from the Growing Up Today Study (mean age 22–29 years in 2010). Three eating behavior outcomes were assessed: Binge eating (eating a large amount of food in a short period of time with loss of control), top quartile of coping-motivated eating (from the Motivations to Eat scale), and top quartile of disinhibited eating (from the Three-Factor Eating Questionnaire).
Results
PTSD symptoms were associated with 2- to 3-fold increases in binge eating and top-quartile coping-motivated eating; having ≥ 4 PTSD symptoms, relative to no PTSD symptoms, was associated with covariate-adjusted RRs of 2.7 (95% CI: 2.1, 3.4) for binge eating, 2.1 (95% CI: 1.9, 2.4) for the top quartile of coping-motivated eating, and 1.5 (95% CI: 1.3, 1.7) for the top quartile of disinhibited eating. There was a trend toward PTSD symptoms in 2007 predicting new onset binge eating in 2010. Having depression symptoms comorbid with PTSD symptoms further increased risk of binge eating and coping-motivated eating. All eating behaviors were associated with obesity.
Conclusion
Clinicians treating patients with PTSD should know of potential comorbid problematic eating behaviors that may contribute to obesity.
Keywords: posttraumatic stress disorder, depression, obesity, eating behavior, binge eating
INTRODUCTION
Posttraumatic stress disorder (PTSD) affects 6.8% of US civilians at some point in their lives (1), with many more affected by subthreshold symptoms (2). PTSD is increasingly recognized as a risk factor for obesity (3–8) and obesity-related diseases (2; 9–11). The mechanisms driving the PTSD–obesity association are not known, but eating as a way to cope with distress may be a contributing factor. The eating-to-cope hypothesis is supported by neurological studies indicating that ingestion of high-fat and high-sugar foods blunts feelings of distress by triggering dopaminergic reward responses (12; 13). Further, a relatively large literature indicates that PTSD is frequently comorbid with eating disorders relevant to weight gain (e.g. binge eating disorder)(14–17). This literature has generally focused on eating disorder diagnoses, the most severe end of a spectrum of potentially problematic eating. Far less is known about how PTSD symptoms are associated with common, problematic overeating behaviors that do not meet diagnostic thresholds but may nonetheless contribute to obesity risk.
PTSD is frequently comorbid with depression (18), and it is plausible that the combination of PTSD symptoms and depression symptoms is associated with additional weight-related risks (19). Depression has been found to predict obesity in some studies (19; 20), potentially due to its association with disordered eating behaviors (21). There is also evidence that depressive symptoms reduce individuals’ capacity to cope with life stressors (22; 23). This reduced coping capacity may put those with a combination of PTSD and depressive symptoms at uniquely high risk for maladaptive coping such as problematic eating.
A greater understanding of the links of PTSD symptoms, with and without comorbid depression, with a range of prevalent obesogenic eating behaviors is needed to understand PTSD-related comorbidities and highlight potential targets for obesity prevention in PTSD-affected populations. To our knowledge the current literature on this topic is limited to a handful of studies. These include two studies on PTSD and emotional eating (24; 25) (one showing a positive finding, the other null), two studies showing PTSD symptoms to be associated with food addiction (26; 27), and four studies finding greater intake of unhealthy foods and/or lower intake of healthy foods among those with PTSD (28–30). One study of earthquake survivors found that meeting PTSD diagnostic criteria was associated with increased craving for sweets or carbohydrates, as well as both increases and decreases in appetite (31). Other common overeating behavior phenotypes have not been examined. Further, few studies have explicitly examined the combination of PTSD and depression symptoms in relation to eating behavior. Finally, none of the existing studies have been able to examine the longitudinal links between PTSD symptoms and new onset of eating behaviors.
We aimed to address these gaps in the literature using data from the Growing Up Today Study, a community-based longitudinal cohort of over 7000 young men and women. Our study aims were to estimate: (1) the cross-sectional associations between PTSD symptoms and three common types of problematic overeating: sub-threshold binge eating (binge eating at any frequency), coping-motivated eating (32), and disinhibited eating (33); (2) the longitudinal association between PTSD symptoms and subsequent new onset of binge eating among women and men without a prior binge eating history; and (3) the independent and joint impacts of PTSD and depression symptoms on eating behaviors. Finally, to establish the relevance of these findings to the PTSD-obesity association, we aimed to (4) examine the association of each type of problematic eating with obesity. Because of prior literature suggesting that the PTSD-obesity association is stronger for women than men (4–6), we examined modifications of associations of PTSD, eating behaviors, and obesity by gender, and present stratified results where appropriate.
METHODS
Study Population
The Growing Up Today Study (GUTS) comprises children of participants in the Nurses’ Health Study II (NHSII). In 1996, NHSII participants with children aged 9–14 years were asked to provide consent for their children to participate in GUTS; 18,526 mothers (54%) consented, and their children (n=26,765) were invited to participate. Of these, 16,882 (9039 girls and 7843 boys) returned the baseline questionnaire and have been followed through a series of 11 online and mailed questionnaires collecting self-reported demographic, behavioral, and medical data.
We excluded participants from our analytic sample if they did not complete the 2007 GUTS questionnaire on which PTSD symptoms were assessed (n=2870 women and 4150 men), if they had missing or incomplete PTSD data (n=285 women and 304 men), or if they did not complete the 2010 questionnaire on which eating behavior outcomes were ascertained (n=924 women and 911 men). This left 7438 participants for our analysis. Participants were aged 22 to 29 years in 2010 when eating behaviors were assessed.
Variables
Exposures
PTSD symptoms in 2007
The 2007 questionnaire assessed PTSD symptoms using Breslau’s Short Screening Scale (34). Participants were asked to report up to 7 lifetime PTSD symptoms, including intrusive thoughts, emotional numbness, and hypervigilance, in reference to the “most distressing” event that they had experienced. Participants were not asked to specify the event they found most distressing. For each PTSD symptom endorsed, participants were asked to report how frequently they had experienced that symptom in the past 4 weeks (5-point scale from “none of the time” to “most or all of the time”). We defined participants as having a current PTSD symptom if they reported experiencing it with any frequency in the past 4 weeks. These symptoms do not designate a clinical PTSD diagnosis; however, self-reported symptoms assessed with this screener have been found to predict important health outcomes such as increases in weight status and type 2 diabetes (3; 35). Breslau et al. recommend ≥4 symptoms as the cut-off for significant PTSD symptoms (34), reporting 80% sensitivity and 97% specificity for this cut-off (34). Preliminary analyses indicated that additional symptoms beyond this cut-off did not confer additional risk for the eating behavior outcomes, and we therefore categorized PTSD symptoms in three levels: 0, 1–3, and ≥4 symptoms.
Joint PTSD and depressive symptoms in 2007
PTSD and depression are frequently comorbid, and depression has been associated with obesity in some, though not all, studies (20; 36). We were therefore interested in examining whether the combination of PTSD and depression symptoms conferred additional risk of problematic eating behaviors. The 2007 questionnaire included the short form of the Centers for Epidemiologic Studies-Depression (CESD-10) scale (37), which assesses the frequency of depressive symptoms over the past week. A score of 10 or more is considered indicative of significant depressive symptoms (37), with research showing 100% sensitivity and 93% specificity for major depression in a community-based sample (38). To examine joint PTSD and depressive symptoms, we dichotomized PTSD symptoms at ≥4 and the CESD-10 score at ≥10 and created a 4-category variable with the following levels: No significant PTSD or depression symptoms (below the cut-offs on both the PTSD screener and the CESD-10 scale), significant depression symptoms only, significant PTSD symptoms only, and both significant PTSD and significant depression symptoms.
Outcomes
We examined the association of 2007 PTSD and depression symptoms with three problematic eating behavior phenotypes, measured in 2010: subthreshold binge eating, coping-motivated eating, and disinhibited eating. In addition, we examined the association of each eating behavior, and PTSD symptoms, with obesity in 2013.
Binge eating in 2010
Binge eating self-report has been collected on each survey over GUTs follow-up. For our main analyses, we used binge eating reported on the 2010 questionnaire (the first questionnaire after the 2007 assessment of PTSD) as our outcome. Binge eating questions asked participants whether, over the past year, they had eaten an extremely large quantity of food over a short period of time (“overeating”), and the frequency with which this had occurred (5-point scale from “Never” to “More than once a week”). Participants who indicated that they had overeaten at any frequency were asked whether they had felt unable to stop eating at that time (“loss of control”). We defined binge eating as overeating with loss of control at any frequency in 2010. This measure captures a relatively common form of problematic eating, as distinct from binge eating disorder (which has specific, severe clinical manifestations). A comparison of this measure with results of clinical interviews assessing binge eating disorder among girls found that the survey measure had a sensitivity of 0.53, specificity of 0.79, and negative predictive value of 0.98.39 These values suggest that whereas the measure is not optimized to capture the small number of individuals with binge eating disorder, it correctly separates most respondents who do not exhibit binge eating symptoms from those who do (i.e., high negative predictive value).
Binge eating was also assessed in 2007. In our longitudinal analysis of PTSD and incident binge eating (described in the analysis section), we excluded participants with prevalent binge eating behavior reported in 2007 (the time of PTSD assessment) so that we could establish the temporal order from PTSD symptoms to binge onset.
Coping-motivated eating in 2010
The coping-motivated eating subscale of the Motivations to Eat Scale (32) was included on the 2010 questionnaire. Jackson et al. report an internal consistency reliability of α=0.88 and have found the subscale scores to be positively correlated with binge eating (32). Items in this 5-item subscale ascertain frequency (never or almost never, rarely, sometimes, often, almost always or always) with which participants eat because they feel depressed, sad, or worthless; to cope; to comfort themselves; or to distract themselves. We examined coping-motivated eating both as a continuous outcome and dichotomized at the gender-specific 75th percentile. In preliminary analyses, we found that those with scores below the 75th percentile were normal weight on average, whereas those above were overweight, suggesting that this is a substantively relevant cut-point.
Disinhibited eating in 2010
The 2010 questionnaire included the disinhibited eating subscale of the Three Factor Eating Questionnaire (33). This 10-item subscale measures the tendency to overeat in response to certain stimuli, such as during social occasions, when with someone who is overeating, in the presence of palatable foods, and as a reaction to dieting. In a recent study, reliability of this subscale ranged from 0.78 in normal-weight individuals to 0.80 in obese individuals (40). As with the coping-motivated eating scale, we examined disinhibited eating both as a continuous measure and dichotomized it at the gender-specific 75th percentile. Similar to coping-motivated eating, we found that the 75th percentile cut-point roughly aligned with the point at which the average body weight changed from normal-weight to overweight.
Obesity in 2013. Obesity was defined as a body mass index (BMI) of >30 kg/m2 based on self-reported height and weight in 2013. Prior work has shown that self-reported BMI is closely correlated with measured BMI (41; 42).
Covariates
For our main adjusted analyses, we included potential confounders of the association between PTSD and disordered eating, including age at baseline (continuous), nonwhite race (dichotomous), perceived social status (participants were asked to indicate their family’s status in American society, from 1, best off, to 10, worst off; continuous), and participant BMI percentile at 1996 study baseline standardized to CDC growth charts (43) (continuous with squared term). We ran subsequent models additionally adjusting for other health behavior correlates of eating behaviors that may be downstream from PTSD, including participant smoking history in 2007 (ever/never), history of dieting (ever/never) in 2005 (the most recent data available prior to 2010 outcomes), and BMI in 2007. In sensitivity analyses, we adjusted for participants’ mothers’ BMI at age 18. Maternal BMI is a strong predictor of child BMI, and potentially a predictor of child eating behaviors (via maternal eating behaviors); it may also be a marker of risk factors (e.g., low socioeconomic status, intergenerational maltreatment) for child exposure to trauma and/or propensity toward PTSD. (Mothers are Nurses’ Health Study II participants; weight at age 18 was asked on the NHSII baseline survey and validated against nursing school entry physical exams (44)).
Data analysis
Our analytic sample included 7438 participants (4960 women and 2478 men) with non-missing PTSD symptom data who had completed the 2010 questionnaire on which the eating behavior outcomes of interest were ascertained. For analyses of each outcome, we excluded the following number of participants because they were missing the outcome variable: 124 women and 122 men missing binge eating, 307 women and 322 men missing coping-motivated eating, and 304 women and 321 men missing disinhibited eating.
Association of PTSD symptoms with eating behaviors
We used modified Poisson regression (45) to estimate risk ratios for each eating behavior outcome in 2010 as a function of the number of reported PTSD symptoms experienced in 2007. We then assessed the longitudinal nature of the relationship between PTSD and binge eating by running models of incident binge eating in 2010 as a function of the number of current PTSD symptoms reported in 2007, among participants with no history of binge eating prior to or in 2007 (when PTSD was assessed; n=1645 women and 303 men). Finally, we examined eating behaviors as a function of joint exposure to significant PTSD symptoms and/or significant depression symptoms.
For each analysis above, we ran a series of models in the following sequence: crude models, addition of main sociodemographic covariates and baseline BMI percentile, and addition of adult health behaviors and 2007 body mass index. We tested for interactions between PTSD symptoms and gender in predicting each outcome. No interactions were statistically significant (Wald test p-values for gender interactions with binge eating, continuous coping-motivated eating, and continuous disinhibited eating were 0.67, 0.21, and 0.30, respectively). We therefore examined these associations in the combined sample of men and women.
We ran three sensitivity analyses of PTSD symptoms and eating behaviors to assess the sensitivity of our results to various modeling decisions. First, we reran PTSD–eating behavior models with additional adjustment for maternal BMI at age 18 years, as a potential shared risk factor for PTSD symptoms and eating behaviors. Second, to assess the impact of attrition out of the sample and related selection bias, we reran PTSD–eating behavior models using stabilized inverse probability weights to adjust for differential non-response to the 2007 and 2010 questionnaires, on which our exposure and outcome were measured. Third, in an attempt to assess whether results were likely to be driven by PTSD-related medication use we reran PTSD–eating behavior models after excluding those who reported past-year use of psychotropic medications (antidepressants or antianxiety medications) on the 2013 questionnaire, the only questionnaire to assess use of these medications.
Associations of eating behaviors and PTSD symptoms with obesity
To examine the extent to which our findings on PTSD and eating behavior would have implications for weight-related health, we examined the association of each eating behavior outcome with obesity. We also estimated risk ratios for obesity in 2013 as a function of 2007 PTSD symptoms. We used modified Poisson regression (45) to estimate risk ratios for obesity in 2013 as a function of (in separate models) binge eating, top quartile of coping-motivated eating, top quartile of disinhibited eating, and categorical PTSD symptoms. Continuous coping-motivated and disinhibited eating were not linearly associated with obesity, and methods to relax the linearity assumption such as inclusion of squared terms or splines make interpretation difficult; we therefore include results for these eating behaviors dichotomized at the top quartile. For each exposure (eating behavior or PTSD symptoms), we ran covariate-adjusted models and an additional model adjusted for baseline BMI. We tested for interactions between gender and each exposure. We found that interactions were significant for gender with PTSD symptoms (Wald p-value for interaction=0.03), but not for gender with eating behaviors. However, for consistency, we present all obesity results stratified by gender.
Results
Of 7438 participants in the analytic sample, 13% reported ≥ 4 PTSD symptoms in 2007. Participants with and without PTSD symptoms were similar in age, race, and perceived social status. Participants with PTSD symptoms were more likely than those without to be female, have a history of smoking or dieting, and to have engaged in binge eating, coping-motivated eating, and disinhibited eating in 2010 (Table 1).
Table 1.
0 symptoms (n=3925) | 1–3 symptoms (n=2544) | 4–7 symptoms (n=969) | |
---|---|---|---|
Outcomes and covariates | % or Mean(SD) |
% or Mean(SD) |
% or Mean(SD) |
Binge eating in 2010a | 4.6 | 7.0 | 12.6 |
Top quartile of coping-motivated eatingb | 3.0(3.7) | 4.0(4.1) | 5.5(4.8) |
Top quartile of disinhibited eatingc | 2.8(2.3) | 3.1(2.4) | 3.5(2.6) |
Age at 1996 baseline (years) | 12.1(1.6) | 12.0(1.6) | 12.0(1.6) |
Gender (female) | 62.1 | 70.9 | 73.9 |
Nonwhite race | 3.0 | 3.2 | 3.6 |
BMI percentile at 1996 baselined | 55.2(29.1) | 54.8(30.2) | 55.9(30.9) |
Perceived social status relative to USe | 3.8(1.2) | 3.9(1.2) | 3.9(1.2) |
Participant history of smoking | 50.9 | 57.1 | 62.1 |
Participant history of dieting | 59.3 | 65.4 | 71.2 |
Participants were defined as having binge eating in 2010 if they responded that they had eaten an extremely large quantity of food over a short period of time with any frequency over the past year, and that during that episode they had felt unable to stop eating.
Coping-motivated eating was measured with 5 items ascertaining the frequency (never or almost never, rarely, sometimes, often, almost always or always) with which participants eat because they feel depressed, sad, or worthless; to cope; to comfort themselves; or to distract themselves. The score was dichotomized at the gender-specific 75th percentile.
Disinhibited eating was assessed with 10 items measuring the tendency to overeat in response to certain stimuli, such as during social occasions, when with someone who is overeating, in the presence of palatable foods, and as a reaction to dieting. The score was dichotomized at the gender-specific 75th percentile.
BMI percentile at baseline was computed from self-reported height and weight, standardized to CDC growth charts.
Perceived social status relative to the US was assessed by asking participants where on a scale of 1 (best off) and 10 (worst off) their family fell in American society.
Association of PTSD symptoms with eating behaviors
In main analyses of the association between PTSD symptoms and eating behaviors, we found that the prevalence of all three eating behavior phenotypes increased with the number of PTSD symptoms (Table 2). Associations were generally stronger for binge eating and coping-motivated eating than for disinhibited eating. For example, having ≥ 4 PTSD symptoms, relative to no PTSD symptoms, was associated with main covariate-adjusted RRs of 2.7 (95% CI: 2.1, 3.4) for binge eating, 2.1 (95% CI: 1.9, 2.4) for the top quartile of coping-motivated eating, and 1.5 (95% CI: 1.3, 1.7) for the top quartile of disinhibited eating (Table 2). In sensitivity analyses, adjusting for maternal BMI made little difference in the associations of PTSD symptoms with eating behaviors. Likewise, results were comparable when adjusted for non-response to key questionnaires using inverse probability weights, and when excluding participants taking psychotropic medications.
Table 2.
Eating behavior outcome | Number of PTSD symptoms | Model 1: Crude | Model 2: Adjusted for main covariatesa | Model 3: Adjusted for health behaviors b | ||||
---|---|---|---|---|---|---|---|---|
Dichotomous outcomes | N (cases) | RR | (95% CI) | RR | (95% CI) | RR | (95% CI) | |
Binge eating | ||||||||
0 | 3810 (174) | 1 | – | 1 | – | 1 | – | |
1–3 | 2451 (172) | 1.5 | (1.3, 1.9) | 1.5 | (1.2, 1.8) | 1.4 | (1.2, 1.8) | |
4–7 | 931 (120) | 2.8 | (2.3, 3.5) | 2.7 | (2.1, 3.4) | 2.5 | (2.0, 3.2) | |
Top quartile of coping-motivated eating | ||||||||
0 | 3604 (589) | 1 | – | 1 | – | 1 | – | |
1–3 | 2315 (566) | 1.5 | (1.3, 1.7) | 1.5 | (1.4, 1.7) | 1.5 | (1.4, 1.7) | |
4–7 | 890 (320) | 2.2 | (2.0, 2.5) | 2.1 | (1.9, 2.4) | 2.0 | (1.8, 2.3) | |
Top quartile of disinhibited eating | ||||||||
0 | 3605 (710) | 1 | – | 1 | – | 1 | – | |
1–3 | 2318 (522) | 1.1 | (1.0, 1.3) | 1.2 | (1.1, 1.3) | 1.2 | (1.0, 1.3) | |
4–7 | 890 (253) | 1.4 | (1.3, 1.6) | 1.5 | (1.3, 1.7) | 1.4 | (1.2, 1.6) | |
Continuous outcomes | N (means) | β | (95% CI) | β | (95% CI) | β | (95% CI) | |
Coping-motivated eating (range 0–20) | ||||||||
0 | 3604 (3.0) | 0 | – | 0 | – | 0 | – | |
1–3 | 2315 (4.0) | 1.0 | (0.8, 1.2) | 0.8 | (0.6, 1.0) | 0.7 | (0.5, 1.0) | |
4–7 | 890 (5.5) | 2.4 | (2.1, 2.8) | 2.1 | (1.8, 2.5) | 2.0 | (1.7, 2.4) | |
Disinhibited eating (range 0–10) | ||||||||
0 | 3605 (2.8) | 0 | – | 0 | – | 0 | – | |
1–3 | 2318 (3.1) | 0.4 | (0.2, 0.5) | 0.3 | (0.2, 0.4) | 0.3 | (0.2, 0.4) | |
4–7 | 890 (3.6) | 0.8 | (0.6, 1.0) | 0.7 | (0.5, 0.9) | 0.6 | (0.4, 0.8) |
Model 2 is adjusted for participant age at baseline (continuous), gender, participant body mass index at baseline (standardized to CDC standard growth curves; continuous with a squared term), nonwhite race, and perceived social status (continuous).
Model 3 is adjusted for Model 2 covariates plus participant body mass index in 2007 when PTSD was ascertained (in kg/m2, continuous with a squared term), participant history of dieting (yes or no, through 2005) and participant history of smoking (yes or no, through 2007).
Longitudinal association of PTSD symptoms with incident binge eating
We ran a longitudinal model restricting our sample to participants with no history of binge eating prior to PTSD symptom ascertainment. We found suggestive evidence that PTSD symptoms in 2007 predicted new onset of binge eating in 2010 (Table 3). Those with ≥ 4 PTSD symptoms in 2007 had twice the risk of reporting binge eating for the first time in 2010, (main covariate-adjusted RR=2.2; 95% CI: 1.1, 4.3). Additional adjustment for maternal BMI attenuated this RR to 1.9 (95% CI: 0.9, 3.9).
Table 3.
Number of PTSD symptoms | N (cases) | Model 1: Crude | Model 2: Adjusted for main covariates a | Model 3: Adjusted for health behaviorsb | Model 4: Adjusted for maternal BMIc | ||||
---|---|---|---|---|---|---|---|---|---|
RR | (95% CI) | RR | (95% CI) | RR | (95% CI) | RR | (95% CI) | ||
0 | 2997 (32) | 1 | – | 1 | – | 1 | – | 1 | – |
1–3 | 1704 (31) | 1.7 | (1.0, 2.8) | 1.6 | (1.0, 2.8) | 1.7 | (1.0, 2.8) | 1.7 | (1.0, 2.8) |
4–7 | 543 (14) | 2.4 | (1.3, 4.5) | 2.2 | (1.1, 4.3) | 2.1 | (1.1, 4.2) | 1.9 | (0.9, 3.9) |
Model 2 is adjusted for participant age at baseline (continuous), gender, participant body mass index at baseline (standardized to CDC standard growth curves; continuous with a squared term), nonwhite race, and perceived social status (continuous).
Model 3 is adjusted for Model 2 covariates plus participant body mass index in 2007 when PTSD was ascertained (in kg/m2, continuous with a squared term), participant history of dieting (yes or no, through 2005) and participant history of smoking (yes or no, through 2007).
Model 4 is adjusted for Model 3 covariates plus participant’s mother’s body mass index at age 18 (mothers are Nurses’ Health Study II participants; weight at age 18 was asked on the NHSII baseline survey and validated against nursing school entry physical exams).
Comorbid PTSD and depression symptoms and eating behaviors
PTSD and depression are frequently comorbid, and their combination may confer additional risk of problematic eating behaviors. Analyses of combinations of significant depressive symptoms (CESD-10 ≥ 10) and PTSD symptoms (≥ 4) indicated that each mental health problem was independently associated with each of the eating behavior phenotypes of interest (Table 4). However, the strength of the associations across the joint PTSD-depression categories differed qualitatively by eating behavior. Main covariate-adjusted binge eating analyses indicated that PTSD with depression (RR= 2.8; 95% CI: 2.3, 3.6) and PTSD without depression (RR=2.2; 95% CI: 1.5, 3.3) were more strongly associated with binge eating than depression alone (RR=1.8; 95% CI: 1.4, 2.2). In analyses of coping-motivated eating, exposure to the combination of PTSD and depression was associated with a larger elevation in risk (RR=2.3; 95% CI: 2.0, 2.6) than either depression alone (RR=1.9; 95% CI: 1.7, 2.1) or PTSD alone (RR=1.8; 95% CI: 1.4, 2.3). RRs were overall smaller for disinhibited eating and were similar across categories of PTSD, depression, and their combination.
Table 4.
Eating behavior outcome | Exposure to PTSD and/or Depression | Model 1: Crude | Model 2: Adjusted for main covariatesa | Model 3: Adjusted for health behaviorsb | |||||
---|---|---|---|---|---|---|---|---|---|
Dichotomous outcomes | PTSD | Depression | N (cases) | RR | (95% CI) | RR | (95% CI) | RR | (95% CI) |
Binge eating | |||||||||
No | No | 4567 (209) | 1 | – | 1 | – | 1 | – | |
No | Yes | 1572 (136) | 1.9 | (1.5, 2.3) | 1.8 | (1.4, 2.2) | 1.6 | (1.3, 2.0) | |
Yes | No | 245 (25) | 2.2 | (1.5, 3.3) | 2.2 | (1.5, 3.3) | 2.1 | (1.4, 3.2) | |
Yes | Yes | 685 (95) | 3.0 | (2.4, 3.8) | 2.8 | (2.2, 3.6) | 2.6 | (2.1, 3.3) | |
Top quartile of coping-motivated eating | |||||||||
No | No | 4325 (679) | 1 | – | 1 | – | 1 | – | |
No | Yes | 1478 (467) | 2.0 | (1.8, 2.2) | 1.9 | (1.7, 2.1) | 1.8 | (1.6, 2.0) | |
Yes | No | 232 (64) | 1.8 | (1.4, 2.2) | 1.8 | (1.4, 2.3) | 1.7 | (1.4, 2.1) | |
Yes | Yes | 657 (255) | 2.5 | (2.2, 2.8) | 2.3 | (2.0, 2.6) | 2.2 | (1.9, 2.5) | |
Top quartile of disinhibited eating | |||||||||
No | No | 4326 (772) | 1 | – | 1 | – | 1 | – | |
No | Yes | 1480 (443) | 1.7 | (1.5, 1.9) | 1.6 | (1.5, 1.8) | 1.5 | (1.4, 1.7) | |
Yes | No | 232 (51) | 1.2 | (1.0, 1.6) | 1.5 | (1.2, 1.9) | 1.4 | (1.1, 1.8) | |
Yes | Yes | 657 (201) | 1.7 | (1.5, 2.0) | 1.7 | (1.4, 1.9) | 1.5 | (1.3, 1.8) | |
Continuous outcomes | PTSD | Depression | N (means) | β | (95% CI) | β | (95% CI) | β | (95% CI) |
Coping-motivated eating (range 0–20) | |||||||||
No | No | 4325 (3.0) | 0 | – | 0 | – | 0 | – | |
No | Yes | 1478 (4.8) | 1.8 | (1.6, 2.1) | 1.6 | (1.4, 1.9) | 1.5 | (1.3, 1.8) | |
Yes | No | 232 (4.6) | 1.7 | (1.1, 2.2) | 1.4 | (0.9, 2.0) | 1.4 | (0.8, 1.9) | |
Yes | Yes | 657 (5.8) | 2.8 | (2.4, 3.2) | 2.5 | (2.1, 2.9) | 2.4 | (2.0, 2.8) | |
Disinhibited eating (range 0–10) | |||||||||
No | No | 4326 (2.7) | 0 | – | 0 | – | 0 | – | |
No | Yes | 1480 (3.5) | 0.8 | (0.7, 1.0) | 0.7 | (0.6, 0.9) | 0.6 | (0.5, 0.7) | |
Yes | No | 232 (3.4) | 0.7 | (0.4, 1.0) | 0.7 | (0.4, 1.0) | 0.7 | (0.3, 1.0) | |
Yes | Yes | 657 (3.6) | 0.9 | (0.7, 1.1) | 0.7 | (0.5, 1.0) | 0.6 | (0.4, 0.8) |
Model 2 is adjusted for participant age at baseline (continuous), gender, participant body mass index at baseline (standardized to CDC standard growth curves; continuous with a squared term), nonwhite race, and perceived social status (continuous).
Model 3 is adjusted for Model 2 covariates plus participant body mass index in 2007 when PTSD was ascertained (in kg/m2, continuous with a squared term), participant history of dieting (yes or no, through 2005) and participant history of smoking (yes or no, through 2007)
Associations of eating behavior and PTSD symptoms with obesity
In analyses assessing the implications of our eating behaviors for obesity risk, we found that binge, coping-motivated, and disinhibited eating in 2010 were each associated with elevated obesity prevalence in both women and men (Table 5). Among women, top quartile of coping-motivated eating and top quartile of disinhibited eating were associated with main covariate- and baseline BMI-adjusted RRs for obesity of 1.5 (95% CI: 1.3, 1.7), and 1.5 (95% CI: 1.3, 1.8), respectively, relative to those without the eating behavior. For binge eating, the sociodemographic-adjusted RR for obesity was 1.8 (95% CI: 1.5, 2.2), but this was attenuated upon adjustment for baseline BMI (RR=1.1; 95% CI: 0.9 1.3). It is possible that baseline BMI was influenced by long-standing binge eating behavior and that adjustment for baseline BMI therefore adjusted away some of the association of interest. For men, the obesity RRs associated with binge, coping-motivated, and disinhibited eating adjusting for sociodemographics and baseline BMI were 1.5 (95% CI: 1.0, 2.2), 1.5 (95% CI: 1.2, 1.9), and 2.0 (95% CI: 1.6, 2.6). PTSD symptoms in 2007 were also associated with obesity in 2013 in women (main covariate-adjusted RR=1.5, 95% CI: 1.2, 1.8; baseline BMI-adjusted RR=1.3, 95% CI: 1.1, 1.6); however, they were unrelated to obesity in 2013 in men.
Table 5.
Eating behavior/PTSD symptoms | N(cases) | Model 1: Crude | Model 2: Adjusted for sociodemographicsa | Model 3: Adjusted for baseline BMIb | |||
---|---|---|---|---|---|---|---|
RR | (95% CI) | RR | (95% CI) | RR | (95% CI) | ||
Women Binge eating | |||||||
No | 3672 (496) | 1 | – | 1 | – | 1 | – |
Yes | 334 (91) | 2.0 | (1.7, 2.4) | 1.8 | (1.5, 2.2) | 1.1 | (0.9, 1.3) |
Top quartile of coping-motivated eating | |||||||
No | 3036 (333) | 1 | – | 1 | – | 1 | – |
Yes | 857 (237) | 2.5 | (2.2, 2.9) | 2.4 | (2.0, 2.8) | 1.5 | (1.3, 1.7) |
Top quartile of disinhibited eating | |||||||
No | 3164 (352) | 1 | – | 1 | – | 1 | – |
Yes | 732 (219) | 2.7 | (2.3, 3.1) | 2.7 | (2.3, 3.2) | 1.5 | (1.3, 1.8) |
PTSD symptoms | |||||||
0 | 2047 (258) | 1 | – | 1 | – | 1 | – |
1–3 | 1465 (233) | 1.3 | (1.1, 1.5) | 1.2 | (1.0, 1.4) | 1.2 | (1.0, 1.4) |
4–7 | 566 (104) | 1.5 | (1.2, 1.8) | 1.5 | (1.2, 1.8) | 1.3 | (1.1, 1.6) |
Men Binge eating | |||||||
No | 1753 (245) | 1 | – | 1 | – | 1 | – |
Yes | 48 (18) | 2.7 | (1.8, 3.9) | 2.7 | (1.8, 4.0) | 1.5 | (1.0, 2.2) |
Top quartile of coping-motivated eating | |||||||
No | 1351 (158) | 1 | – | 1 | – | 1 | – |
Yes | 340 (94) | 2.4 | (1.9, 3.0) | 2.5 | (1.9, 3.1) | 1.5 | (1.2, 1.9) |
Top quartile of disinhibited eating | |||||||
No | 1226 (109) | 1 | – | 1 | – | 1 | – |
Yes | 466 (143) | 3.5 | (2.8, 4.3) | 3.4 | (2.6, 4.3) | 2.0 | (1.6, 2.6) |
PTSD symptoms | |||||||
0 | 1136(176) | 1 | – | 1 | – | 1 | – |
1–3 | 551(73) | 0.9 | (0.7, 1.1) | 0.8 | (0.6, 1.1) | 0.7 | (0.6, 0.9) |
4–7 | 185(30) | 1.0 | (0.7, 1.5) | 1.1 | (0.7, 1.5) | 1.0 | (0.7, 1.3) |
Discussion
In this large cohort of young men and women, we found strong associations between reported PTSD symptoms and three eating behavior phenotypes that are associated with obesity. We also found suggestive evidence of longitudinal associations between PTSD symptoms and new onset of binge eating. In general, PTSD symptoms were less strongly associated with disinhibited eating than with binge eating and coping-motivated eating, perhaps because binge and coping-motivated eating appear to be driven by emotional states (32; 46), whereas disinhibited eating may be more strongly influenced by external cues and/or problems with impulse control (33). Most eating behaviors examined were associated with increases in obesity risk. (Among women, the RR for binge eating was close to 1 after adjustment for baseline BMI; it is possible that this null value is due to overadjustment, as baseline BMI have been influenced by pre-baseline binge eating.)
Overall, these findings support hypotheses suggesting that overconsumption of food may be a common coping strategy for those experiencing significant psychological distress (47), which may in turn drive elevated obesity risks associated with PTSD symptoms. However, we did not find that PTSD symptoms were universally associated with obesity; only women in our cohort showed an association between PTSD symptoms in 2007 and obesity in 2013. It is unclear why no such association emerged for men; we explored the possibility that men with PTSD and problematic overeating behaviors disproportionately compensated for their overeating (e.g., by fasting or purging). However, we found that few men reported compensatory behaviors. It seems likely that PTSD-related obesity may emerge over time in men if those with PTSD symptoms continue to engage in obesogenic eating behaviors at a high rate (3). Consistent with our findings, prior studies have found stronger PTSD–obesity links among women (4–6), and there is recent evidence for an emergence of PTSD-related obesity among men at older ages (48).
Our findings also suggest that the combination of PTSD and depressive symptoms may put men and women at particularly high risk for problematic eating. The combination of PTSD and depressive symptoms may be more strongly associated with problematic eating simply because it indicates especially severe psychiatric distress. It is also possible that depressive symptoms undermine individuals’ capacity to cope with PTSD symptoms in constructive ways (22), exacerbating reliance on unhealthy coping strategies such as overeating.
Our study expands on existing literature linking PTSD to eating disorder diagnoses (14; 16; 17; 49–51), by highlighting common and less severe eating behaviors that may nonetheless be of concern for long-term health. In particular, revealing the associations between PTSD symptoms and sub-clinical behaviors may help better pinpoint targets for earlier intervention. The limitations of our study include lack of information about time of onset of coping-motivated eating and disinhibited eating, which prevented us from establishing the temporal relationship between PTSD symptoms and these eating behaviors. Further, our measures were survey-based and are therefore subject to errors in reporting. In particular, our measure of obesity is based on self-report and may be subject to error, despite several studies suggesting this error is likely to be minimal (41; 42). PTSD symptoms were assessed at a single time-point, limiting our ability to discriminate those with longer-term versus shorter-term experiences of PTSD. Nevertheless, our study has several notable strengths including a large community-based sample of both men and women and rich covariate data including baseline BMI. In addition, we were able to address prior gaps in the literature by examining the joint contributions of PTSD and depressive symptoms, looking at several eating behavior phenotypes, and exploring longitudinal relationships between PTSD symptoms and subsequent incidence of binge eating.
Our findings add to a growing body of evidence suggesting that problematic eating behaviors are common among those suffering from PTSD symptoms, and provide suggestive support for a temporal sequence from PTSD to initiation of problematic eating. Our results further suggest that these eating behaviors may be important contributors to obesity, highlighting a potentially important mechanism by which PTSD symptoms increase obesity risk. This potential mechanism warrants additional research, especially because it will clarify eating behaviors as possible targets for intervention in the prevention of obesity among those with PTSD. Clinicians treating those with traumatic stress symptoms should be aware that many patients, particularly those with both PTSD and depressive symptoms, have comorbid eating behaviors that may put them at risk for weight gain and obesity. Screening for eating problems in PTSD care, and development of integrated treatments that address trauma and problematic eating simultaneously, may be warranted. Further, clinicians treating people with obesity might consider screening for traumatic stress symptoms and associated eating phenotypes. Such screening could lead to more sensitive and effective care than lifestyle counseling alone. Finally, our results emphasize the need for health promotion and obesity prevention strategies that address the potential influence of psychological factors on individuals’ health behaviors.
Acknowledgments
Funding: Dr. Mason is supported by the Building Interdisciplinary Research Careers in Women’s Health Grant (# K12HD055887) from the Eunice Kennedy Shriver National Institutes of Child Health and Human Development (NICHD), the Office of Research on Women’s Health, and the National Institute on Aging, NIH, administered by the University of Minnesota Deborah E. Powell Center for Women’s Health. The content is solely the responsibility of the authors and does not necessarily represent the office views of the NICHD or NIH. This work was also supported by National Institutes of Health grants R01 HD066963 and R01 HD049889. Dr. Austin is supported by the U.S. Maternal and Child Health Bureau, Health Resources and Services Administration grants 6T71-MC00009 and T76-MC00001.
Footnotes
Conflict of Interest: The authors declare that they have no conflict of interest.
Ethical approval: All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
Informed consent: Informed consent was obtained from all individual participants included in the study.
References
- 1.Kessler RC, Berglund P, Demler O, Jin R, Merikangas KR, Walters EE. Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey Replication. Arch Gen Psychiatry. 2005;62:593–602. doi: 10.1001/archpsyc.62.6.593. [DOI] [PubMed] [Google Scholar]
- 2.Roberts AL, Agnew-Blais JC, Spiegelman D, et al. Posttraumatic Stress Disorder and Incidence of Type 2 Diabetes Mellitus in a Sample of Women: A 22-Year Longitudinal Study. JAMA Psychiatry. 2015 doi: 10.1001/jamapsychiatry.2014.2632. Epub ahead. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Kubzansky LD, Bordelois P, Jun HJ, et al. The Weight of Traumatic Stress: A Prospective Study of Posttraumatic Stress Disorder Symptoms and Weight Status in Women. JAMA Psychiatry. 2013 Nov; doi: 10.1001/jamapsychiatry.2013.2798. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Pagoto SL, Schneider KL, Bodenlos JS, et al. Association of post-traumatic stress disorder and obesity in a nationally representative sample. Obesity (Silver Spring) 2012;20:200–205. doi: 10.1038/oby.2011.318. [DOI] [PubMed] [Google Scholar]
- 5.Mitchell KS, Aiello AE, Galea S, Uddin M, Wildman D, Koenen KC. PTSD and obesity in the Detroit neighborhood health study. Gen Hosp Psychiatry. 2013;35:671–673. doi: 10.1016/j.genhosppsych.2013.07.015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Perkonigg A, Owashi T, Stein MB, Kirschbaum C, Wittchen HU. Posttraumatic Stress Disorder and Obesity. Evidence for a Risk Association. Am J Prev Med. 2009;36:1–8. doi: 10.1016/j.amepre.2008.09.026. [DOI] [PubMed] [Google Scholar]
- 7.Dedert EA, Becker ME, Fuemmeler BF, Braxton LE, Calhoun PS, Beckham JC. Childhood traumatic stress and obesity in women: the intervening effects of PTSD and MDD. J Trauma Stress. 2010;23:785–763. doi: 10.1002/jts.20584. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Brewerton T, O’Neil P, Dansky B, Kilpatrick D. Extreme Obesity and its Associations with Victimization, PTSD, Major Depression and Eating Disorders in a National Sample of Women. J Obes Eat Disord. 2015;1(2):2–9. [Google Scholar]
- 9.Ahmadi N, Hajsadeghi F, Mirshkarlo HB, Budoff M, Yehuda R, Ebrahimi R. Post-traumatic stress disorder, coronary atherosclerosis, and mortality. Am J Cardiol. 2011;108:29–33. doi: 10.1016/j.amjcard.2011.02.340. [DOI] [PubMed] [Google Scholar]
- 10.Boyko EJ, Jacobson IG, Smith B, et al. Risk of diabetes in U.S. military service members in relation to combat deployment and mental health. Diabetes Care. 2010;33:1771–1777. doi: 10.2337/dc10-0296. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Dedert EA, Calhoun PS, Watkins LL, Sherwood A, Beckham JC. Posttraumatic stress disorder, cardiovascular, and metabolic disease: a review of the evidence. Ann Behav Med. 2010;39(1):61–78. doi: 10.1007/s12160-010-9165-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Cota D, Tschöp MH, Horvath TL, Levine AS. Cannabinoids, opioids and eating behavior: the molecular face of hedonism? Brain Res Rev. 2006;51:85–107. doi: 10.1016/j.brainresrev.2005.10.004. [DOI] [PubMed] [Google Scholar]
- 13.Ulrich-Lai YM, Christiansen AM, Ostrander MM, et al. Pleasurable behaviors reduce stress via brain reward pathways. Proc Natl Acad Sci U S A. 2010;107:20529–20534. doi: 10.1073/pnas.1007740107. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Striegel-Moore RH, Garvin V, Dohm FA, Rosenheck RA. Eating disorders in a national sample of hospitalized female and male veterans: Detection rates and psychiatric comorbidity. Int J Eat Disord. 1999;25:405–414. doi: 10.1002/(SICI)1098-108X(199905)25:4<405::AID-EAT5>3.0.CO;2-F. [DOI] [PubMed] [Google Scholar]
- 15.Swinbourne JM, Touyz SW. The co-morbidity of eating disorders and anxiety disorders: a review. Eur Eat Disord Rev. 2007;15:253–274. doi: 10.1002/erv.784. [DOI] [PubMed] [Google Scholar]
- 16.Mitchell KS, Mazzeo SE, Schlesinger MR, Brewerton TD, Smith BN. Comorbidity of partial and subthreshold ptsd among men and women with eating disorders in the national comorbidity survey-replication study. Int J Eat Disord. 2012;45:307–315. doi: 10.1002/eat.20965. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Grilo CM, White Ma, Barnes RD, Masheb RM. Posttraumatic stress disorder in women with binge eating disorder in primary care. J Psychiatr Pract. 2012;18:408–412. doi: 10.1097/01.pra.0000422738.49377.5e. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Campbell DG, Felker BL, Liu C-F, et al. Prevalence of depression-PTSD comorbidity: implications for clinical practice guidelines and primary care-based interventions. J Gen Intern Med. 2007;22(6):711–718. doi: 10.1007/s11606-006-0101-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.de Wit L, Luppino F, van Straten A, Penninx B, Zitman F, Cuijpers P. Depression and obesity: A meta-analysis of community-based studies. Psychiatry Res. 2010;178:230–235. doi: 10.1016/j.psychres.2009.04.015. [DOI] [PubMed] [Google Scholar]
- 20.Pan A, Sun Q, Czernichow S, et al. Bidirectional association between depression and obesity in middle-aged and older women. Int J Obes. 2012;36:595–602. doi: 10.1038/ijo.2011.111. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Puccio F, Fuller-Tyszkiewicz M, Ong D, Krug I. A systematic review and meta-analysis on the longitudinal relationship between eating pathology and depression. Int J Eat Disord. 2016;49(5):439–454. doi: 10.1002/eat.22506. [DOI] [PubMed] [Google Scholar]
- 22.Orzechowska A, Zajączkowska M, Talarowska M, Gałecki P. Depression and ways of coping with stress: a preliminary study. Med Sci Monit. 2013;19:1050–1056. doi: 10.12659/MSM.889778. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Pu S, Nakagome K, Yamada T, et al. The relationship between the prefrontal activation during a verbal fluency task and stress-coping style in major depressive disorder: a near-infrared spectroscopy study. J Psychiatr Res. 2012;46(11):1427–1434. doi: 10.1016/j.jpsychires.2012.08.001. [DOI] [PubMed] [Google Scholar]
- 24.Talbot LS, Maguen S, Epel ES, Metzler TJ, Neylan TC. Posttraumatic stress disorder is associated with emotional eating. J Trauma Stress. 2013;26(4):521–525. doi: 10.1002/jts.21824. [DOI] [PubMed] [Google Scholar]
- 25.Michopoulos V, Powers A, Moore C, Villarreal S, Ressler KJ, Bradley B. The mediating role of emotion dysregulation and depression on the relationship between childhood trauma exposure and emotional eating. Appetite. 2015;91:129–136. doi: 10.1016/j.appet.2015.03.036. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Mason SM, Flint AJ, Roberts AL, Agnew-Blais J, Koenen KC, Rich-Edwards JW. Posttraumatic stress disorder symptoms and food addiction in women by timing and type of trauma exposure. JAMA psychiatry. 2014;71(11):1271–1278. doi: 10.1001/jamapsychiatry.2014.1208. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Mitchell KS, Wolf EJ. PTSD, food addiction, and disordered eating in a sample of primarily older veterans: The mediating role of emotion regulation. Psychiatry Res. 2016;243:23–29. doi: 10.1016/j.psychres.2016.06.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Hirth JM, Rahman M, Berenson AB. The association of posttraumatic stress disorder with fast food and soda consumption and unhealthy weight loss behaviors among young women. J Womens Health (Larchmt) 2011;20:1141–1149. doi: 10.1089/jwh.2010.2675. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Vilija M, Romualdas M. Unhealthy food in relation to posttraumatic stress symptoms among adolescents. Appetite. 2014;74:86–91. doi: 10.1016/j.appet.2013.12.002. [DOI] [PubMed] [Google Scholar]
- 30.Godfrey KM, Lindamer LA, Mostoufi S, Afari N. Posttraumatic stress disorder and health: a preliminary study of group differences in health and health behaviors. Ann Gen Psychiatry. 2013;12(1):30. doi: 10.1186/1744-859X-12-30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Carmassi C, Antonio Bertelloni C, Massimetti G, et al. Impact of DSM-5 PTSD and gender on impaired eating behaviors in 512 Italian earthquake survivors. Psychiatry Res. 2015;225(1–2):64–69. doi: 10.1016/j.psychres.2014.10.008. [DOI] [PubMed] [Google Scholar]
- 32.Jackson B, Cooper ML, Mintz L, Albino A. Motivations to eat: Scale development and validation. J Res Pers. 2003;37:297–318. doi: 10.1016/S0092-6566(02)00574-3. [DOI] [Google Scholar]
- 33.Stunkard AJ, Messick S. The three-factor eating questionnaire to measure dietary restraint, disinhibition and hunger. J Psychosom Res. 1985;29(1):71–83. doi: 10.1016/0022-3999(85)90010-8. http://www.ncbi.nlm.nih.gov/pubmed/3981480. Accessed May 7, 2015. [DOI] [PubMed] [Google Scholar]
- 34.Breslau N, Peterson EL, Kessler RC, Schultz LR. Short screening scale for DSM-IV posttraumatic stress disorder. Am J Psychiatry. 1999;156:908–911. doi: 10.1176/ajp.156.6.908. [DOI] [PubMed] [Google Scholar]
- 35.Roberts AL, Agnew-Blais J, Spiegelman D, et al. Posttraumatic stress disorder and type 2 diabetes incidence in women: An 18-year longitudinal study. Under Rev. 2013 doi: 10.1001/jamapsychiatry.2014.2632. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Goodman E, Whitaker RC. A prospective study of the role of depression in the development and persistence of adolescent obesity. Pediatrics. 2002;110:497–504. doi: 10.1542/peds.110.3.497. [DOI] [PubMed] [Google Scholar]
- 37.Andresen EM, Malmgren JA, Carter WB, Patrick DL. Screening for depression in well older adults: Evaluation of a short form of the CES-D. Am J Prev Med. 1994;10:77–84. [PubMed] [Google Scholar]
- 38.Irwin M, Artin KH, Oxman MN. Screening for depression in the older adult: criterion validity of the 10-item Center for Epidemiological Studies Depression Scale (CES-D) Arch Intern Med. 159(15):1701–1704. doi: 10.1001/archinte.159.15.1701. http://www.ncbi.nlm.nih.gov/pubmed/10448771. Accessed October 21, 2016. [DOI] [PubMed] [Google Scholar]
- 39.Field AE, Taylor CB, Celio A, Colditz GA. Comparison of Self-Report to Interview Assessment of Bulimic Behaviors among Preadolescent and Adolescent Girls and Boys. Int J Eat Disord. 2004;35:86–92. doi: 10.1002/eat.10220. [DOI] [PubMed] [Google Scholar]
- 40.Bohrer BK, Forbush KT, Hunt TK. Are common measures of dietary restraint and disinhibited eating reliable and valid in obese persons? Appetite. 2015;87:344–351. doi: 10.1016/j.appet.2014.12.226. [DOI] [PubMed] [Google Scholar]
- 41.Goodman E, Hinden BR, Khandelwal S. Accuracy of teen and parental reports of obesity and body mass index. Pediatrics. 2000;106:52–58. doi: 10.1542/peds.106.1.52. [DOI] [PubMed] [Google Scholar]
- 42.Strauss RS. Comparison of measured and self-reported weight and height in a cross-sectional sample of young adolescents. Int J Obes Relat Metab Disord. 1999;23:904–908. doi: 10.1038/sj.ijo.0800971. [DOI] [PubMed] [Google Scholar]
- 43.Kuczmarski RJ, Ogden CL, Guo SS, et al. 2000 CDC Growth Charts for the United States: methods and development. Vital Health Stat 11. 2002:1–190. [PubMed] [Google Scholar]
- 44.Troy LM, Hunter DJ, Manson JE, Colditz GA, Stampfer MJ, Willett WC. The validity of recalled weight among younger women. Int J Obes Relat Metab Disord. 1995;19(8):570–572. [PubMed] [Google Scholar]
- 45.Zou G. A Modified Poisson Regression Approach to Prospective Studies with Binary Data. Am J Epidemiol. 2004;159(7):702–706. doi: 10.1093/aje/kwh090. [DOI] [PubMed] [Google Scholar]
- 46.Berg KC, Crosby RD, Cao L, et al. Facets of Negative Affect Prior to and Following Binge-Only, Purge-Only, and Binge/Purge Events in Women With Bulimia Nervosa. J Abnorm Psychol. 2012 doi: 10.1037/a0029703. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Brewerton TD. Posttraumatic stress disorder and disordered eating: food addiction as self-medication. J Womens Health (Larchmt) 2011;20:1133–1134. doi: 10.1089/jwh.2011.3050. [DOI] [PubMed] [Google Scholar]
- 48.Smith BN, Tyzik AL, Neylan TC, Cohen BE. PTSD and obesity in younger and older veterans: Results from the mind your heart study. Psychiatry Res. 2015;229(3):895–900. doi: 10.1016/j.psychres.2015.07.044. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.von Lojewski A, Boyd C, Abraham S, Russell J. Lifetime and recent DSM and ICD psychiatric comorbidity of inpatients engaging in different eating disorder behaviours. Eat Weight Disord. 2012;17:e185–e193. doi: 10.1007/BF03325346. http://search.ebscohost.com/login.aspx?direct=true&db=psyh&AN=2012-28996-005&site=ehost-live. [DOI] [PubMed] [Google Scholar]
- 50.Hoerster KD, Jakupcak M, Hanson R, et al. PTSD and depression symptoms are associated with binge eating among US Iraq and Afghanistan veterans. Eat Behav. 2015;17:115–118. doi: 10.1016/j.eatbeh.2015.01.005. [DOI] [PubMed] [Google Scholar]
- 51.Dansky BS, Brewerton TD, Kilpatrick DG, O’Neil PM. The National Women’s Study: relationship of victimization and posttraumatic stress disorder to bulimia nervosa. Int J Eat Disord. 1997;21:213–228. doi: 10.1002/(SICI)1098-108X(199704)21:3<213::AID-EAT2>3.0.CO;2-N. [DOI] [PubMed] [Google Scholar]