Acute Daily Stress,Daily Food Consumption,and the Moderating Effect of Disordered Eating among Adolescents with Overweight/Obesity

Abstract

Objective:

The purpose of this study was to examine the associations between acute daily stress dimensions (frequency, sum) and food intake in adolescents with overweight/obesity, and to explore the potential moderating effect of disordered eating behaviors on these associations.

Methods:

One hundred eighty-two adolescents with overweight/obesity (65% females; 68.7% non-white; 16.2 ± 1.8 years of age) were included in this analysis. Acute daily stress was measured using the Daily Stress Inventory, and daily caloric intake was measured using a food frequency questionnaire. Disordered eating behavior was assessed using the Eating Attitudes Test (EAT-26).

Results:

Acute daily stress frequency (B = 0.013 ± 0.003; p < 0.001) and acute daily stress sum (B = 0.003 ± 0.001; p < 0.001) were associated with greater daily caloric intake. Disordered eating behavior moderated the association between acute daily stress frequency and caloric intake (p_interaction = 0.039), with greater daily caloric intake among those with higher levels of disordered eating. Disordered eating behavior did not significantly moderate the association between acute daily stress sum and daily caloric intake (p_interaction = 0.053).

Conclusions:

These findings suggest that greater exposure to acute daily stressors may increase daily food intake in adolescents with overweight/obesity, with greater susceptibility among those engaging in high levels of disordered eating. Longitudinal research is warranted to elucidate the long-term effect of acute daily stressors and disordered eating on food intake among adolescents with overweight/obesity. The Health and Culture Project is registered at www.clinicaltrials.gov (No. NCT02938663).

Introduction

Acute daily stressors are brief, time-limited exposures that are common and often include events such as rushing, deadlines, and arguments.¹ Under normal circumstances, acute stress and subsequent stress responses typically do not have long-term health consequences.¹ However, if an individual has repeated exposure to an acute stressor or perceives a continual threat of a stressor reoccurring, then this acute stressor may manifest into a chronic stressor, increasing susceptibility to long-term health risk.

Adolescence is a particularly vulnerable period of heightened stress due, in part, to biological changes, interpersonal conflicts with peers, increased academic load at school, and greater performance-related stressors.^2–5 Adolescents with overweight/obesity often experience additional psychosocial stressors related to their weight. Unlike their peers with lower body weights, adolescents with overweight/obesity are more likely to receive criticism from close family members because of their weight, have higher incidences of teasing and rejection, and are twice as likely to be bullied.^6–9 The heightened exposure to stress observed in this high weight group increases their susceptibility to stress eating, a maladaptive behavioral response to both acute and chronic stressors.¹⁰

Adolescents with overweight/obesity display several characteristics that may further increase their risk of stress eating. In comparison with their lower weight peers, adolescents with overweight/obesity have greater sensitivity and exaggerated cortisol responses to acute stress.¹¹ Like adults with obesity, youth with overweight/obesity have been shown to exhibit lower satiety and greater sensitivity and responsiveness to appetizing food cues (e.g., smell) compared with lower weight youth.^12,13

Evidence also suggests that adolescents with obesity are prone to exhibiting greater dietary restraint,^14,15 a characteristic associated with counterregulatory eating when exposed to acute stressors.^16,17 Lastly, adolescents with overweight/obesity have also been shown to engage in emotional eating in response to stress.¹⁶ In sum, these characteristics suggest that adolescents with overweight/obesity may have greater vulnerability to stress-induced eating.

While adolescents with overweight/obesity may be more susceptible to stress eating, empirical data and results supporting this association within higher weight (higher BMI) sample populations are mixed. In adults with overweight, Lemmens et al. observed increased caloric intake after an acute psychological stressor.¹⁸ Herhaus et al. more recently observed increased food consumption among men and women with obesity following an acute stressor in a laboratory environment.¹⁹

In contrast, Nagy et al. observed decreased food intake following an acute laboratory stressor in adolescents with overweight/obesity and heightened cortisol reactivity.²⁰ Appelhans et al. also noticed reduced food intake among women with obesity and high stress responses following an acute stressor.²¹ Given this inconsistency in food intake among sample populations with higher weight/BMI and limited studies examining this association among higher weight/BMI adolescents, more studies are needed to identify factors that contribute or otherwise alter eating behavior during periods of heightened acute stress within this group.

One plausible link that may influence the relationship between stress and food intake is disordered eating behavior. Disordered eating behaviors are defined as problematic eating behaviors (e.g., restrictive, bingeing, self-control, and compensatory behaviors) occurring at a lower frequency or severity than clinically diagnosed eating disorders.²² Engagement in disordered eating often stems from body dissatisfaction and a desire to control one's weight/shape. Furthermore, body dissatisfaction, shape/weight concerns, dietary restriction, and the use of extreme or otherwise unhealthy weight control behaviors have been found to be more common among adolescents with higher weight status.²³ For example, epidemiological evidence suggests that ∼40% of adolescent girls and 20% of adolescent boys with overweight/obesity engage in disordered eating behaviors over a year.²⁴

Empirical evidence also suggests that disordered eating predicts further weight gain among children and adolescents.^25–27 It has also been previously demonstrated that adolescents with overweight/obesity are at greater risk of engaging in disordered eating behaviors due to increased stress exposure.²⁸ More recently, our group observed cross-sectional associations between increased psychological stress and greater dieting behavior in adolescents with overweight/obesity.²⁹ Examining the potential moderating role of disordered eating behaviors in the association between stress and eating behavior in adolescents with overweight/obesity may provide additional insight into how acute stress predicts eating behavior in this at-risk group.

The purpose of this study was to examine the associations between acute daily stress and daily caloric food intake in adolescents with overweight/obesity, and to determine the potential moderating effect of disordered eating behavior on these associations. Given the racial/ethnic diversity of the sample, an exploratory aim examining how race/ethnicity may influence acute stress and eating behaviors was also conducted. The primary hypothesis was that higher acute daily stress (as defined by frequency and sum) would be associated with higher daily caloric intake. In addition, it was hypothesized that the associations between acute daily stress and daily caloric intake would be modified by disordered eating behavior.

Methods

Sample Population and Procedure

Baseline data from two studies conducted at the University of Michigan Childhood Disparities Research Laboratory (CDRL) were included in this analysis: the Health and Culture Project (HCP) and the Stress, Obesity, and Diabetes in Adolescents (SODA) study. The primary aim of the HCP study was to examine the social, cultural, and psychological predictors of obesity in adolescents. The primary aim of the SODA study was to examine cross-sectional associations among different indices of stress (exposure, perception, biological response) on type 2 diabetes risk in Hispanic and non-Hispanic black and white adolescents with overweight/obesity. Descriptions of both studies have been previously published.^29,30 Data collection procedures for both studies occurred at the CDRL, off-site at local schools, afterschool programs, community centers, or at the University of Michigan Clinical Research Unit.

Participants were included in the present analysis if: (1) they were between the ages of 13 and 19 years; (2) they were individuals with overweight/obesity (BMI percentile ≥85th percentile); (3) they had complete dietary intake data; and (4) they had completed acute daily stress data. In total, a subset of 182 adolescents with overweight/obesity were included in this analysis. There were no significant differences in distribution of race/ethnicity, gender, mean age, BMI percentile, BMI, pubertal development score, or acute daily stress among those who were included and those excluded from the analysis (p > 0.05).

Before starting both studies (HCP and SODA studies), all experimental procedures were approved by the University of Michigan Institutional Review Board (HUM00083264 and HUM00080820). Informed assent was obtained from all participants younger than the age of 18, and written informed consent obtained from each of their caregivers before the start of data collection procedures. For participants 18 years and older, solely written informed consent was obtained.

Dependent Variable

Total daily caloric intake

Total daily caloric intake is the average amount of energy consumed via food and beverage daily, and was obtained using the 2012 version of the Youth/Adolescent Food Frequency Questionnaire (YAQ) developed by Harvard University.^31,32 The YAQ is a self-reported questionnaire that asks questions related to food and drink consumption throughout the preceding year. All completed YAQs were shipped to the T.H. Chan School of Public Health at Harvard University, where they were analyzed for mean total daily calories (kcals) along with other measures (e.g., total fat) using the Nutrition Data System for Research Windows-based dietary analysis program. The questionnaire is designed for adolescents, and has been previously validated and calibrated as a semiquantitative food frequency questionnaire.³²

Independent Variable

Acute daily stress

Acute daily stress was conceptualized as a brief, short-term stressor involving a specific event.³³ Acute daily stress was assessed using the Daily Stress Inventory,³⁴ a measure of stress designed to assess daily sources and individualized impact of stressful events. The Daily Stress Inventory is a 58-item self-reported questionnaire that allows individuals to indicate the stressful events that they may have experienced over the past day (24 hours). An example question from the Daily Stress Inventory included: “In the past 24 hours…I was criticized or verbally attacked.” Participants were asked to respond honestly, indicating whether or not the event occurred, and if so, how much stress did it cause on a Likert-type scale anchored by (1) “occurred but was not stressful” to (7) “caused me to panic.”

Two scores were derived from the Daily Stress Inventory: an acute daily stress frequency score (FREQ) and an acute daily stress sum (SUM). FREQ was calculated as the total number of stressful events that occurred in the past 24 hours with possible scores ranging from 0 to 58. SUM was calculated as the summation of the impact ratings of the endorsed events (i.e., how much stress did each event cause), with a maximum possible score of 406.³⁴ The Daily Stress Inventory demonstrated excellent internal consistency within the study sample for acute stress FREQ (Cronbach's α = 0.94) and SUM (Cronbach's α = 0.97). The Daily Stress Inventory has been previously used in adolescent samples.^35,36

Covariates

Given the possible confounding influence of gender, race/ethnicity, pubertal development, and BMI on the association between psychological stress and caloric intake, these variables were added as covariates and controlled for in all of the statistical models.^10,37–39 Anthropometric measures of height and weight were measured in the CDRL by trained study staff members. Height (cm) was measured to the nearest 0.1 cm using ShorrBoard^® (Weigh and Measure, LLC., Olney, MD) with standardized procedures. Body weight (kg) was measured to the nearest 0.1 kg, using an electronic scale (Doran Scales, Inc., Batavia, IL). BMI for age and sex percentile was calculated using the CDC guidelines (www.cdc.gov/growthcharts).

Pubertal development was measured using the Pubertal Development Scale (PDS).⁴⁰ Based on the self-reported completion of developmental stages on the PDS, pubertal category scores were calculated and classified into the following categories: prepubertal, early pubertal, midpubertal, late pubertal, or postpubertal. Given the hormonal changes that occur with puberty, the variable was used as a covariate rather than biological age. Race and gender were self-reported.

Moderator

Disordered eating was defined and conceptualized as problematic eating behaviors (e.g., restrictive, bingeing, compensatory behaviors) occurring at a lower frequency or severity than clinically diagnosed eating disorders.²² Disordered eating behavior was measured using the total Eating Attitudes Test (EAT-26) score from the 26-item EAT-26.⁴¹ The EAT-26 questionnaire is a screening instrument used to identify participants with various levels and types of eating pathology, who may be at risk for eating disorders and may therefore need to be referred to a qualified professional.⁴² It is an abbreviated version of the EAT-40 questionnaire created by Garner et al.⁴¹

The EAT-26 contains three subscales; the dieting subscale, bulimia and food preoccupation subscale, and an oral control subscale altogether assessing restrictive, bingeing, self-control, and compensatory behaviors. For each question, participants were asked to indicate how often each statement applied to them on a Likert-type scale anchored by 1 (never) to 6 (always). Example items from each of the EAT-26 subscales included: “I am occupied with a desire to be thinner” (dieting), “I have gone on eating binges where I feel that I may not be able to stop” (food preoccupation), and “I display self-control around food” (oral control).

Total EAT-score was obtained by adding all items from the questionnaire (including a reverse scored item), to get a general score of disordered eating behavior. Higher self-reported scores on EAT-26 denoted greater concern regarding body weight, body shape, and eating.⁴¹ The questionnaire demonstrated acceptable internal consistency for the study sample (Cronbach's α = 0.79). The EAT-26 questionnaire has been previously used to assess disordered eating risk in adolescent samples and among individuals with overweight/obesity.^43,44

Statistical Analysis

All data analyses were conducted using Stata 16.0 (StataCorp LLC, College Station, TX). Before all analyses, variables were evaluated for normality, and total daily energy intake was natural log-transformed to help normalize its distribution. Means and frequencies were determined for age, gender, race, BMI, BMI percentile, pubertal development, FREQ, SUM, and total EAT-26 score. For main effects models, multivariable linear regression was used to examine the association between the independent acute daily stress dimensions (FREQ, SUM) and the dependent variable (total daily caloric intake), adjusted for all a priori covariates (race, gender, BMI, and pubertal development) in each model.

In the interaction models exploring the moderating effects of disordered eating behavior on the relationship between the acute daily stress dimensions and daily caloric intake, an interaction term was added to each of the acute daily stress dimension (e.g., acute daily stress FREQ × EAT-score) in the regression models to test moderation. Similarly, for the exploratory aim examining how race/ethnicity may influence acute stress and caloric intake, race/ethnicity was added as an interaction term. A continuous disordered eating variable was used in place of a dichotomized or categorized variable to avoid the common issues with such variables (i.e., arbitrariness of dichotomizing at the median).^45,46 A p-value of p < 0.05 was set to determine significance for all models.

Results

Participant Characteristics

Participant characteristics in mean and standard deviation are seen in Table 1. In total, study participants were 65% female (35% male) and 68.7% non-white. Among the nonwhite adolescents, 63.2% of participants self-identified as non-Hispanic black or African American, 4.4% as Hispanic, and 1.1% as other (e.g., Asian, Pacific-Islander). Participants had a mean age of 16.2 ± 1.8 years and a mean BMI (36.8% white) percentile of 94.3% ± 4.2% (BMI: 31.1 ± 6.1 kg/m²). Approximately 44% (n = 80) of sample participants were classified as adolescents with overweight and 56% (n = 102) were classified as adolescents with obesity. On average, participants consumed 2007 kcals/day, consistent with the national average of 2100 kcals for boys and 1755 kcals for girls.⁴⁷ There were, however, 16 participants with daily caloric intake values above 3500 kcals/day, and 4 participants with a daily caloric intake value below 500 kcals/day.

Table 1.

Participant Characteristics

Variables	N	Mean (SD) or frequency (%)
Age (years)	182	16.2 (1.8)
Gender (% female)	182	65%
Race/ethnicity (% nonwhite)	182	68.7%
Non-Hispanic black	115	63.2%
Non-Hispanic white	57	31.3%
Hispanic	8	4.4%
Other races/ethnicities	2	1.1%
BMI (kg/m²)	182	31.1 (6.1)
BMI percentile	182	94.3 (4.3)
Pubertal category score	182	9.8 (2.0)
Acute daily stress frequency (FREQ)	182	18.0 (11.9)
Total acute daily stress (SUM)	182	47.3 (47.7)
EAT-26 score (disordered eating)	182	7.8 (7.1)
Total daily caloric intake (kcals)	182	2007.1 (1187.5)

Statistics are reported in either mean and standard deviation (SD) or frequency (%).

EAT, Eating Attitudes Test.

FREQ score was 18.0 ± 11.9, with participants experiencing on average 18 acute stress events over a 24-hour period. Mean SUM score of the impact of acute daily stressors was 47.3 ± 47.7 over a 24-hour period. Both the mean FREQ and SUM scores of acute daily stressors are consistent with acute daily stress FREQ and SUM scores observed in other adolescent samples.^36,48 The average sample score for the EAT-26 was 7.8 ± 7.1 for disordered eating behavior, a score that is also consistent with a prior study conducted in an adolescent sample.⁴⁹ Furthermore, there were 170 participants (93.4% of total sample) with EAT-26 scores under 20 (the suggested cutoff value for a recommendation to see a qualified professional for a possible eating disorder), and 12 participants (4.6% of total sample) with EAT-26 scores at or above the 20-point cutoff.

Associations between Acute Daily Stress FREQ, SUM, and Total Daily Caloric Intake

The associations between acute daily stress FREQ and SUM, with that of total daily caloric intake, can be seen in Table 2. The FREQ score of acute daily stress was positively associated with total daily caloric intake, with a higher number of acute daily stress events, associated with greater total daily caloric intake (B = 0.013 ± 0.003; p < 0.001). The SUM score of acute daily stress was also associated with total daily caloric intake among the study sample, with higher total distress from acute stress events associated with greater total daily caloric intake (B = 0.003 ± 0.001; p < 0.001).

Table 2.

Association between Acute Daily Stressors and Daily Caloric Intake

Predictor	Caloric intake (B ± SE)	p
Total distress (SUM)	0.003 ± 0.001	<0.001^*
BMI percentile	−0.017 ± 0.009	0.06
Gender	−0.255 ± 0.086	0.003^*
Race	0.021 ± 0.063	0.74
Pubertal development score	−0.006 ± 0.021	0.78
Acute stress frequency (FREQ)	0.013 ± 0.003	<0.001^*
BMI percentile	−0.017 ± 0.01	0.053
Gender	−0.243 ± 0.086	0.005^*
Race	0.026 ± 0.063	0.68
Pubertal development score	−0.001 ± 0.021	0.95

Natural log-transformed caloric intake values.

indicates significant difference at p < 0.05.

Acute Daily Stress Dimensions and the Moderating Effects of Disordered Eating

Interaction models were conducted to examine the moderating effects of disordered eating behavior in the association between acute daily stress dimensions and daily caloric intake. It was observed that disordered eating behavior (EAT-26 score) modified the association between acute daily stress FREQ and daily caloric intake (p_interaction = 0.039). The model suggested that for individuals with higher EAT-26 scores, a higher number of acute daily stress events (FREQ) was associated with greater daily caloric intake compared with adolescents with lower EAT-26 scores (Table 3). Disordered eating behavior, however, did not significantly modify the association between acute daily stress SUM and daily caloric intake, but trended toward significance (p_interaction = 0.053; Table not shown).

Table 3.

Moderating Effect of Disordered Eating on the Association between Acute Daily Stress FREQ and Daily Caloric Intake

Predictor	Caloric intake (B ± SE)	p
Acute daily stress FREQ × total EAT-score	0.001 ± 0.000	0.039^*
BMI percentile	−0.017 ± 0.01	0.055
Gender	−0.23 ± 0.085	0.008^*
Race	0.02 ± 0.06	0.78
Pubertal development score	0.004 ± 0.02	0.84

Natural log-transformed caloric intake values.

Exploratory Analyses of the Influence of Race/Ethnicity on Acute Stress and Eating Behaviors

Given the small representation of participants who were Hispanic or other (those self-identifying with another racial/ethnic group apart from Hispanic, non-Hispanic black, or white), no racial/ethnic comparisons were conducted including these two groups in the exploratory analyses. However, among non-Hispanic white and black participants, there were significant racial differences between scores for acute daily stress. Higher acute daily stress SUM (non-Hispanic black: 53.7 ± 5.1 vs. non-Hispanic white: 34.2 ± 3.7; p = 0.01) and daily stress FREQ scores (non-Hispanic black: 19.4 ± 1.2 vs. non-Hispanic white: 15.2 ± 1.1; p = 0.03) were noted among non-Hispanic black participants compared with their white peers.

A racial difference in disordered eating was also observed, with higher disordered eating levels (EAT-26 scores) among non-Hispanic black participants compared with their non-Hispanic white peers (non-Hispanic black: 8.58 ± 0.75 vs. non-Hispanic white: 6.28 ± 0.65; p = 0.048). Non-Hispanic black participants also reported greater total daily caloric intake compared with their white counterparts, but this difference was not statistically significant (non-Hispanic whites: 1785.1 ± 96.3 kcals/day vs. non-Hispanic blacks: 2126.7 ± 125.7 kcals/day; p = 0.075). Race/ethnicity did not modify the association between acute daily stress FREQ and total daily caloric intake, nor the association between acute daily stress SUM and total daily caloric intake (p's > 0.05).

Discussion

Consistent with our primary hypothesis, significant associations were observed between acute daily psychological stress dimensions and caloric intake, where higher daily stress FREQ and SUM were associated with greater daily caloric intake within our diverse sample of adolescents with overweight/obesity. In addition, adolescents who reported engaging in more disordered eating behaviors (as indicated by higher EAT-26 scores) were at an even greater risk of engaging in stress eating than those engaging in less disordered eating behaviors. Lastly, race/ethnicity differences were observed across acute stress and disordered eating levels, however, race/ethnicity did not moderate the association between acute daily stress dimensions and daily caloric intake. Given the higher stress exposure of adolescents with overweight/obesity and increased susceptibility to stress eating, it is essential to find ways to facilitate and equip higher weight adolescents with healthy stress coping strategies.

While many researchers have focused on chronic stress,^50–52 much less research has examined the effects of acute daily psychological stress on eating behavior in adolescents. One such study, however, was conducted by Michaud et al., who previously observed that increased acute stress stemming from an academic examination was associated with increased energy intake among adolescents.⁵³ Similarly, the present study extends these findings by observing increased daily caloric intake among adolescents with higher weight status in response to acute daily stress. Although acute stress responses are typically associated with suppression of appetite,⁵⁴ it is plausible that under circumstances where there is high cumulative burden of acute stress (as measured here in number of events and amount of distress), maladaptive behavioral responses may occur. Adolescents with overweight/obesity may, in turn, seek to cope with their stress via increased food intake of palatable comfort foods.

Eating palatable foods triggers increased secretion of dopamine, a hormone that aids in reducing feelings of stress and lowers the hypothalamic pituitary adrenal (HPA)-axis response.⁵⁵ With evidence suggesting that adolescents with overweight/obesity display high dietary restraint^14,17—a characteristic associated with overeating when faced with an acute stressor—it is possible that adolescents may overeat when facing a high cumulative burden of acute daily stress. In addition to a higher cumulative burden of acute daily stress, evidence from the stress literature also points to the salience of stressors such as weight stigma as a social/stereotype threat known to increase food intake among those with higher weight.⁸ Considering that unhealthy behaviors persist throughout adolescence and may track into adulthood,⁵⁶ it is necessary to provide social-emotional resources to help adolescents with overweight/obesity cope with the pressing acute stressors they may experience on a daily basis.

In the present study, it was also observed that adolescents engaging in disordered eating were especially vulnerable to higher food intake when encountering acute daily stress. These findings are noteworthy, given the increased prevalence and susceptibility of developing eating disorders, dysfunctional attitudes to food, and heightened body dissatisfaction among adolescents with overweight/obesity.^23,24 It is possible that participants who exhibited high disordered eating behavior may already have a pathological relationship with food, and thus, when experiencing acute bouts of stress (e.g., acute daily stressors) may experience reduced self-control, resulting in increased food intake. This phenomenon has been found to be particularly noticeable among those with higher weight, where stress from weight stigma, for example, has been found to deplete self-regulatory resources.⁵⁷

Emerging evidence also suggests that children with overweight/obesity with coexisting disordered eating behavior (i.e., loss of control eating) may display even lower self-regulation than those with overweight/obesity without disordered eating.⁵⁸ Given the higher prevalence of disordered eating behaviors among youth with overweight/obesity, in addition to the strong link between stress and self-regulation,^59,60 the combination of decreased self-regulation and increased disordered eating behavior may affect food intake in this group of adolescents when experiencing acute stress. Nevertheless, future research should explore the long-term influence of acute stress on caloric intake with specific disordered eating behaviors (e.g., binge eating) as moderators among adolescents with overweight/obesity.

Due to the racial/ethnic diversity of the study sample, exploratory analyses examining how race/ethnicity may influence stress and eating behaviors were conducted. Although non-Hispanic black adolescents had higher levels of acute daily stressors, in addition to higher levels of disordered eating behavior, we observed that race/ethnicity did not moderate the associations between acute daily stressors and daily caloric intake. It should be noted, however, that the higher levels of acute daily stress observed among non-Hispanic black participants compared with non-Hispanic white adolescents are consistent with higher social and environmental stress exposure observed in the existing stress literature (e.g., Ref.⁶¹).

Although race/ethnicity can be used as a proxy measure of understanding race-related stress exposure (e.g., racial discrimination), it is plausible that the lack of a significant moderating effect of race/ethnicity could be due to race/ethnicity not capturing specific race-related social/environmental stressors. Nevertheless, future studies examining potential racial/ethnic differences in the association between acute daily stress and eating behaviors should use specific race-related stressors, which may provide additional mechanistic insight into the pathways of stress and eating behavior across racial groups.

Important strengths and limitations of this study should be noted. Strengths of the study include the assessment of relevant dependent variables in a racially diverse sample of adolescents at a higher weight status. In addition, the comprehensive assessment of acute daily stress (utilizing frequency and total distress) is an additional strength of this study. Despite these strengths, the present findings should be interpreted in light of a few limitations. First, the cross-sectional study design limits inferences on the directionality of the associations between the acute daily stress dimensions and caloric intake, and causality cannot be inferred. Second, funding limitations precluded the use of a gold standard measurement of current energy intake (e.g., doubly labeled water or direct calorimetry), and as a result, the study utilized the commonly used self-reported food frequency questionnaire.

Although the food frequency questionnaire is not a gold standard measurement of current caloric intake, it has been found to be a valid measure of caloric intake across various adolescent samples.^62,63 Moreover, the EAT-26 is a self-reported screening measure of disordered eating behavior and did not explicitly capture specific types of disordered eating behaviors that may be common to adolescents with overweight/obesity such as binge eating and loss of control overeating. Nonetheless, the EAT-26 is among the most widely used standardized measure of symptoms and concerns of eating pathology,⁴² and was used presently given the wide array of pathological eating behaviors that can be experienced by adolescents with overweight/obesity. Additional research using an interview method and the contextualization of specific types of disordered eating behavior among adolescents with higher weight is necessary to gain a deeper mechanistic understanding of the interrelationship between disordered eating, stress, and food intake in this group.

Lastly, it should be noted that the use of self-reported daily stress, and the recruitment of a convenience sample of adolescents with overweight/obesity, limits generalizability to other adolescents without overweight/obesity. Despite these limitations, we observed significant associations between acute daily stress and higher daily caloric intake along with moderating effects of disordered eating among adolescents with overweight/obesity.

Conclusions and Implications

Overall, these findings suggest that acute daily stressors are associated with increased daily caloric intake among adolescents with overweight/obesity, and that those who engage in disordered eating behaviors may be particularly at risk of increasing food intake when experiencing acute bouts of stress. If experiencing acute stress in the long term, this group of adolescents may further gain weight and face higher cardiometabolic disease risk. This possible long-term effect of stress on food intake has been previously demonstrated, as chronic stressors have been shown to be associated with increased food intake, particularly sweetened and highly processed foods (e.g., Ref.⁶⁴).

Given that modest changes in energy intake such as an additional 100 kcals/day can lead to continuous weight gain,⁶⁵ higher consistent caloric intake overtime as a result of acute stress may be detrimental to the long-term health and well-being of adolescents with overweight/obesity. While the vast majority of the sample participants had an EAT-26 score below the 20-point cutoff, engaging in higher levels of the behaviors listed in the EAT-26 questionnaire points to a higher risk of eating disorders and an advanced progression along the disordered eating continuum.⁶⁶

Adolescents with overweight/obesity with higher levels of disordered eating behavior, who when experiencing high daily stress, may be especially vulnerable to increased food intake. When engaging in behaviors such as restriction and bingeing at higher levels overtime (chronically), it is highly plausible that this may lead to increased weight gain in addition to actual manifestation of an eating disorder. While our findings may not have direct clinical or treatment significance, it is important for clinicians and researchers alike to find ways to prevent this at-risk group of adolescents from being caught in a vicious and dangerous cycle of weight gain and pathological eating behaviors.

With the increased stress and mental health challenges among youth in the 21st century⁶⁷ and easier access to unhealthy foods,⁶⁸ the challenge of finding adaptive multipronged strategies and interventions focused on promoting healthy coping in this group is of the utmost importance. Strategies such as adopting antibullying policies within school settings to protect students from weight-based bullying, as well as providing families with the education and resources to recognize the harmful impact of weight stigma, should be prioritized.⁶⁹

Fostering social support among families and friends of those experiencing weight stigma and encouraging adolescents to use and practice strategies such as positive self-talk and healthy lifestyle behaviors are additional approaches that should be considered among several others. In doing so, researchers, clinicians, and families alike may ultimately help adolescents break the cycle of stress-related eating behaviors among those at greatest risk.

Footnotes

Acknowledgments

We would like to thank the Health and Culture Study team (HCP), the Stress, Obesity, and Diabetes in Adolescents (SODA) study team, Michigan Clinical Research Unit, and Michigan Consulting for Statistics, Computing and Analytics Research. We are also grateful for our study participants and their families for their involvement. The results of this study are presented clearly, honestly, and without fabrication, falsification, or inappropriate data manipulation.

Funding Information

This work was supported by the American Diabetes Association (Grant No. 7-13-JF-24), the University of Michigan School of Public Health Momentum Center, and the University of Michigan Office of Research. The Health and Culture Project is registered at (No. NCT02938663).

Author Disclosure Statement

No competing financial interests exist.

References

Epel

, Crosswell

, Mayer

, et al. More than a feeling: A unified view of stress measurement for population science. Front Neuroendocrinol, 2018; 49:146–169.

Markova

, Nikitskaya

. Coping strategies of adolescents with deviant behaviour. Int J Adolesc Youth, 2017; 22:36–46.

Rith-Najarian

, McLaughlin

, Sheridan

, Nock

. The biopsychosocial model of stress in adolescence: Self-awareness of performance versus stress reactivity. Stress, 2014; 17:193–203.

Roemmich

, Lambiase

, Balantekin

, et al. Stress, behavior, and biology: Risk factors for cardiovascular diseases in youth. Exerc Sport Sci Rev, 2014; 42:145–152.

Steinberg

, Morris

. Adolescent development. Annu Rev Psychol, 2001; 52:83–110.

Pont

, Puhl

, Cook

, et al. Stigma experienced by children and adolescents with obesity. Pediatr, 2017; 140. [Epub ahead of print]; DOI: 10.1542/peds.2017–3034.

Puhl

, Luedicke

. Weight-based victimization among adolescents in the school setting: Emotional reactions and coping behaviors. J Youth Adolesc, 2012; 41:27–40.

Tomiyama

. Weight stigma is stressful. A review of evidence for the cyclic obesity/weight-based stigma model. Appetite, 2014; 82:8–15.

Van Den Berg

, Neumark-Sztainer

, Eisenberg

, Haines

. Racial/ethnic differences in weight-related teasing in adolescents. Obesity, 2008; 16(S2):S3-S10.

10.

Jääskeläinen

, Nevanperä

, Remes

, et al. Stress-related eating, obesity and associated behavioural traits in adolescents: A prospective population-based cohort study. BMC Public Health, 2014; 14:321.

11.

Verdejo-Garcia

, Moreno-Padilla

, Garcia-Rios

, et al. Social stress increases cortisol and hampers attention in adolescents with excess weight. PLoS One, 2015; 10. [Epub ahead of print]; DOI:10.1371/journal.pone.0123565.

12.

Carnell

, Wardle

. Appetite and adiposity in children: Evidence for a behavioral susceptibility theory of obesity. Am J Clin Nutr, 2008; 88:22–29.

13.

Jansen

, Theunissen

, Slechten

, et al. Overweight children overeat after exposure to food cues. Eating Behav, 2003; 4:197–209.

14.

Hofmann

, Ardelt-Gattinger

, Paulmichl

, et al. Dietary restraint and impulsivity modulate neural responses to food in adolescents with obesity and healthy adolescents. Obesity, 2015; 23:2183–2189.

15.

Snoek

, van Strien

, Janssens

, Engels

. Restrained eating and BMI: A longitudinal study among adolescents. Health Psychol, 2008; 27:753.

16.

Kubiak

, Vögele

, Siering

, et al. Daily hassles and emotional eating in obese adolescents under restricted dietary conditions—The role of ruminative thinking. Appetite, 2008; 51:206–209.

17.

Roemmich

, Wright

, Epstein

. Dietary restraint and stress-induced snacking in youth. Obes Res, 2002; 10:1120–1126.

18.

Lemmens

, Rutters

, Born

, Westerterp-Plantenga

. Stress augments food ‘wanting’ and energy intake in visceral overweight subjects in the absence of hunger. Physiol Behav, 2011; 103:157–163.

19.

Herhaus

, Ullmann

, Chrousos

, Petrowski

. High/low cortisol reactivity and food intake in people with obesity and healthy weight. Transl Psychiatry, 2020; 10:1–8.

20.

Nagy

, Gill

, Adams

, et al. Stress-induced suppression of food intake in overweight and obese adolescents. Psychosom Med, 2019; 81:814–820.

21.

Appelhans

, Pagoto

, Peters

, Spring

. HPA axis response to stress predicts short-term snack intake in obese women. Appetite, 2010; 54:217–220.

22.

Pereira

, Alvarenga

. Disordered eating: Identifying, treating, preventing, and differentiating it from eating disorders. Diabetes Spectr, 2007; 20:141–148.

23.

Goldschmidt

, Aspen

, Sinton

, et al. Disordered eating attitudes and behaviors in overweight youth. Obesity, 2008; 16:257–264.

24.

Neumark-Sztainer

, Wall

, Haines

, et al. Shared risk and protective factors for overweight and disordered eating in adolescents. Am J Prev Med, 2007; 33:359–369.

25.

Field

, Austin

, Taylor

, et al. Relation between dieting and weight change among preadolescents and adolescents. Pediatrics, 2003; 112:900–906.

26.

Stice

, Presnell

, Shaw

, Rohde

. Psychological and behavioral risk factors for obesity onset in adolescent girls: A prospective study. J Consult Clin Psychol, 2005; 73:195.

27.

Tanofsky-Kraff

, Cohen

, Yanovski

, et al. A prospective study of psychological predictors of body fat gain among children at high risk for adult obesity. Pediatrics, 2006; 117:1203–1209.

28.

Libbey

, Story

, Neumark-Sztainer

, Boutelle

. Teasing, disordered eating behaviors, and psychological morbidities among overweight adolescents. Obesity, 2008; 16(S2):S24–S29.

29.

Ajibewa

, Zhou

, Barry

, et al. Adolescent stress: A predictor of dieting behaviors in youth with overweight/obesity. Appetite, 2020; 147:104560.

30.

Nelson

, Gerras

, McGlumphy

, et al. Racial discrimination and low household education predict higher body mass index in African American youth. Childhood Obes, 2018; 14:114–121.

31.

Rockett

HRH

, Wolf

, Colditz

. Development and reproducibility of a food frequency questionnaire to assess diets of older children and adolescents. J Am Diet Assoc, 1995; 95:336–340.

32.

Rockett

HRH

, Breitenbach

, Frazier

, et al. Validation of a youth/adolescent food frequency questionnaire. Prev Med, 1997; 26:808–816.

33.

Þórarinsdóttir

, Faurholt-Jepsen

, Ullum

, et al. The validity of daily self-assessed perceived stress measured using Smartphones in healthy individuals: Cohort study. JMIR Mhealth Uhealth, 2019; 7. [Epub ahead of print]; DOI: 10.2196/13418.

34.

Brantley

, Waggoner

, Jones

, Rappaport

. A daily stress inventory: Development, reliability, and validity. J Behav Med, 1987; 10:61–73.

35.

Moskow

, Addington

, Bearden

, et al. The relations of age and pubertal development with cortisol and daily stress in youth at clinical risk for psychosis. Schizophr Res, 2016; 172:29–34.

36.

Tessner

, Mittal

, Walker

. Longitudinal study of stressful life events and daily stressors among adolescents at high risk for psychotic disorders. Schizophr Bull, 2011; 37:432–441.

37.

Harris

, Flexeder

, Thiering

, et al. Changes in dietary intake during puberty and their determinants: Results from the GINIplus birth cohort study. BMC Public Health, 2015; 15. [Epub ahead of print]; DOI: 10.1186/s12889-015-2189-0.

38.

Laugero

, Falcon

, Tucker

. Relationship between perceived stress and dietary and activity patterns in older adults participating in the Boston Puerto Rican Health Study. Appetite, 2011; 56:194–204.

39.

Xie

, Gilliland

, Li

Y-F

, Rockett

. Effects of ethnicity, family income, and education on dietary intake among adolescents. Prev Med, 2003; 36:30–40.

40.

Petersen

, Crockett

, Richards

, Boxer

. A self-report measure of pubertal status: Reliability, validity, and initial norms. J Youth Adolesc, 1988; 17:117–133.

41.

Garner

, Olmsted

, Bohr

, Garfinkel

. The Eating Attitudes Test: Psychometric features and clinical correlates. Psychol Med, 1982; 12:871–878.

42.

McEwen

, Flouri

. Fathers' parenting, adverse life events, and adolescents' emotional and eating disorder symptoms: The role of emotion regulation. Eur Child Adolesc Psychiatry, 2009; 18:206–216.

43.

Attie

, Brooks-Gunn

. Development of eating problems in adolescent girls: A longitudinal study. Dev Psychol, 1989; 25:70–79.

44.

Edwards

, Walk

, Thompson

, et al. Disordered eating attitudes and behavioral and neuroelectric indices of cognitive flexibility in individuals with overweight and obesity. Nutrients, 2018; 10:1902.

45.

Naggara

, Raymond

, Guilbert

, et al. Analysis by categorizing or dichotomizing continuous variables is inadvisable: An example from the natural history of unruptured aneurysms. Am J Neuroradiol, 2011; 32:437–440.

46.

Royston

, Altman

, Sauerbrei

. Dichotomizing continuous predictors in multiple regression: A bad idea. Stat Med, 2006; 25:127–141.

47.

Ervin

, Ogden

. Trends in intake of energy and macronutrients in children and adolescents from 1999–2000 through 2009–2010. NCHS Data Brief, 2013; 113:1–8.

48.

McGlumphy

, Shaver

, Ajibewa

, Hasson

. Perceived stress predicts lower physical activity in African-American boys, but not girls. Am J Health Behav, 2018; 42:93–105.

49.

Zerwas

, Holle

, Watson

, et al. Childhood anxiety trajectories and adolescent disordered eating: Findings from the NICHD study of early child care and youth development. Int J Eating Disord, 2014; 47:784–792.

50.

Cartwright

, Wardle

, Steggles

, et al. Stress and dietary practices in adolescents. Health Psychol, 2003; 22:362.

51.

Jeong

E-Y

, Kim

K-N

. Influence of stress on snack consumption in middle school girls. Nutr Res Pract, 2007; 1:349–355.

52.

Kim

, Yang

, Kim

A-J

, Lim

. Academic stress levels were positively associated with sweet food consumption among Korean high-school students. Nutrition, 2013; 29:213–218.

53.

Michaud

, Kahn

, Musse

, et al. Relationships between a critical life event and eating behaviour in high-school students. Stress Med, 1990; 6:57–64.

54.

Torres

, Nowson

. Relationship between stress, eating behavior, and obesity. Nutrition, 2007; 23:887–894.

55.

Epel

, Tomiyama

, Dallman

. Stress and reward. In: KD Brownell

Gold (eds.), Food and Addiction: A Comprehensive Handbook. 2012, Oxford University Press, Oxford, United Kingdom, p. 266.

56.

Wiium

, Breivik

, Wold

. Growth trajectories of health behaviors from adolescence through young adulthood. Int J Environ Res Public Health, 2015; 12:13711–13729.

57.

Major

, Eliezer

, Rieck

. The psychological weight of weight stigma. Soc Psychol Pers Sci, 2012; 3:651–658.

58.

Goldschmidt

, O'Brien

, Lavender

, Pearson

, Le Grange

, Hunter

. Executive functioning in a racially diverse sample of children who are overweight and at risk for eating disorders. Appetite, 2018; 124:43–49.

59.

Duckworth

, Kim

, Tsukayama

. Life stress impairs self-control in early adolescence. Front Psychol, 2013; 3. [Epub ahead of print]; DOI: 10.3389/fpsyg.2012.00608.

60.

Miller

, Gearhardt

, Fredericks

, et al. Targeting self-regulation to promote health behaviors in children. Behav Res Ther, 2018; 101:71–81.

61.

Williams

. Stress and the mental health of populations of color: Advancing our understanding of race-related stressors. J Health Soc Behav, 2018; 59:466–485.

62.

Araujo

, Yokoo

, Pereira

. Validation and calibration of a semiquantitative food frequency questionnaire designed for adolescents. J Am Dietetic Assoc, 2010; 110:1170–1177.

63.

Slater

, Philippi

, Fisberg

, Latorre

MRDO

. Validation of a semi-quantitative adolescent food frequency questionnaire applied at a public school in São Paulo, Brazil. Eur J Clin Nutr, 2003; 57:629–635.

64.

Adam

, Epel

. Stress, eating and the reward system. Physiol Behav, 2007; 91:449–458.

65.

Hill

, Peters

, Wyatt

. Using the energy gap to address obesity: A commentary. J Am Dietetic Assoc, 2009; 109:1848–1853.

66.

Sundgot-Borgen

, Torstveit

. Aspects of disordered eating continuum in elite high-intensity sports. Scand J Med Sci Sports, 2010; 20(s2):112–121.

67.

Bor

, Dean

, Najman

, Hayatbakhsh

. Are child and adolescent mental health problems increasing in the 21st century? A systematic review. Aust N Z J Psychiatry, 2014; 48:606–616.

68.

de Vet

, de Wit

, Luszczynska

, et al. Access to excess: How do adolescents deal with unhealthy foods in their environment?. Eur J Public Health, 2013; 23:752–756.

69.

Puhl

, Himmelstein

, Gorin

, Suh

. Missing the target: Including perspectives of women with overweight and obesity to inform stigma-reduction strategies. Obes Sci Pract, 2017; 3:25–35.