Adverse experiences as predictors of maladaptive and adaptive eating: Findings from EAT 2018

Abstract

Adverse experiences, such as childhood abuse and other violence victimization, are associated with problematic eating. However, whether different types of adversity relate to both maladaptive and adaptive eating behaviors is unclear. This study examined the associations of different adverse experiences with maladaptive (i.e., overeating and binge eating) and adaptive (i.e., intuitive eating and mindful eating) eating by gender. Data were derived from the EAT-2018 (Eating and Activity over Time) study (N=1411, aged 18-30 years in 2017-2018). Modified Poisson regressions were used to examine the associations between adverse experiences and the prevalence of maladaptive eating. Linear regressions were used to examine the associations between adverse experiences and adaptive eating scores.

Each adverse experience was associated with greater prevalence of maladaptive eating and lower adaptive eating scores. Among women, intimate partner sexual violence was strongly associated with more overeating (PR=2.1 [95% CI=1.4-3.1]) and binge eating (PR=2.4 [95% CI=1.5-3.9]), and less mindful eating (β=−0.6, [95% CI= −0.8, −0.3]); being attacked, beaten, or mugged was most associated with less intuitive eating (β=−0.5, [95% CI= −0.8, −0.2]). Among men, being attacked, beaten, or mugged was strongly associated with more overeating (PR= 2.1 [95% CI=1.2-3.5]) and binge eating (PR=3.2 [95% CI=1.6-6.5]); intimate partner physical violence was strongly associated with less intuitive eating (β=−0.6, [95% CI= −0.9, −0.2]); childhood emotional abuse was strongly associated with less mindful eating (β=−0.8, [95% CI= −1.0, −0.5]). To improve eating behaviors, adverse life experiences and the potential impact on maladaptive and adaptive eating should be considered.

1. Introduction

Eating behaviors fall on a continuum ranging from adaptive to maladaptive eating. Overeating (eating an unusually large quantity of food) and binge eating (overeating with a sense of loss of control) exist on the maladaptive eating side of the spectrum and are characterized by reliance on external or emotional cues to determine when, what, and how much to eat. Overeating and binge eating are public health concerns given their association with weight-related health outcomes [1-4] and psychological well-being [5-7]. Adaptive eating, such as intuitive eating and mindful eating, has recently gained attention as a potential way to improve nutrition and health outcomes [8-9-18-10-17]. Both intuitive and mindful eating share common features, such as reliance on internal physiological hunger and satiety cues rather than emotional cues to eat [19-20]. Despite these similarities, intuitive eating and mindful eating differ in important ways. Intuitive eating involves giving oneself permission to eat desired foods when hungry and being attuned to the physical feelings of the body to determine what, when, and how much to eat [19-21-22]. Meanwhile, mindful eating emphasizes that individuals are aware of the present moment when eating, pay attention to the effect of food on their senses, and notice the physical and emotional sensations occurring in response to eating [20].

Given the impacts of eating behaviors on health and well-being [1-2-11-18-3-10], there is a need to identify people who are at greatest risk of overeating and binge eating, and those who struggle the most to adopt adaptive eating, such as intuitive and mindful eating. Adverse experiences such as exposure to violence have been shown to predict overeating and binge eating [23-32]. Studies suggest that adverse experiences may lead survivors to dissociate from negative emotional states [33-36] by engaging in maladaptive eating such as binge eating [26-29-31-37]. However, the spectrum of adverse experiences is broad, ranging from child abuse to witnessing community or interpersonal violence, and occur in various contexts (e.g., family and work), and there may be heterogeneity in the impacts of different adversities. Thus far, research has largely been compartmentalized, focusing on a single domain of adversities at a time. For example, most studies of adverse experiences and eating behaviors have been conducted on adversities known as “adverse childhood experiences” (ACEs) and on intimate partner violence (IPV). Within the domains of ACEs, childhood sexual and physical abuse have been identified as risk factors for maladaptive eating [26-29-31-37]. Other types of ACEs, such as childhood emotional abuse [25-26-38], household substance abuse, household mental illness, and household incarceration [39-40], have received far less attention. Furthermore, only a few studies have examined the association of other adverse experiences, such as witnessing violence [38-41-42], with maladaptive eating.

Compared to the large body of literature that has examined adverse experiences as predictors of maladaptive eating, the association between adverse experiences and adaptive eating has attracted substantially less attention. Although a few studies have examined predictors of adaptive eating (i.e., intuitive eating and mindful eating) [43-45], to the best of our knowledge, no study has explicitly examined adverse experiences as predictors of intuitive eating or mindful eating. In addition to the specific gaps in the literature on maladaptive and adaptive eating, research on adverse experiences and maladaptive eating has often been exclusive to women [46-48] or to men [49-50] or included both women and men but have not reported gender-stratified results [28-30-51]. Therefore, whether and how the association between adverse experiences and maladaptive and adaptive eating differs by gender remains unclear.

Building upon existing studies that have examined the association between adverse experiences and eating behaviors, the primary goal of this study was to comprehensively examine the associations between a broad range of adverse experiences and eating behaviors. To this end, the current study aimed to (1) describe the prevalence of a broad range of adverse experiences, including ACEs, IPV, and experience or witnessing of interpersonal violence; (2) examine the extent to which adverse experiences are associated with overeating, binge eating, intuitive eating, and mindful eating; and (3) explore whether these associations differ by gender in a large population-based sample of emerging adults. In this study, we hypothesized that adverse experiences would be associated with increased risks of overeating and binge eating and lower intuitive and mindful eating scores. Based on the prior literature [27-31-37-42-50], we hypothesized that there would be gender differences in the association between adversities and eating behavior outcomes, with adversities being more strongly related to eating behaviors in women than in men.

2. Methods

2.1. Study Design and Population

Eating and Activity over Time (EAT-2018) is the follow-up to Eating and Activity in Teens (EAT-2010), a population-based study of eating patterns, physical activity, weight-related behaviors, and weight status of 2793 middle and high school students in 20 urban public schools in Minneapolis and St Paul, Minnesota [52-54]. In the current analysis, participants were restricted to those who completed EAT 2018 (N=1568). Participants who were missing information regarding adverse experiences (n=70), eating behaviors (n=31) or covariates (n=56) at EAT 2018 were further excluded, leaving a final analytic sample of 1411 (827 women and 584 men) emerging young adults in 2017-2018. The analytic sample in this study represents 90% of the 1568 participants who responded to EAT 2018. Because attrition from EAT 2010 to EAT 2018 did not occur completely at random, inverse probability weighting (IPW) was used in all analyses to account for missing data [55-56]. IPW minimizes potential response bias due to missing data and enables extrapolation to the original EAT 2010 school-based sample. Weights for IPW were derived as the inverse of the estimated probability that an individual responded at the two time points based on several characteristics reported in 2010, including demographics, past year frequency of dieting, and weight status. In the weighted analytic sample in this study, the mean age was 22 (SD=2.0) years and 54% were women, and the ethnicity/race distribution was 20% White, 28% Black or African American, 20% Asian American, 17% Hispanic, 3% Native American, and 12% mixed or other. All study protocols were approved by the University of Minnesota’s Institutional Review Board Human Subjects Committee.

2.2. EAT Survey Development

Development of the surveys were guided by previous Project EAT surveys [57], social cognitive theory [58], experts from various disciplines, and extensive pilot testing. Test-retest reliability of the measures was examined among 129 adolescents over a week and among 112 young adults over three weeks.

2.3. Measurements

2.3.1. Exposures

2.3.1.1. Adverse childhood experiences

Adverse childhood experiences were assessed at EAT 2018 by asking the participants about their experiences of abuse (sexual, physical, and emotional) and household dysfunction during childhood (<18 years). Childhood sexual, physical, and emotional abuse items were adapted from the Childhood Trauma Questionnaire [59]. Household dysfunction items were adapted from the Adverse Childhood Experiences Study [39-40]. To maximize each adverse childhood experiences items’ sensitivity and specificity, as recommended by others [60], various cutoff points were chosen to dichotomize each type of adverse childhood experiences.

2.3.1.1.1. Childhood sexual abuse

Childhood sexual abuse experienced before age 18 was assessed with two items “Did someone in your family member touch you in a sexual way against your wishes or force you to touch them in a sexual way?” (Test-retest % agreement =95% at EAT 2018) and “Did someone not in your family touch you in a sexual way against your wishes or force you to touch them in a sexual way?” (Test-retest % agreement =90% at EAT 2018). Response options for each item used to assess childhood sexual abuse were “yes” or “no.”

2.3.1.1.2. Childhood physical abuse

Childhood physical abuse was assessed with the item “An adult in my family hit me so hard it left me with bruises or marks.” Response options for childhood physical abuse were “never,” “rarely,” “sometimes,” “often,” and “very often.” Childhood physical abuse was dichotomized into “never” or “ever.” Test-retest % agreement = 96% at EAT 2018.

2.3.1.1.3. Childhood emotional abuse

Childhood emotional abuse experienced before age 18 was assessed with the item “an adult in my family said hurtful or insulting things to me.” Response options were “never,” rarely,” “sometimes,” “often,” and “very often.” Participants who responded that an adult in their family said hurtful or insulting things “often” or more were categorized as experience of childhood emotional abuse (Test-retest r=.77 at EAT 2018).

2.3.1.1.3. Household dysfunction

Household dysfunction was assessed with three items, “Prior to age 18, did you live with a family member who was a problem drinker or alcoholic, who used street drugs, or who abused prescription drugs (Test-retest % agreement = 87% at EAT 2018),” “live with a household member who was depressed, mentally ill, or attempted suicide (Test-retest % agreement = 81% at EAT 2018),” and “live with a household member went to prison (Test-retest % agreement = 94% at EAT 2018).” Response options for each item were “yes” or “no.”

2.3.1.2. Intimate Partner Violence

Intimate partner violence was retrospectively assessed at EAT 2018. Intimate partner sexual violence and intimate partner physical violence items were adapted and modified from EAT-II (Eating Among Teens-II) Study survey [37].

2.3.1.2.1. Intimate Partner Sexual Violence

Intimate partner sexual violence was assessed with the item “Have you been forced to touch a dating partner or spouse sexually or had some type of sexual behavior forced on you.” Response options for intimate partner sexual violence item were “yes, in the past year,” “yes, more than a year ago,” and “no”. Intimate partner sexual violence was categorized as “yes” or “no” for easier interpretation of the data. Test-retest % agreement = 93% at EAT 2018.

2.3.1.2.2. Intimate Partner Physical Violence

Intimate partner physical violence was assessed with the item “Have you been hit shoved, held down, or had some other physical forced used against you by a spouse or someone you were dating.” Response options for intimate partner physical violence item were “yes, in the past year,” “yes, more than a year ago,” and “no”. Intimate partner physical violence was categorized as “yes” or “no” for easier interpretation of the data. Test-retest % agreement = 89% at EAT 2018.

2.3.1.3. Other adverse experiences

Other adverse experiences items were adapted and modified from the Life Events Questionnaire [61] and Brief Trauma Questionnaire [62]. Items used to assess other adverse experiences included “had problems with the police (Test-retest % agreement =92% at EAT 2018),” “had been attacked, beaten, or mugged (Test-retest % not available due to lack of power),” “had a close family member or friend die violently for example in a serious car crash, mugging, homicide, or suicide (Test-retest % agreement = 89% at EAT 2018),” and “witnessed a situation in which someone was seriously injured or killed, or in which you feared someone would be seriously injured or killed (Test-retest % agreement = 90% at EAT 2018).” Response options for each item were “yes, in the past year,” “yes, more than a year ago,” and “no.” In this study, each item was categorized as “yes” or “no” for easier interpretation of the data.

2.3.2. Outcomes

2.3.2.1. Maladaptive Eating Behaviors

2.3.2.1.1. Overeating

Overeating was assessed by the item “In the past year, have you ever eaten so much food in a short period of time that you would be embarrassed if others saw you (binge eating?). Available response options were “yes” or “no.” (Test-retest % agreement = 90% at EAT 2018).

2.3.2.1.2. Binge eating

Binge eating was assessed among participants who endorsed overeating by subsequently asking, “During the times when you ate this way, did you feel you couldn’t stop eating or control what or how much you were eating?” Response options were “yes” or “no.” (Test-retest percent k=.58 at EAT 2018)

2.3.2.2. Adaptive Eating Behaviors

2.3.2.2.1. Intuitive Eating

Adaptive eating behaviors included intuitive eating and mindful eating. Intuitive eating was assessed with three items adapted from the Intuitive Eating Scale [22], including “I stop eating when I feel full,” “I eat everything that is on my plate, even if I’m not that hungry,” and “I trust my body to tell me how much to eat.” Response options were “hardly ever,” “sometimes,” “much of the time,” “almost always” and were assigned 0-3 points, respectively. The second item, “I eat everything that is on my plate even if I’m not hungry” was reverse scored. A summary score was created for intuitive eating (0-9 points), with a higher score indicating higher level of intuitive eating (McDonald’s ω = .56 at EAT 2018 and test-retest r=.57 at EAT 2018).

2.3.2.2.2. Mindful Eating

Mindful eating was assessed with four items adapted from the Mindful Eating Questionnaire [20] including “I eat so quickly that I don’t taste what I’m eating,” “I snack without noticing that I am eating,” “Before I eat I take a moment to appreciate the colors and smells of my food,” and “I taste every bite of food that I eat.” Response options were “hardly ever,” “sometimes,” “much of the time,” “almost always” and were assigned 0 to 3 points respectively. The first and second item, I eat so quickly that I don’t taste what I’m eating” and “I snack without noticing that I am eating” were reverse scored. A summary score was created for mindful eating (0-12 points), with a higher score indicating higher level of mindful eating (McDonald’s ω = .50 at EAT 2018 and test-retest r=.67 at EAT 2018).

2.3.3. Covariates

Age (based on the date of birth), ethnicity/race, and family socioeconomic status were self-reported in EAT 2010 and were considered potential confounders of the association between adversities and eating behaviors. Ethnicity/race was assessed at baseline with the question, “Do you think of yourself as (1) White, (2) Black or African American, (3) Hispanic or Latino, (4) Asian American, (5) Native Hawaiian or Pacific Islander, (6) American Indian or Native American, or (7) Other.” The very few participants who self-described themselves as “Hawaiian or Pacific Islander” or did not report their ethnicity/race were coded as “mixed/other.” The categorization of SES was determined at baseline primarily based on the highest education level of either parent with adjustments made for student eligibility for free/reduced price school meals, family receipt of public assistance, and parents’ employment status (Test-retest % agreement =98-100%).

2.4. Statistical Analysis

Demographic characteristics of the sample are presented as descriptive statistics. T-tests and chi-square tests were conducted to examine the differences between participants with and without adverse experiences. To examine the associations between each type of adverse experience with maladaptive eating (i.e., overeating and binge eating), modified Poisson regressions [63] were performed to estimate prevalence ratios (PRs) comparing the prevalence of overeating and binge eating in those with versus without adverse experience. A PR of 1 indicates the null (no association). Ninety-five percent confidence intervals (CIs) are presented for each estimate to show its precision and statistical compatibility with the null hypothesis. To examine associations between each type of adverse experience and adaptive eating (i.e, intuitive eating and mindful eating), multivariable linear regressions were conducted, with β indicating the difference in average score of those with a particular type of adversity versus those without. A β coefficient of 0 indicates the null (no association). Sensitivity analyses were conducted to examine dose-response associations of cumulative adverse experiences of ACEs and IPV with maladaptive and adaptive eating. In the sensitivity analyses, maladaptive and adaptive eating were each regressed on the number of exposures to ACEs and IPV and the number of overall adverse experiences (ACEs and IPV).

Preliminary analyses indicated that tests for interaction between adverse experiences and gender on eating behaviors were nonsignificant (p_interaction =.05-.97). However, due to prior evidence for differences in associations of adverse experiences with eating behaviors by gender [27-31-37-42-50], we present gender-stratified results to allow qualitative comparisons between men and women. All models were adjusted for age, ethnicity/race, and family socioeconomic status, and weighted by nonresponse propensity to reflect the original EAT 2010 sample population [55-56]. All analyses were performed using SAS 9.4 (SAS Institute Inc., Cary, NC).

3. Results

3.1. Prevalence of Each Adverse Experience

As shown in Table 1, 65% (n=222) of white participants, 57% (n=177) of Black/African American participants, 57% (n=141) of Hispanic/Latinx participants, 39% (n=124) of Asian American participants, 75% (n=6) of Native American participants, and 71% (n=37) of participants of mixed or other race/ethnicity reported experiencing one or more adverse experiences.

Table 1.

Demographic characteristics of the analytic sample (N=1411)

	Any adverse experiences
	Ever N=791 (56%)	Never N=620 (44%)	All participants N= 1411	P value
Age in years M±SD	22.1±2.0	22.0±2.0	22.0±2.0	0.3
Race n (row %)
White	222 (65)	119 (35)	341 (20)	<0.001
Black/African American	177 (57)	128 (43)	305 (28)
Hispanic/Latinx	141 (57)	106 (43)	247 (17)
Asian American	124 (39)	196 (61)	320 (20)
Native American	6 (75)	2 (25)	8 (1)
Mixed/other	37 (71)	15 (29)	52 (3)
Gender n (row %)	54 (61)	84 (39)	138 (11)
Women	472 (57)	355 (43)	827 (54)	0.3
Men	319 (54)	265 (46)	584 (46)
SES n (row %)
Low	288 (56)	227 (44)	515 (39)	0.9
Low-middle	168 (54)	142 (46)	310 (22)
Middle	130 (55)	105 (45)	235 (18)
Middle-high	130 (59)	94 (41)	224 (13)
High	75 (58)	52 (43)	127 (8)

Note: % weighted to reflect the original EAT 2010 participants

Regarding gender, 57% (n=472) of women and 54% (n=319) of men reported experiencing one or more adverse experiences. Across family socioeconomic statuses, the prevalence of experience of adverse experiences ranged from 54% (n=168) among participants with a low-middle family socioeconomic status to 59% (n=130) among participants with a middle-high family socioeconomic status.

The two types of adverse experiences with the highest prevalence were living with a family member with a mental illness (29% and 24% of the women and men, respectively) and living with a family member with a substance use disorder (23% and 21% of the women and men, respectively; Table 2). Childhood sexual abuse, emotional abuse, living with an incarcerated household member, IPV, and violent death of a close family member or friend were more prevalent among the women (p =0.04), whereas other adverse experiences (i.e., problems with police or a history of being attacked, beaten, or mugged) were more prevalent among the men (p <0.01).

Table 2.

Prevalence of each adverse experience by gender

Column n (%)	Women N=827	Men N=584	P value
Any adverse experience	472 (57)	319 (55)	<0.001
Childhood sexual abuse	172 (21)	34 (6)	<0.001
Childhood physical abuse	139 (17)	89 (15)	0.4
Childhood emotional abuse	130 (18)	49 (8)	<0.001
Household substance abuse	184 (23)	121 (21)	0.1
Mental illness in household	241 (29)	137 (24)	0.1
Incarcerated household member	106 (13)	51 (9)	0.1
Intimate partner sexual violence	74 (9)	15 (3)	<0.001
Intimate partner physical violence	89 (11)	23 (4)	<0.001
Problems with police	75 (9)	112 (19)	<0.001
Attacked, beaten or mugged	43 (5)	51 (9)	<0.001
Violent death of a close family member or friend	148 (18)	81 (14)	0.1
Witnessing a serious injury or death	84 (10)	69 (12)	0.3

Note: % weighted to reflect the original EAT 2010 participants

3.2. Associations between Adverse Experiences and Maladaptive Eating

3.2.1. Overeating

Among women, nearly all types of adverse experiences, except for household substance abuse and having problems with police, were associated with greater prevalence of overeating, after adjustment for sociodemographic variables, although not all associations were statistically significant. Of the predictors, intimate partner sexual violence was most strongly associated with overeating (PR= 2.1; 95% CI: 1.4-3.1) (Table 3). Analyses of cumulative exposure to adversities were suggestive of a dose-response relationship between multiple IPV exposures and overeating (i.e., the greater the number of exposures to IPV, the greater prevalence of overeating). Cumulative ACEs, and cumulative number of all adversities, were not associated with overeating in women in a dose-response manner.

Table 3.

Maladaptive and adaptive eating by adverse experiences among women (N=827)

	Maladaptive eating		Adaptive eating
	Overeating	Binge eating	Intuitive eating	Mindful eating
	PR (95% CI)		β (95% CI)
Any adverse experience (n=472)	1.4 (0.8-2.4)	2.2 (1.5-3.3)	−0.3 (−0.4, −0.1)	−0.2 (−0.3, −0.1)
Childhood sexual abuse (n=172)	1.5 (1.1-2.1)	1.6 (1.1-2.3)	−0.2 (−0.3, −0.1)	0.1 (−0.1, 0.2)
Childhood physical abuse (n=139)	1.3 (0.9-1.9)	1.6 (1.1-2.5)	−0.4 (−0.6, −0.23)	−0.3 (−0.5, −0.1)
Childhood emotional abuse (n=130)	1.4 (1.0-2.0)	1.6 (1.1-2. 5)	−0.3 (−0.5, −0.2)	−0.1 (−0.2, 0.2)
Household substance abuse (n=184)	1.2 (0.8-1.7)	1.2 (0.8-1.8)	0.01 (−0.1, 0.2)	0.1 (−0.2, 0.12)
Mental illness in household (n=241)	1.4 (1.1-2.0)	1. 5 (1.0-2.1)	−0.1 (−0.3, −0.1)	−0.1 (−0.3, 0. 1)
Incarcerated household member (n=106)	1.3 (0.9-1.9)	1.0 (0.6-1.6)	−0.2 (−0.4, −0.1)	0.1 (−0.1, 0.3)
Intimate partner sexual violence (n=74)	2.1 (1.4-3.1)	2.4 (1.5-3.9)	−0.4 (−0. 6, −0.1)	−0.6 (−0.8, −0.3)
Intimate partner physical violence (n=89)	1.4 (0.9-2.1)	1. 7 (1.1-2.7)	0.1 (−0.2, 0.2)	−0.2 (−0.4, −0.1)
Problems with police (n=75)	1.2 (0.7-1.9)	1.4 (0.8-2.4)	0.01 (−0.2, 0.2)	−0.3 (−0.6, −0.1)
Attacked, beaten or mugged (n=43)	1.9 (1.1-3.0)	2.2 (1.3-3.4)	−0.5 (−0.8, −0.2)	−0.2 (−0.5, 0.1)
Violent death of a close family member or friend (n=148)	1.2 (0.9-1.7)	1.1 (0.7-1.7)	0.2 (0.1, 0.4)	0.4 (0.2, 0.6)
Witnessing a serious injury or death (n=84)	1.7 (1.1-2.5)	2.0 (1.2-3.1)	−0.3 (−0.5, −0.1)	−0.2 (−0.4, 0.1)

Note: PR = prevalence ratio; CI= confidence interval; Boldfaced values indicate significance

Bold denotes statistical significance

Models adjusted for age, race, and socioeconomic status.

Among men, fewer adverse experiences were associated with overeating. A history of being attacked, beaten, or mugged was strongly associated with overeating (PR= 2.1; 95% CI= 1.2-3.5). Additional predictors included childhood physical abuse, mental illness in the household, and a history of being attacked, beaten, or mugged (PR range= 1.3-1.8) (Table 4). Sensitivity analyses did not indicate dose-response associations between any type of adverse experiences and overeating.

Table 4.

Maladaptive and adaptive eating by adverse experiences among men (N=584)

	Maladaptive eating		Adaptive eating
	Overeating	Binge eating	Intuitive eating	Mindful eating
	PR (95% CI)		β (95% CI)
Any adverse experience (n=319)	1.2 (0.8-1. 8)	1.66 (0.9-3.1)	−0.2 (−0.3, −0.1)	−0.1 (−0.2, 0.1)
Childhood sexual abuse (n=34)	1.0 (0.4-2.4)	1.3 (0.4-4.1)	−0.3 (−0.6, 0.1)	−0.2 (−0.6, 0.1)
Childhood physical abuse (n=89)	1.7 (1.0-2.7)	2.8 (1.5-5.3)	−0.4 (−0.6, −0.2)	−0.1 (−0.3, 0.2)
Childhood emotional abuse (n=49)	1.8 (1.0-3.2)	2.5 (1.2-5.3)	−0.5 (−0.7, −0.3)	−0.8 (−1.0, −0.5)
Household substance abuse (n=121)	0.9 (0.5-1.5)	0.9 (0.4-2.0)	−0.1 (−0.2, 0.1)	−0.1 (−0.2, 0.2)
Mental illness in household (n=137)	1.7 (1.1-2.6)	2.5 (1.4-4.6)	−0.3 (−0.5, −0.2)	−0.2 (−0.4, −0.1)
Incarcerated household member (n=51)	1.2 (0.6-2.3)	1.0 (0.4-2.8)	−0.3 (−0.6, −0.1)	0.3 (0.1, 0.6)
Intimate partner sexual violence (n=15)	0.8 (0.2-3.1)	1.8 (0.4-7.4)	−0.3 (−0.7, 0.12)	0.1 (−0.5, 0.6)
Intimate partner physical violence (n=23)	1.6 (0.8-3.6)	3.1 (1.3-7.7)	−0.6 (−0.9, −0.2)	0.2 (−0.2, 0.6)
Problems with police (n=112)	1.2 (0.7-1.8)	1.0 (0.4-2.0)	−0.1 (−0.2, 0.1)	0.2 (−0.1, 0.4)
Attacked, beaten or mugged (n=51)	2.1 (1.2-3.5)	3.2 (1.6-6.5)	−0.4 (−0.7, −0.2)	0.2 (−0.1, 0.5)
Violent death of a close family member or friend (n=81)	1.3 (0.8-2.3)	1.6 (0.8-3.4)	−0.1 (−0.2, 0.2)	−0.2 (−0.5, 0.1)
Witnessing a serious injury or death (n=69)	1.5 (0.9-2.6)	2.3 (1.2-4.7)	−0.1 (−0.3, 0.1)	0.3 (−0.1, 0.5)

Note: PR = prevalence ratio; CI= confidence interval; Boldfaced values indicate significance

Bold denotes statistical significance

Models adjusted for age, race, and socioeconomic status.

3.2.2. Binge Eating

Among both men and women, the majority of the adverse experiences assessed were associated with greater prevalence of binge eating. Exceptions included living with a household member with a substance use disorder, living with an incarcerated household member, and having problems with police, the PRs for which were all close to the null for both men and women. Among women, intimate partner sexual violence was most strongly associated with binge eating (PR= 2.4; 95% CI= 1.5-3.9) (Table 3). Among men, being attacked, beaten or mugged was strongly associated with binge eating (PR= 3.2; 95% CI= 1.6-6.5) (Table 4). Among women, sensitivity analyses suggested dose-response associations between the number of exposures to IPV with binge eating. There was no indication of a dose-response association between ACEs or total adversities and binge eating (Table 5). Among men, cumulative experiences of ACEs and total cumulative adverse experiences each showed a dose-response association with binge eating, while cumulative IPV did not show dose-response association (Table 6).

Table 5.

Maladaptive and adaptive eating by cumulative adverse experiences among women (N=827)

	Maladaptive eating		Adaptive eating
	Overeating	Binge eating	Intuitive eating	Mindful eating
	PR (95% CI)		β (95% CI)
Cumulative adverse childhood experiences (ACEs)
0 (n=427)	1.0 (ref)	1.00 (ref)	0.0 (ref)	0.0 (ref)
1 (n=184)	1.4 (0.8-2.7)	1.9 (1.2-2.9)	−0.2 (−0.6, 0.1)	−0.3 (−0.7, 0.1)
≥ 2 (n=216)	1.3 (0.8-2.7)	1.9 (1.2-2.9)	−0.4 (−0.7, −0.1)	−0.2 (−0.5, 0.2)
Cumulative intimate partner violence (IPV)
0 (n=704)	1.0 (ref)	1.0 (ref)	0.0 (ref)	0.0 (ref)
1 (n=83)	0.5 (0.2-1.6)	1.3 (0.8-2.3)	0.1 (−0.4, 0.5)	0.1 (−0.4, 0.6)
2 (n=40)	1.8 (0.7-4.4)	2.8 (1.6-4.8)	−0.3 (−0.9, 0.2)	−0.9 (−1.5, −0.2)
Cumulative adverse experiences of ACEs and IPV
0 (n=401)	1.0 (ref)	1.0 (ref)	0.0 (ref)	0.0 (ref)
1 (n=178)	1.2 (0.6-2.4)	1.9 (1.2-3.1)	−0.4 (−0.7, −0.1)	−0.2 (−0.6, 0.1)
≥2 (n=248)	1.4 (0.8-2.5)	2.0 (1.3-3.1)	−0.4 (−0.7, −0.1)	−0.2 (−0.5, 0.1)

Note: Bold denotes statistical significance

Table 6.

Maladaptive and adaptive eating by cumulative adverse experiences among men (N=584)

	Maladaptive eating		Adaptive eating
	Overeating	Binge eating	Intuitive eating	Mindful eating
	PR (95% CI)		β (95% CI)
Cumulative adverse childhood experiences (ACEs)
0 (n=343)	1.0 (ref)	1.0 (ref)	0.0 (ref)	0.0 (ref)
1 (n=153)	1.5 (0.9-2.8)	1.1 (0.5-2.4)	−0.3 (−0.6, 0.1)	−0.1 (−1.4, 1.2)
≥ 2 (n=88)	0.9 (0.4-2.2)	2.6 (1.3-5.2)	−0.4 (−0.8, 0.0)	−0.2 (−0.7, 0.3)
Cumulative intimate partner violence (IPV)
0 (n=554)	1.0 (ref)	1. 0 (ref)	0.0 (ref)	0.0 (ref)
1 (n=22)	0.7 (0.1-3.7)	2.1 (0.7-6.5)	−0.8 (−1.5, −0.1)	−0.1 (−1.0, 0.9)
2 (n=8)	0.7 (0.1-3.7)	2.1 (0.7-6.5)	−0.8 (−1.5, −0.1)	−0.1 (−1.0, 0.9)
Cumulative adverse experiences of ACEs and IPV
0 (n=334)	1.0 (ref)	1.0 (ref)	0.0 (ref)	0.0 (ref)
1 (n=151)	1.4 (0.8-2.6)	1.1 (0.5-2.3)	−0.3 (−0.6, 0.1)	−0. 1 (−0.4, 0.4)
≥2 (n=99)	1.0 (0.4-2.2)	2.7 (1.4-5.4)	−0.4 (−0.8, −0.1)	−0.2 (−0.7, 0.2)

Note: Bold denotes statistical significance

3.3. Associations between Adverse Experiences and Adaptive Eating

3.3.1. Intuitive Eating

After adjustment for sociodemographic variables, compared to the participants without adverse experiences, parameter estimates indicated lower mean scores of intuitive eating for nearly all types of adverse experiences, although not all associations were significant. Among women, the experience of being attacked, beaten, or mugged was most strongly associated with less intuitive eating (β= −0.5; 95% CI= −0.8, −0.2). Additional statistically significant associations were found with childhood sexual, physical, and emotional abuse; mental illness in households; incarcerated household members; intimate partner sexual violence; a history of being attacked, beaten, or mugged; and witnessing a serious injury or death (Table 3). ACEs and IPV were each cumulatively associated with lower intuitive scores (i.e., the greater the number of exposures to ACEs and IPV, the lower the intuitive eating score) (Table 5).

Among men, intimate partner physical violence was most strongly associated with less intuitive eating (β= −0.6; 95% CI= −0.9, −0.2). Additional statistically significant associations were found with childhood physical and emotional abuse, mental illness in households, incarcerated household members, and history of being attacked, beaten, or mugged (Table 4). Sensitivity analyses indicated dose-response associations of the number of ACEs and total adverse experiences (i.e., combination of ACEs and IPV) with lower scores of intuitive eating (Table 6).

3.3.2. Mindful Eating

Among women, intimate partner sexual violence was most strongly associated with less mindful eating (β =−0.6; 95% CI= −0.8, −0.3). Additional statistically significant associations were found with childhood emotional abuse, intimate partner physical violence, and problems with police (Table 3). Sensitivity analyses indicated a dose-response association between IPV and lower scores of mindful eating, but not for other adversities (Table 5).

Among men, childhood emotional abuse was strongly and significantly associated with mindful eating (β= −0.8; 95% CI= −1.0, −0.5) (Table 4). An additional significant association was found with living with a household member with a mental illness. Sensitivity analyses indicated dose-response associations of the number of ACEs and total adverse experience (i.e., a combination of ACEs and IPV) with lower scores of mindful eating (Table 6).

4. Discussion

The aims of the current study were to: comprehensively assess the prevalence of a range of adverse experiences, including ACEs, IPV, and experiencing or witnessing other adversities; to examine the extent to which these adverse experiences were related to maladaptive eating (i.e., overeating and binge eating) and adaptive eating (i.e., intuitive and mindful eating); and explore whether the associations differed by gender. We present three main findings. First, adverse experiences were found to be prevalent among emerging adults. In particular, living with a family member with a mental illness and living with a family member with a substance use disorder were the two most prevalent types of adverse experiences. Second, various adverse experiences were associated with greater maladaptive eating and lower adaptive eating scores. Third, the extent to which adverse experiences were associated with maladaptive and adaptive eating qualitatively differed between women and men. Among women, intimate partner sexual violence was strongly predictive of both maladaptive and adaptive eating. Among men, a history of being attacked, beaten, or mugged was strongly predictive of overeating and binge eating, while childhood emotional abuse was strongly predictive of adaptive eating.

Regarding the prevalence of adverse experiences in this study, 56% of the emerging adults had experienced one or more adverse experiences; specifically, 45% of the participants had experienced one or more ACEs (child maltreatment or household dysfunction), 11% had experienced IPV (either sexual or physical), and 36% had experienced other adverse experiences. Although nearly half of the participants reported a history of ACEs, this prevalence is lower than the estimates reported in the original ACE study [39] and other reports [64]. The lower prevalence of ACEs may be due to the more limited number of items and types of adverse experiences included and assessed in our study (e.g., our study did not assess parental separation/divorce and child witnessing domestic violence against their mother). Furthermore, the 11% of the participants in our study who reported experience of IPV is lower than the estimates reported in a national intimate partner and sexual violence survey among adults [65]. The younger age of the participants in our study may have resulted in a difference in estimating the lifetime prevalence of IPV. However, the 36% of the participants in our study who experienced other types of interpersonal violence was comparable to a study that reported that 38% of the adolescents had witnessed any type of interpersonal violence [66].

Our findings indicate that the direction of the associations of adverse experiences with maladaptive eating were similar across gender. However, patterns of the associations qualitatively differed by gender. For instance, among women, intimate partner sexual violence was the type of adverse experience most strongly associated with overeating and binge eating. This finding is consistent with studies that have shown IPV and sexual violence to be associated with binge eating [31-37] and eating disorders related to binge eating [25]. Among men, childhood physical abuse was a suggestive predictor of overeating and binge eating in our study, which concurs with prior literature [27-31-37]. We further add to the literature by suggesting being attacked, beaten, and mugged, another form of physical violence that has received less attention, as a predictor of overeating and binge eating among men. Interestingly, most adverse experiences were more strongly associated with binge eating than overeating. Given that sense of loss of control is the central characteristic of binge eating, the stronger association with binge eating suggests that adverse experiences may specifically affect survivors’ perceived ability to control what and how much they eat.

Similar to maladaptive eating, patterns of the associations of adverse experiences with adaptive eating qualitatively differed by gender. Intimate partner violence (sexual and physical for women and men, respectively), childhood emotional abuse in men, and history of being attacked, beaten, and mugged in women were strongly associated with lower levels of adaptive eating. This finding of intimate partner violence, child maltreatment, and other interpersonal violence as suggestive risk factors for lower levels of adaptive eating in both women and men extends studies that have explored factors for higher adaptive eating scores and have found nonspecific perceived stress to be a predictor [43-45]. To the best of our knowledge, no study has explicitly examined adverse experiences as predictors of intuitive or mindful eating.

The strengths of the present study include expanding the more conventionally studied adverse experiences in the eating behavior literature (i.e., child maltreatment and intimate partner violence) by assessing household dysfunction and experiencing or witnessing other adversities, types of experiences that have been largely understudied as risk factors for eating behaviors. Notably, we advance the field of adaptive eating behaviors, by finding evidence that adverse experiences may predict lower engagement with adaptive eating behaviors, in addition to increased risk for maladaptive eating. Further, the large study population of emerging young adults allowed us to stratify analyses by gender. Although most studies exploring the association between adverse experiences and eating behaviors are focused on women [46-48] or do not report stratified results by gender [28-30-51], we document that associations between adversities and eating behaviors are not exclusive to women. Although the direction of the association between adverse experiences and eating behaviors is consistent between women and men, the magnitudes and patterns of these associations appear to differ by gender. Relatedly, the present study contributes to the understanding of the relationship between adverse experiences and eating behaviors in emerging adulthood, a relatively understudied developmental period between adolescence and young adulthood.

Despite these strengths, limitations of the present study should be acknowledged. First, participants were drawn from the St. Paul-Minneapolis metropolitan area, which limits generalizability to other geographical regions outside the Midwest. Second, all adverse experiences were retrospectively assessed and self-reported, which increases the possibility of recall bias. Relatedly, although living with someone with a mental illness is a common measure of household dysfunction in the ACE literature, the measure in this study did not distinguish between different psychiatric conditions (e.g., schizophrenia, bipolar disorder, or obsessive-compulsive disorder). Because the impact of mental illness on eating behaviors may vary by the type of mental illness a family member is diagnosed with, future work may consider expanding upon this work to clarify how exposure to family members with varied psychiatric conditions in childhood relates to later eating behaviors. Third, our assessment of adverse experiences did not cover all potentially relevant domains. For example, our study did not assess childhood witnessing of domestic violence, physical bullying by peers, or cyber bullying. Fourth, we were unable to distinguish whether adverse experiences were discrete, one-time experiences, or experiences that repeatedly occurred over time. Future studies should distinguish between acute and chronic adverse experiences when examining associations with eating behaviors. Fifth, the design of this study was cross-sectional, which limits the ability to draw temporal or causal inferences. Last, low internal consistency of the intuitive and mindful eating measures also represents an important limitation of the present study. However, low internal consistency is expected with low numbers of items, and the ability to assess these constructs along with several other eating- and weight-related constructs at the population level necessitated brief measures.

Our findings have important implications for research and clinical practice. The substantial prevalence of adverse experiences reported in our study illustrates the critical need to prevent adverse experiences. Given that living as a child with a family member who had depression or had another mental illness, or attempted suicide, and living with a family member with a substance use disorder were the two most prevalent types of adverse experiences, our findings illustrate the need for a more comprehensive assessment of adverse experiences including adverse experiences related to household dysfunction that are often neglected in research studies on eating behaviors. Notably, mental illness is often treatable, and so an important implication of this finding is the need to improve access to mental health care and reduce the stigma associated with mental illnesses among families with children. The greater prevalence of other adverse experiences less commonly examined in the eating behaviors literature (e.g., experience of a violent death of a close family member or friend, witnessing a serious injury or death) and their associations with maladaptive and adaptive eating suggest such adverse experiences deserve greater attention from researchers. Our findings may also be useful for practitioners, particularly those who work closely with individuals who engage in maladaptive eating. The finding that adverse experiences are predictors of eating behaviors may provide insight allowing for tailoring of treatments of those engaging in problematic eating. For example, a young person engaging in binge eating who has been exposed to adverse life events may need different strategies than one whose binge eating is primarily due to dietary restrictions and social pressures to be thin [67-68], without the accompaniment of adverse life experiences. Furthermore, the assessment of a more comprehensive range of adverse experiences may aid in the identification of people who are at the greatest risk of engaging in overeating or binge eating, and those who are less likely to adopt intuitive or mindful eating. Therefore, health care providers might consider inquiring about past adverse experiences when assessing patients’ overall well-being and their risk for maladaptive eating and other problematic behaviors as well as adopting adaptive eating. Gaining greater insight into the types of adverse experiences that influence eating behaviors is critical for interventionists aiming to identify where help is most needed and areas to target to protect those at the greatest risk of engaging in maladaptive eating and least likely to adopt adaptive eating. Interventions and programs are needed to help encourage recovery from adverse experiences and develop healthier coping skills and strategies.