Abstract
Adverse Childhood Experiences (ACEs) have been associated with detrimental long-term health outcomes, including obesity risk. Existing research has yet to examine whether early life ACEs are associated with diet in early childhood within socioeconomic subgroups. Data were drawn from the Early Childhood Longitudinal Study-Birth Cohort (2001–2002). Mother-child dyads (n = 7,000) were recruited when children were 9-months old, and followed longitudinally at 2 years, and 4 years. Mothers reported children’s exposure to five ACEs at 9-months and 2 years and children’s daily intake of fruits, vegetables, sweet snacks, and sugar-sweetened beverages (SSBs) at 4 years. Weighted multiple linear regression models tested the effect of cumulative and individual ACEs on child diet in full, low-, and high-SES samples. Cumulative ACE score was inversely associated with frequency of fruit intake in full (b=−0.08, p=0.005) and low-SES samples (b=−0.10, p<0.001). Domestic violence was associated with less frequent fruit intake in full (b=−0.21, p=0.01) and low-SES samples (b=−0.29 p=0.008). In the full sample, incarceration was associated with less frequent fruit intake (b=−0.24, p=0.02), and domestic violence was associated with higher sweet snack (b=0.22, p=0.01) and SSB intake (b=0.27, p=0.009). Results provide preliminary evidence on the association between cumulative and specific ACEs and child diet, and how this relationship varies by SES context. Future research is needed to understand the complex multi-level mechanisms operating along this pathway in order to inform interventions supporting behavior change and to build evidence for policies that may reduce diet-related disparities in ACE exposure.
1. Introduction
The association between Adverse Childhood Experiences (ACEs), stressful or traumatic events, and detrimental long-term health outcomes has been well documented (1–4). ACEs, which typically refer to 7 to 10 traumatic experiences (e.g., forms of child maltreatment, domestic violence, substance abuse, mental health concerns, parental divorce/separation, incarceration) (1, 2), affect over 60% of adults in the U.S. (5) and increase risk of obesity in adulthood (6, 7). Poorer quality diet in childhood increases risk for obesity and a number of other chronic health conditions (8), particularly for those of lower socioeconomic status (SES) (9). However, little is known about whether ACEs in the first two years of life are associated with changes in dietary intake in early childhood or whether this association is influenced by SES (2). Understanding whether specific early life ACEs and/or cumulative ACE exposure are associated with child diet, and whether this is unique to SES is essential to intervention development that promotes healthy eating specific to families’ socioeconomic context. This is particularly important in early childhood, a period of rapid development in which children are highly sensitive to external environments (10) and eating habits that may be carried into adulthood are established (11). The purpose of the present study is to examine whether both specific and cumulative ACEs in the first two years of life are associated with child diet and to assess associations within SES subgroups.
1.2. ACEs and Child Diet
Poor nutrition is a key driver of obesity risk in children, namely the overconsumption of foods high in saturated fats and added sugars and failure to meet Dietary Guidelines for Americans (DGA) recommendations for consumption of healthful foods like fruits and vegetables (8). DGA recommends limiting intake of sugary beverages and snacks, promoting nutrient-dense snacks, and exchanging fruit products with added sugars for whole fruit (8). Children between the ages of 2 to 5 years should be fed 1.5 to 2.5 cups of vegetables and 1 to 2 cups of fruits daily. Intake of added sugars and saturated fat should be limited to less than 10% of total calories (8). Many children do not meet recommended dietary patterns and exceed limits for added sugar, with the majority of added sugars coming from snacks, sweets, and sugar-sweetened beverages (SSBs) (8, 12).
Research has shown increased risk of poorer quality diet, obesity, and higher body mass index (BMI) among adults with a history of ACEs (13–17). Adults reporting childhood physical, emotional, or sexual abuse were more likely to eat food in response to stress (13), engage in emotional eating (14) or binge eating (15, 16), and have higher risk for obesity (13). Although studies have found positive associations between ACEs and obesity risk among children and adolescents (17–20), research on how ACEs may affect dietary intake in early childhood is limited. Two Australian studies found inverse associations between parental mental health concerns and child fruit/vegetable consumption (21, 22). A national study found parental incarceration was associated with increased child consumption of sweet snacks, salty snacks, starch, and soda among 5-year-old children (23). Prospectively examining the effects of ACEs on subsequent child dietary health will determine how early childhood stressors may affect health behaviors linked to chronic illness.
1.3. Frequency and Type of ACE
Research has established a gradient relationship between cumulative ACEs and risk of obesity (1, 2). However, different forms of adversity can differently effect child stress pathways and neural development (24, 25) and may uniquely impact child diet. For example, childhood emotional abuse, but not physical or sexual abuse, was associated with increased risk of emotional eating in adulthood (14). Because little is known about links between ACEs and child diet, the present study examines associations between both cumulative ACEs and individual forms of ACEs.
1.4. SES Disparities in ACEs and Child Health
Understanding SES disparities and the effects of ACEs on child diet is critical as families of low-SES are more likely to be exposed to ACEs (5) and risk factors for less healthy eating practices (26, 27), such as eating more highly palatable, energy-dense foods in response to stress, purchasing more fast-food, or having a decreased desire to cook (13, 28, 29). Prospectively assessing associations between early life ACEs and child diet within SES groups may elucidate important pathways to obesity risk and inform clinical intervention, policy, and future research to improve diet among children who experience ACEs across the socioeconomic spectrum.
1.5. Current Study
The present study examines prospective effects of ACEs on child diet and whether cumulative and specific ACEs in the first two years of life predict child diet in high- and low-SES samples. Specifically, we assess: 1) the association between cumulative and specific ACEs from infancy to age 2 years and daily intake of healthy (fruit, vegetables) and unhealthy (sweet snacks, SSBs) foods at 4-years, and 2) whether these associations differ within high- versus low-SES subgroups.
2. Method
2.1. Procedure and Sample
The study was a secondary analysis of the Early Childhood Longitudinal Study-Birth Cohort (ECLS-B) (30). Data were collected from 7,000 mother-child dyads when children were 9 months (2001–2002), and follow-up occurred when children were 2 years (2003–2004), 4 years (2005–2006), and 5 years-old (2006–2007) via in-home observations and self-assessments. Data for this study were drawn from the 9-month, 2-year, and 4-year data collection waves. The ECLS-B used a multistage, clustered, list-frame design to randomly select children born in 2001 as registered in the National Center for Health Statistics (30). Eligibility criteria for initial sampling excluded deceased mothers, mothers younger than 15 years at time of birth, and children adopted prior to 9-months of age. The present study additionally excluded those without a primary maternal figure, mothers under 18 years at time of birth, and children with conditions that might impact development (children born prematurely, non-singleton births, developmental disability). In almost all cases (97.8%) the respondent was the child’s biological mother or female guardian and will henceforth be referred to as mothers.
2.2. Measures
2.2.1. Dependent Variable
Average daily intake of healthy (fruit, vegetables) and unhealthy (sweet snacks, SSBs) items at 4-years of age were collected during in-home assessments. Interviewers asked mothers to recall how many times per day, on average, their child consumed each food type over the past week: fresh fruit (apples, bananas, oranges, berries, or other fruit such as applesauce, canned peaches, canned fruit cocktail, frozen berries, or dried fruit), vegetables (including in stir fry, soup or stew, excluding French fries and fried potatoes), sweet snacks (candy, fruit roll-ups, ice cream, cookies, other sweets), and sugar-sweetened beverages (soda, sports drinks, <100% fruit juice). Response options included: none, 1/day, 2/day, 3/day, 4+/day, 1–3 times/week, and 4–6 times/week. For this study, weekly servings were recoded as daily servings to represent average daily intake of each food group on a scale of 0 to 4 (0 servings/day, 0.5 servings/day, 1 serving/day, 2 servings/day, or 4 servings/day). Each food type was included as a separate continuous dependent variable.
2.2.2. Independent Variables
Adverse Childhood Experiences.
ACEs commonly recognized in the literature (31) were reported by mothers via self-administered questionnaire at the 9-month or 2-year wave in-home assessments. Mothers indicated if they experienced the following in the past year: 1) domestic violence (physical abuse or fear of physical abuse), 2) substance use (history of drinking or drug problem or being told they have drinking/drug problem), 3) mental/emotional distress (mild, moderate or severe depressive symptoms per a modified version of the Center for Epidemiologic Studies Depression Scale (CES-D) (30); overnight stay in facility for a psychological/mental health problem, speaking with a mental health professional for an emotional or psychological problem, or feeling/being told that help is needed for any emotional or psychological problem), 4) marital divorce or separation, and 5) incarceration (put in jail, arrested, or convicted of a crime other than drunk driving). Due to data availability, domestic violence was measured at the 9-month wave and all other ACEs were measured at the 2-year wave. Each of the five ACEs were examined individually as dichotomous variables (exposure = 1, no exposure = 0), as well as cumulatively (sum score of ACE exposures; possible scores range from 0 = no exposure to 5 = exposure to all 5 ACEs). The cumulative ACE scoring method is the norm for measuring cumulative ACES and has internal and external reliability (31).
Socioeconomic Status.
SES was calculated as a composite based on z-scores of self-reported family income, highest household education, and parental occupation (coded according to 2000 Standard Occupational Classification Manual and the 1989 General Social Survey) (32). SES was computed by averaging z-scores for all available components and stratifying SES into quintiles (Q1-lowest through Q5-highest). In order to assess group-specific associations between ACEs and diet, the full sample was stratified by low-SES (Q1 and Q2) and high-SES (Q4 and Q5) subgroups. Q3 cases were not included in the present study because income, education, and occupational characteristics were distinct from both the low- and high-SES groups and Q3 cases would not be representative of what constitutes risk in a low- or high-SES context (33). For parameter estimates for Q3 models, see Supplemental Table 1.
Control Variables.
Control variables included child sex (male/female), child age (in months), child race (Black, White, other), urban/rural residence (derived from sampling frame data and 2000 Census of Population summary tapes containing urban/rural information), mother’s country of origin (U.S. born, not U.S. born), and maternal weight status (normal weight, overweight or obese weight).
2.3. Data Analysis
Demographic characteristics were assessed using standard descriptive statistics. Multiple linear regression analysis was weighted to be representative of children born in 2001, applying weights for primary sampling units (PSU) and strata identifiers. Birth certificates by county and county groups were randomly sampled within each PSU and stratified based on region, median household income, proportion minority population, and metro versus non-metro area. Regression specifications and weighting procedures were then applied to SES subgroups. Weighted regression models examined effects of individual and cumulative ACEs on child diet in the full sample. A subgroup analysis approach (34) assessed associations within low-SES and high-SES subsamples using the domain analysis function. Models adjusted for child sex, age, race, urban or rural residence, mother’s country of origin, and maternal weight status due to possible confounding (33). Full sample models also adjusted for SES. The ECLS-B used imputation to handle missing SES data (35). In the present study, all other measures had less than 5% missing data, aside from domestic violence (7.8%), substance use (10.1%), mental/emotional distress (9.4%), incarceration (10.1%), and maternal weight status (7.4%); missing data were handled via list-wise deletion in each analysis (36). Data analysis was carried out using STATA 15.
3. Results
3.1. Sample Characteristics
The final full sample consisted of 3,800 cases with 1,550 in the low-SES sample (lower 2 quintiles) and 1,850 in the high-SES sample (upper 2 quintiles). The low-SES group included disproportionately higher rates of Black participants (21.8% vs. 8.0%); there were higher rates of White participants in the high-SES group (28.5% vs. 70.2%; Table 1). In the full sample, nearly half (49.4%) of children were exposed to one or more ACE (cumulative ACE score M = 0.54, SD = 0.77) by the age of 2 years. Average cumulative ACE scores were significantly higher in the low-SES group (M = 0.81, SD = 0.75) compared to the high-SES group (M = 0.47, SD = 0.02), and exposure to each individual ACE was significantly higher in the low-SES group. Average daily servings of food were 2.41 (SD = 1.10) for fruit, 2.21 (SD = 1.00) for vegetables, 1.78 (SD = .90) for sweet snacks, and 1.20 (SD = 1.00) for SSBs. Only SSB intake varied by SES group (low-SES M = 1.62, SD = 0.04; high-SES M = 0.96, SD = 0.04).
Table 1.
Sample Characteristics, Food Intake, and ACE Exposure by SES, Early Childhood Longitudinal Study-Birth Cohort 2001–2004, U.S.
| Full Sample (n=3,800) | Low-SES (n=1,550) | High-SES (n=1,850) | Low vs. High-SES t/χ2a | |
|---|---|---|---|---|
| Sample Characteristics (n [%])b | ||||
| Child male | 2,150 (50.6) | 800 (51.0) | 950 (50.5) | 0.08 (p = 0.78) |
| Child race | 148.54 (p < 0.001) | |||
| Black | 600 (14.8) | 350 (21.8) | 150 (8.0) | |
| White | 1,450 (50.3) | 350 (28.5) | 800 (70.2) | |
| Other | 2,150 (34.8) | 900 (49.7) | 900 (21.8) | |
| Rural residence | 600 (13.8) | 300 (16.1) | 200 (11.0) | 10.21 (p = 0.002) |
| Mother foreign born | 1,300 (22.6) | 500 (33.4) | 650 (14.3) | 114.42 (p < 0.001) |
| Mother overweight or obese | 1,950 (54.4) | 850 (61.0) | 700 (46.9) | 33.25 (p < 0.001) |
| Food Intakes Per Day (M±SD) | ||||
| Fruit (0–4) | 2.41 (1.1) | 2.51 (1.1) | 2.41 (1.1) | 1.70 (p = .09) |
| Vegetable (0–4) | 2.21 (1.0) | 2.24 (1.1) | 2.19 (1.0) | 1.26 (p = .21) |
| Sweet snacks (0–4) | 1.78 (0.9) | 1.76 (1.0) | 1.82 (0.9) | −0.21 (p = .84) |
| Sugar-sweetened beverages (0–4) | 1.29 (1.1) | 1.62 (0.04) | 0.96 (0.04) | 11.76 (p < .001) |
| ACEs | ||||
| Cumulative ACE score (0–5) M±SD | 0.54 (0.77) | 0.81 (0.75) | 0.47 (0.02) | −10.30 (p < .001) |
| Domestic violence n (%) | 300 (6.8) | 150 (10.8) | 100 (3.3) | 76.75 (p < .001) |
| Substance use n (%) | 200 (4.6) | 100 (5.1) | 50 (3.0) | 11.53 (p = .02) |
| Mental/emotional distress n (%) | 1,700 (42.9) | 700 (48.5) | 650 (36.8) | 49.76 (p < .001) |
| Divorce n (%) | 200 (5.5) | 150 (8.0) | 50 (2.9) | 48.21 (p < .001) |
| Incarceration n (%) | 200 (5.1) | 150 (8.5) | 50 (1.7) | 86.54 (p < .001) |
Abbreviations: ACE=Adverse Childhood Experience, SES=Socioeconomic status.
Chi square test of independence and t-tests were used to assess differences by low- vs. high-SES.
Ns are unweighted and rounded to the nearest 50 to conform to reporting guidelines. All other statistics are weighted at the child level.
SOURCE: U.S. Department of Education, National Center for Education Statistics, Early Childhood Longitudinal Study, Birth Cohort. Selected years 2001–2004.
3.2. Full Sample Multiple Regression Results
In the full sample, cumulative ACE scores were associated with decreased frequency of fruit intake (b = −0.08, p = 0.005, F[12, 77] = 6.96, p < 0.001; Table 2). Domestic violence (b = −0.21, p = 0.01, F[12, 77] = 6.32, p < 0.001) and incarceration (b = −0.24, p = 0.024, F[12, 77] = 7.83, p < 0.001) were associated with decreased frequency of fruit intake and domestic violence was associated with increased sweet snack intake (b = 0.22, p = 0.013, F(12, 77) = 2.71, p = 0.004) and frequency of SSBs (b = 0.27, p = 0.009, F(12, 77) = 19.85, p < 0.001).
Table 2.
Multivariate Linear Regression Results for Full Sample and by SES, Early Childhood Longitudinal Study-Birth Cohort 2001–2004, U.S.
| Cumulative ACE Score Modelsa | ||||
| Fruit b (95% CI) | Vegetables b (95% CI) | Sweet Snacks b (95% CI) | SSBs b (95% CI) | |
| Full Sample | −0.08 (−0.14, −0.02) | 0.001(−0.06, 0.07) | 0.02 (−0.03, 0.06) | 0.05 (−0.01, 0.09) |
| Low-SES | −0.10 (−0.19, −0.02) | −0.03 (−0.13, 0.08) | −0.0001 (−0.08, 0.08) | −0.01 –0.11, 0.09) |
| High-SES | −0.06 (−0.15, 0.03) | −0.003 (−0.12, 0.11) | 0.004 (−0.08, 0.09) | 0.08 (−0.03, 0.19) |
| Individual ACE Score Models | ||||
| Domestic Violence | ||||
| Fruit | Vegetables | Sweet Snacks | SSBs | |
| Full Sample | −0.21 (−0.38, −0.49) | −0.07 (−0.27, 0.12) | 0.22 (0.05, 0.39) | 0.27 (0.07, 0.47) |
| Low-SES | −0.29 (−0.50, −0.08) | −0.09 (−0.34, 0.15) | 0.20 (−0.03, 0.43) | 0.30 (−0.01, 0.62) |
| High-SES | −0.16 (−0.46, 0.14) | −0.03(−0.33, 0.27) | 0.15 (−0.16, 0.46) | −0.05 (−0.27, 0.16) |
| Substance Use | ||||
| Full Sample | −0.19 (−0.44, 0.56) | 0.02 (−0.17, 0.22) | −0.10 (−0.25, 0.43) | −0.03 (−0.27, 0.20) |
| Low-SES | −0.27 (−0.63, 0.08) | −0.15 (−0.28, 0.09) | −0.01 (−0.23, 0.21) | −0.06 (−0.50, 0.38) |
| High-SES | 0.03 (−0.31, 0.38) | 0.30 (−0.10, 0.71) | −0.30 (−0.57, −0.03) | 0.25 (−0.04, 0.54) |
| Mental/Emotional Distress | ||||
| Full Sample | −0.07 (−0.17, 0.30) | −0.04 (−0.12, 0.43) | 0.03 (−0.03, 0.10) | 0.003 (−0.06, 0.68) |
| Low-SES | −0.10 (−0.26, 0.06) | −0.06 (−0.20, 0.08) | −0.05 (−0.19, 0.09) | −0.09 (−0.24, 0.07) |
| High-SES | −0.11 (−0.24, 0.01) | −0.06 (−0.19, 0.06) | 0.04 (−0.05, 0.13) | 0.08 (−0.06, 0.22) |
| Divorce | ||||
| Full Sample | 0.02 (−0.19, 0.23) | 0.20 (−0.03, 0.42) | −0.11 (−0.28, 0.56) | 0.05 (−0.16, 0.25) |
| Low-SES | 0.10 (−0.23, 0.43) | 0.08 (−0.23, 0.38) | −0.11 (−0.37, 0.16) | −0.15 (−0.43, 0.14) |
| High-SES | 0.07 (−0.40, 0.54) | 0.30 (−0.17, 0.78) | −0.04 (−0.43, 0.34) | 0.26 (−0.23, 0.76) |
| Incarceration | ||||
| Full Sample | −0.24 (−0.45, −0.32) | 0.04 (−0.15, 0.24) | 0.04 (−0.15, 0.23) | 0.10 (−0.10, 0.30) |
| Low-SES | −0.17 (−0.47, 0.12) | 0.08 (−0.17, 0.34) | 0.08 (−0.16, 0.32) | 0.05 (−0.19, 0.30) |
| High-SES | −0.14 (−0.65, 0.37) | −0.09 (−0.63, 0.45) | −0.10 (−0.62, 0.42) | 0.22 (−0.32, 0.76) |
Abbreviations: ACE=Adverse Childhood Experience, SES=Socioeconomic status, SSBs=sugar-sweetened beverages
Values are coefficients for parameter estimates of multiple regression models (95% confidence interval); p values <0.05 are in boldface; see manuscript text for actual p value. All estimates are weighted at the child level and adjusted for control variables: Child sex, child age (months), child race, urban versus rural residence, mother born in US versus mother not born in US, mother overweight or obese versus mother healthy weight. Full sample models additionally controlled for SES.
SOURCE: U.S. Department of Education, National Center for Education Statistics, Early Childhood Longitudinal Study, Birth Cohort. Selected years 2001–2004.
3.3. Stratified Sample Multiple Regression Results in Low- and High-SES Subgroups
Similar to the full sample, in low-SES families, cumulative ACE score was associated with decreased frequency of fruit intake (b = −0.10, p = 0.016, F[8, 81] = 4.77, p < 0.001; Table 2). In the low-SES sample, domestic violence was associated with decreased frequency of fruit intake (b = −0.29, p = 0.008, F[8, 81] = 4.56, p < 0.001). In high-SES families, substance use was associated with decreased frequency of sweet snack intake (b = −0.30, p < 0.001, F[8, 81] = 5.37, p < 0.001); no other associations were observed. Sensitivity analysis showed the main effect of substance use on sweet snack consumption in the high-SES group was not significant in the simple regression model, nor after adjusting for each ACE and control variable separately, indicating that substance use is only associated with sweet snack consumption in the high-SES sample, after accounting for other ACEs and control variables.
4. Discussion
Study findings demonstrate prospective effects of exposure to cumulative and individual early life ACEs on dietary intake two years later within general and low-SES samples. Specifically, cumulative ACE score was inversely associated with fruit consumption after adjusting for SES and other covariates. These findings are consistent with prior research showing a gradient association between ACEs and poorer dietary or eating outcomes (18, 37). Literature reviews have also indicated that effects of ACEs on obesogenic risks may take at least two years to materialize (38). Further exploration revealed that specific ACEs—domestic violence and parental incarceration— may be particularly relevant in relation to child fruit intake. Domestic violence was also associated with more frequent intake of sugary snacks and beverages. Subgroup analysis showed effects of domestic violence on lower fruit consumption within the low-SES sample only. Research has established distinct effects of certain forms of ACEs on child mental health (39, 40). Present study findings contribute to limited research on the distinct effects of ACEs on child physical health (40).
Several potential mechanisms may partly explain associations between cumulative and specific ACEs with child diet outcomes. Exposure to ACEs can increase stress internalization among children and parents, which may make parents more likely to give in to a child’s request for less healthy foods and children more likely to consume more palatable, energy-dense sugary food and drink as a stress coping mechanism (41, 42). Severe stress, particularly during early childhood development, may affect feeding behaviors (43). Overconsumption of highly caloric and palatable foods may be a form of coping that aids in emotion regulation under stress, such that maternal eating-to-cope may lead to emotional child feeding practices (44). ACEs may also affect breastfeeding practices and earlier introduction of solid foods (45, 46), which may could lead to earlier and higher intake of sweets and SSBs. ACEs can affect the entire family system, increasing risks for maternal depression and PTSD which may also impact feeding behaviors (47). In children, exposure to extreme stressors can interrupt endocrine processes that control appetite and may impair hunger responsiveness (48) leading to negative effects on child diet quality, including development of disordered eating (47).
In relation to specific ACE type, witnessing domestic violence in the home setting or being separated from a parent due to incarceration are stressful experiences for a young child, which may be more likely than other ACEs to effect stress-eating mechanisms and child diet. Findings are in line with past research linking domestic violence before age five to increased risk for overweight/obesity from ages 12 to 20 years (49) and parental incarceration with higher consumption of salty snacks and sugary food and drink at age 5 years (23). Prior research has also found that certain adverse experiences may have more of an effect on obesogenic risk than others. Life events related to family physical or mental health, including serious illness and substance use when children were 4, 9, and 11 years-old had more robust associations with risk of overweight at age 15 years than other negative events related to financial instability, family discord, or changes in household structure (50). Greater negative valence regarding certain ACEs may link to higher obesogenic risk (50). Present study findings expand existing research by demonstrating the effects of early exposure specifically to domestic violence and parental incarceration on dietary intake as early as age 4 years.
Subgroup analysis revealed effects of ACE score and domestic violence on fruit intake in the low-SES group, and parental substance use was associated with decreased intake of sweet snacks in the high-SES group. This is in line with past research suggesting the importance of SES in the association between ACEs and child health. In a study of children with neurodevelopmental delays between the ages of 2 and 7 years, for example, ACEs were only associated with increased risk for overweight/obesity after accounting for poverty (51). The unexpected finding between substance use and diet in the high-SES group may be indicative of shifts in dietary intake patterns for children exposed to parental substance use, although not in the hypothesized direction. Those of higher SES may be protected from the effects of cumulative ACE exposure, because of the presence of social and communal protective factors that promote health. This includes access to quality and affordable food, childcare, housing, transportation, and social supports (2). Overall, findings from subgroup analysis suggest that the consideration of SES may provide a more accurate understanding of the role of ACEs in affecting child diet outcomes and highlight the need for additional supports for low-SES groups which may be more vulnerable to risk.
Non-significant findings regarding vegetable intake may be explained by several factors, including child age or taste preferences, food accessibility, or measurement limitations. For example, in younger children, taste preference (i.e., pickiness) may be a relatively strong predictor of vegetable consumption versus fruit consumption. In general, U.S. children are not consuming adequate intake of vegetables and this could impact the ability to detect differences in intake by ACE exposure. Future research, including qualitative work, can tease apart if and how intake of different types of foods are differentially affected by ACEs.
4.1. Strengths and Limitations
This study has several strengths, including assessment of how ACEs experienced in the first two years of life may influence diet two years later. Prior research has been highly limited in this area and has primarily assessed retrospective reports of ACEs in the first 18 years of life. Given that the first few years of life are critical for long-term development, study results can inform future research to mitigate early life adversity and obesity-associated risks. Further, our study sheds light on unique effects of ACEs on diet-related health risks specific to SES context, a highly understudied area (2).
There are several limitations. First, present study findings may not reflect current trends, as data were collected from 2001–2006. However, poor dietary trends among U.S. children have been relatively stable since this time (8). Second, measures of child diet were based on mother-report of select food items, which do not reflect the contribution of energy intake to overall child diet and may not reflect intake outside the home. Diet was only available at the 4-year wave, limiting the study to cross-sectional analysis across multiple time points. Therefore, causality cannot be inferred. However, given the dearth of information on the relationship between ACE exposure and dietary intake, assessing key food groups at this developmental stage contributes important information about the signals between ACEs and diet quality. Third, ACEs were assessed using only five risk factors due to data availability. Future studies should use more rigorous dietary intake methods and outcome measures (e.g., Healthy Eating Index) as well as a more comprehensive assessment of ACEs. Fourth, other factors that may play a role along the pathway from ACEs to child diet, including race and ethnicity, parent feeding practices and weight status, food and physical environments, and child temperament should be further explored. Finally, unweighted cell sizes are small in the high-SES subsample, limiting power to determine what may be driving these effects. Future research should include larger diverse samples in order to better understand environmental and contextual characteristics that may increase risk of poor diet outcomes across the SES spectrum.
5. Conclusions and Implications
This study provides preliminary evidence of how cumulative and specific ACEs may affect child dietary intake and how effects might be specific to SES context. These dietary risk patterns are an upstream marker of serious health problems later in development, including obesity and other chronic diseases. Screening and supports for those with ACEs, including trauma-informed service provision (2) inclusive of parenting supports (52) could be key to obesity prevention among parents of young children, particularly for those of low-SES. Pathways between ACEs and diet and obesity risk are complex likely explained by physiological, psychological, and behavioral factors (2, 25) that should be explored in future research. Future work can inform clinical interventions that support behavior change and inform policy change aimed at reducing diet-related disparities in ACE exposure.
Supplementary Material
Highlights.
Cumulative and specific ACEs predict poorer quality child diet
Domestic violence and incarceration uniquely predicted poorer quality diet
Effects of ACEs on poorer quality diet were observed in low, not high, SES group
Acknowledgements:
Support for this study was provided in part by NIH NIMHD K01MD015326, NICHD K23HD101554, and the University of Michigan Vivian A. and James L. Curtis School of Social Work Research and Training Center. The views expressed here are solely the responsibility of the authors and do not necessarily represent the official views of funders.
Abbreviations
- ACEs
Adverse Childhood Experiences
- BMI
Body Mass Index
- ECLS-B
Early Childhood Longitudinal Study-Birth Cohort
- SES
Socioeconomic Status
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Felitti VJ, Anda RF, Nordenberg D, Williamson DF, Spitz AM, Edwards V, et al. Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults - The adverse childhood experiences (ACE) study. American Journal of Preventive Medicine. 1998. May;14(4):245–58. [DOI] [PubMed] [Google Scholar]
- 2.Suglia SF, Koenen KC, Boynton-Jarrett R, Chan PS, Clark CJ, Danese A, et al. Childhood and adolescent adversity and cardiometabolic outcomes: A scientific statement from the American Heart Association. Circulation. 2018;137(5):e15–e28. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Sonu S, Post S, Feinglass J. Adverse childhood experiences and the onset of chronic disease in young adulthood. Preventive Medicine. 2019;123:163–70. [DOI] [PubMed] [Google Scholar]
- 4.Dube SR, Felitti VJ, Dong M, Giles WH, Anda RF. The impact of adverse childhood experiences on health problems: evidence from four birth cohorts dating back to 1900. Preventive Medicine. 2003;37(3):268–77. [DOI] [PubMed] [Google Scholar]
- 5.Merrick MT, Ford DC, Ports KA, Guinn AS. Prevalence of adverse childhood experiences from the 2011–2014 behavioral risk factor surveillance system in 23 states. JAMA Pediatrics. 2018;172(11):1038–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Hemmingsson E, Johansson K, Reynisdottir S. Effects of childhood abuse on adult obesity: a systematic review and meta-analysis. Obesity Reviews. 2014. November;15(11):882–93. [DOI] [PubMed] [Google Scholar]
- 7.Palmisano GL, Innamorati M, Vanderlinden J. Life adverse experiences in relation with obesity and binge eating disorder: A systematic review. Journal of Behavioral Addictions. 2016. March;5(1):11–31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.US Department of Health and Human Services. Dietary guidelines for Americans 2015–2020: Skyhorse Publishing Inc.; 2017. [Google Scholar]
- 9.Min J, Xue H, Wang Y. Association between household poverty dynamics and childhood overweight risk and health behaviours in the United States: a 8‐year nationally representative longitudinal study of 16800 children. Pediatric Obesity. 2018;13(10):590–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Black MM, Gove A, Merseth KA. Platforms to Reach Children in Early Childhood. In: In: Bundy DAPSN, Horton S, Jamison DT, Patton GC, editor. Child and Adolescent Health and Development 3rd ed Washington, DC: The International Bank for Reconstruction and Development / The World Bank; 2017. [PubMed] [Google Scholar]
- 11.Nicklaus S, Remy E. Early origins of overeating: tracking between early food habits and later eating patterns. Current Obesity Reports. 2013;2(2):179–84. [Google Scholar]
- 12.Kachurak A, Bailey RL, Davey A, Dabritz L, Fisher JO. Daily Snacking Occasions, Snack Size, and Snack Energy Density as Predictors of Diet Quality among US Children Aged 2 to 5 Years. Nutrients. 2019;11(7):1440. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Greenfield EA, Marks NF. Violence from parents in childhood and obesity in adulthood: Using food in response to stress as a mediator of risk. Social Science & Medicine. 2009. March;68(5):791–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.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;91:129–36. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Feinson MC, Hornik-Lurie T. Binge eating & childhood emotional abuse: The mediating role of anger. Appetite. 2016;105:487–93. [DOI] [PubMed] [Google Scholar]
- 16.Conley TD, Garza MR. Gender and Sequelae of Child Versus Adult Onset of Sexual Victimization: Body Mass, Binge Eating, and Promiscuity. Journal of Applied Social Psychology. 2011. November;41(11):2551–72. [Google Scholar]
- 17.Noll JG, Zeller MH, Trickett PK, Putnam FW. Obesity risk for female victims of childhood sexual abuse: a prospective study. Pediatrics. 2007;120(1):e61–e7. [DOI] [PubMed] [Google Scholar]
- 18.McKelvey LM, Saccente JE, Swindle TM. Adverse childhood experiences in infancy and toddlerhood predict obesity and health outcomes in middle childhood. Childhood Obesity. 2019;15(3):206–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Lynch BA, Agunwamba A, Wilson PM, Kumar S, Jacobson RM, Phelan S, et al. Adverse family experiences and obesity in children and adolescents in the United States. Preventive Medicine. 2016;90:148–54. [DOI] [PubMed] [Google Scholar]
- 20.Lynch BA, Rutten LJF, Wilson PM, Kumar S, Phelan S, Jacobson RM, et al. The impact of positive contextual factors on the association between adverse family experiences and obesity in a national survey of children. Preventive Medicine. 2018;116:81–6. [DOI] [PubMed] [Google Scholar]
- 21.Webb HJ, Zimmer-Gembeck MJ, Scuffham PA, Scott R, Barber B. Family stress predicts poorer dietary quality in children: Examining the role of the parent-child relationship. Infant and Child Development. 2018. Jul-Aug;27(4). [Google Scholar]
- 22.Renzaho AMN, Kumanyika S, Tucker KL. Family functioning, parental psychological distress, child behavioural problems, socio-economic disadvantage and fruit and vegetable consumption among 4–12 year-old Victorians, Australia. Health Promotion International. 2011. September;26(3):263–75. [DOI] [PubMed] [Google Scholar]
- 23.Jackson DB, Vaughn MG. Parental Incarceration and Child Sleep and Eating Behaviors. Journal of Pediatrics. 2017. June;185:211–7. [DOI] [PubMed] [Google Scholar]
- 24.McLaughlin KA. Future Directions in Childhood Adversity and Youth Psychopathology. Journal of Clinical Child and Adolescent Psychology. 2016. May-Jun;45(3):361–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Doom JR, Mason SM, Suglia SF, Clark CJ. Pathways between childhood/adolescent adversity, adolescent socioeconomic status, and long-term cardiovascular disease risk in young adulthood. Social Science & Medicine. 2017;188:166–75. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Faith MS, Dennison BA, Edmunds LS, Stratton HH. Fruit juice intake predicts increased adiposity gain in children from low-income families: Weight status-by-environment interaction. Pediatrics. 2006. November;118(5):2066–75. [DOI] [PubMed] [Google Scholar]
- 27.Di Noia J, Byrd-Bredbenner C. Determinants of fruit and vegetable intake in low-income children and adolescents. Nutrition Reviews. 2014. September;72(9):575–90. [DOI] [PubMed] [Google Scholar]
- 28.Parks EP, Kazak A, Kumanyika S, Lewis L, Barg FK. Perspectives on Stress, Parenting, and Children’s Obesity-Related Behaviors in Black Families. Health Education & Behavior. 2016. December;43(6):632–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Schuler BR, Daundasekara SS, Hernandez DC, Dumenci L, Clark M, Fisher J, et al. Economic hardship and child intake of foods high in saturated fats and added sugars: The Mediating role of parenting stress among high-risk families. Public Health Nutrition. 2020:1–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Snow K, Thalji L, Derecho A, Wheeless S, Lennon J, Kinsey S, et al. Early Childhood Longitudinal Study, Birth Cohort (ECLS-B), Preschool Year Data File User’s Manual (2005–06)(NCES 2008-024). Washington, DC: National Center for Education Statistics, Institute of Education Sciences, US Department of Education. 2007:1–190. [Google Scholar]
- 31.Bethell CD, Carle A, Hudziak J, Gombojav N, Powers K, Wade R, et al. Methods to assess adverse childhood experiences of children and families: toward approaches to promote child well-being in policy and practice. Academic Pediatrics. 2017;17(7):S51–S69. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Snow K, Thalji L, Derecho A, Wheeless A, Lennon J, Kinsey S, et al. Early Childhood Longitudinal Study, Birth Cohort (ECLS-B), Preschool Year Data File User’s Manual (2005–06) (NCES 2008–024) Washington, DC: National Center for Education Statistics, Institute of Education Sciences, US Department of Education, 2007. [Google Scholar]
- 33.Schuler BR. Social work and toddler overweight risk: Identifying modifiable child and parent factors across the socioeconomic gradient. Social Work in Health Care. 2019;58(10):952–69. [DOI] [PubMed] [Google Scholar]
- 34.Wang R, Ware JH. Detecting moderator effects using subgroup analyses. Prevention science. 2013;14(2):111–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Schafer JL, editor. Imputation procedures for missing data. INCLEN XVI Global Meeting, Bangkok, Thailand Retrieved November; 1999. [Google Scholar]
- 36.Allison PD. Chapter 3: Conventional Methods. In Quantitative Applications in the Social Sciences: Missing data. The SAGE Handbook of Quantitative Methods in Psychology. London: SAGE Publications Ltd; 2002. [Google Scholar]
- 37.Elsenburg LK, van Wijk KJ, Liefbroer AC, Smidt N. Accumulation of adverse childhood events and overweight in children: A systematic review and meta‐analysis. Obesity. 2017;25(5):820–32. [DOI] [PubMed] [Google Scholar]
- 38.Authors. The association between adverse childhood experiences and childhood obesity: A systematic review. Under review.
- 39.McLaughlin KA, Sheridan MA, Lambert HK. Childhood adversity and neural development: deprivation and threat as distinct dimensions of early experience. Neuroscience & Biobehavioral Reviews. 2014;47:578–91. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Westermair AL, Stoll AM, Greggersen W, Kahl KG, Hüppe M, Schweiger U. All unhappy childhoods are unhappy in their own way—Differential impact of dimensions of adverse childhood experiences on adult mental health and health behavior. Frontiers in Psychiatry. 2018;9:198. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Baskind MJ, Taveras EM, Gerber MW, Fiechtner L, Horan C, Sharifi M. Peer Reviewed: Parent-Perceived Stress and Its Association With Children’s Weight and Obesity-Related Behaviors. Preventing Chronic Disease. 2019;16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Suglia SF, Duarte CS, Chambers EC, Boynton-Jarrett R. Social and Behavioral Risk Factors for Obesity in Early Childhood. Journal of Developmental and Behavioral Pediatrics. 2013. October;34(8):549–56. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Sinha R, Jastreboff AM. Stress as a common risk factor for obesity and addiction. Biological Psychiatry. 2013;73(9):827–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Rodgers RF, Paxton SJ, McLean SA, Campbell KJ, Wertheim EH, Skouteris H, et al. Maternal negative affect is associated with emotional feeding practices and emotional eating in young children. Appetite. 2014;80:242–7. [DOI] [PubMed] [Google Scholar]
- 45.Miller-Graff LE, Ahmed AH, Paulson JL. Intimate partner violence and breastfeeding outcomes in a sample of low-income women. Journal of Human Lactation. 2018;34(3):494–502. [DOI] [PubMed] [Google Scholar]
- 46.Moss BG, Yeaton WH. Early childhood healthy and obese weight status: potentially protective benefits of breastfeeding and delaying solid foods. Maternal and Child Health Journal. 2014;18(5):1224–32. [DOI] [PubMed] [Google Scholar]
- 47.Mason SM, Austin SB, Bakalar JL, Boynton-Jarrett R, Field AE, Gooding HC, et al. Child maltreatment’s heavy toll: The need for trauma-informed obesity prevention. American journal of preventive medicine. 2016;50(5):646. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Chuang J-C, Perello M, Sakata I, Osborne-Lawrence S, Savitt JM, Lutter M, et al. Ghrelin mediates stress-induced food-reward behavior in mice. The Journal of Clinical Investigation. 2011;121(7):2684–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Jun H-J, Corliss HL, Boynton-Jarrett R, Spiegelman D, Austin SB, Wright RJ. Growing up in a domestic violence environment: relationship with developmental trajectories of body mass index during adolescence into young adulthood. J Epidemiol Community Health. 2012;66(7):629–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Lumeng JC, Wendorf K, Pesch MH, Appugliese DP, Kaciroti N, Corwyn RF, et al. Overweight adolescents and life events in childhood. Pediatrics. 2013;132(6):e1506–e12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.Mehari K, Iyengar SS, Berg KL, Gonzales JM, Bennett AE. Adverse Childhood Experiences and Obesity Among Young Children with Neurodevelopmental Delays. Maternal and Child Health Journal. 2020. [DOI] [PubMed] [Google Scholar]
- 52.Campbell JA, Farmer GC, Nguyen-Rodriguez S, Walker RJ, Egede LE. Using path analysis to examine the relationship between sexual abuse in childhood and diabetes in adulthood in a sample of US adults. Preventive Medicine. 2018;108:1–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.