The accelerator, the brake, and the terrain: associations of reward-related eating, self-regulation, and the home food environment with diet quality during pregnancy and postpartum in the pregnancy eating attributes study (PEAS) cohort

Socioeconomic Characteristics and the Home Food Environment Are Associated With Feeding Healthful and Discretionary Foods During the First Year of Life in the Pregnancy Eating Attributes Study

BACKGROUND

Feeding of nutrient-poor foods begins in infancy and may adversely influence long-term food preferences.

OBJECTIVE

To examine associations of socioeconomic characteristics, childbearing parent eating behaviors, and home food environment with infant feeding characteristics.

DESIGN

Secondary analysis of a prospective cohort study from first trimester of pregnancy through 12 months postpartum.

PARTICIPANTS/SETTING

Participants with uncomplicated singleton pregnancies and no major chronic illness were enrolled from November 2014 through October 2016 from 2 university-based obstetrics clinics in Chapel Hill, NC. Of 458 enrolled, 321 were retained through 12 months postpartum.

MAIN OUTCOME MEASURES

Participants completed infant food frequency questionnaires indicating age at introduction and frequency of consuming multiple food groups. Exposures included childbearing parent socioeconomic characteristics, hedonic hunger, addictive-like eating, Healthy Eating Index 2015 calculated from three 24-hour diet recalls, and home food environment fruit/vegetable and obesogenic scores.

STATISTICAL ANALYSES PERFORMED

Multiple imputation using Heckman selection model; linear and logistic regressions examining associations with infant feeding characteristics.

RESULTS

Lower education and income were associated with later infant age at introduction to, and lower frequency of consuming fruits and vegetables at age 12 months. Socioeconomic characteristics were not associated with age at introduction to discretionary solid foods; however, lower education and income were associated with greater infant frequency of intake of discretionary foods and greater odds of introducing fruit juice and sweetened beverages by age 12 months. Childbearing parent Healthy Eating Index 2015, hedonic hunger, and addictive-like eating were not consistently associated with infant feeding characteristics. A more obesogenic food environment was associated with greater frequency of intake of discretionary foods, lower frequency of intake of fruit, and greater odds of fruit juice introduction by age 12 months.

CONCLUSIONS

Infant feeding characteristics may be important intervention targets for addressing socioeconomic disparities in child diet quality. Efforts to reduce routine feeding of discretionary foods across socioeconomic groups are needed; modifying the home food environment may promote healthful infant feeding.

Diet quality from early pregnancy through 1-y postpartum: a prospective cohort study

BACKGROUND

Although diet quality during pregnancy and postpartum is important for multiple parent and child outcomes, within-person changes in diet quality throughout these periods have not been extensively examined.

OBJECTIVES

This study investigated diet quality from early pregnancy through 1 y postpartum and examined differences by sociodemographic characteristics in participants receiving obstetric care in Chapel Hill, North Carolina, United States.

METHODS

Participants completed 24-h dietary recalls at 6 study visits (each pregnancy trimester and 6 wk, 6 mo, and 1 y postpartum) (n = 383). Between-visit mean Healthy Eating Index-2015 (HEI) scores (min = 0, max = 100) were compared using the population ratio (PR) method. The NCI usual intakes method estimated the distribution of HEI scores in pregnancy and postpartum; unpaired t-tests compared usual mean HEI scores by covariates; paired t-tests compared differences between mean pregnancy and postpartum.

RESULTS

The total HEI mean ± standard error scores (NCI method) were 61.4 ± 0.8 in pregnancy and 61.7 ± 0.9 in postpartum. Mean HEI scores differed by sociodemographic characteristics, particularly education, marital status, and federal assistance participation. The highest scores were observed in participants with at least a bachelor's degree (64.1 ± 0.9 in pregnancy, 64.5 ± 1.0 in postpartum, n = 257) and those with other non-Hispanic White race/ethnicity (64.7 ± 1.8 in pregnancy, 66.4 ± 2.2 in postpartum). There were no between-visit differences in mean total HEI scores (PR). Although differences were observed in some mean HEI component scores between pregnancy and postpartum visits, they were small (mostly <1 point) in magnitude and in inconsistent directions.

CONCLUSIONS

Stable total HEI mean scores suggest that adherence to dietary guidelines was consistent throughout pregnancy and postpartum in this sample. Although some sociodemographic characteristics may identify individuals at greater risk of diet-related pregnancy complications, low diet quality was pervasive throughout all subgroups and reflects an urgent need for widespread improvement.

Dietary Sugar and Saturated Fat Consumption Associated with the Gastrointestinal Microbiome during Pregnancy

BACKGROUND

Growing evidence supports changes in the gastrointestinal microbiome over the course of pregnancy may have an impact on the short- and long-term health of both the mother and the child.

OBJECTIVE

Our objective was to explore the association of diet quality, as measured by the Healthy Eating Index (HEI), with the composition and gene ontology (GO) representation of microbial function in the maternal gastrointestinal microbiome during pregnancy.

METHODS

We conducted a retrospective, observational analysis of n = 185 pregnant participants in the Pregnancy Eating Attributes Study. Maternal dietary intake was assessed in the first trimester using the automated self-administered 24-h recall method, from which the HEI 2015 was calculated. Rectal swabs were obtained in the second trimester and sequenced using the NovaSeq 6000 system shotgun platform. We used unsupervised clustering to identify microbial enterotypes representative of maternal taxa and GO functional term composition. Multivariable linear models were used to identify associations between taxa, functional terms, and food components while controlling for relevant covariates. Multinomial regression was then used to predict enterotype membership based on a participant's HEI food component score.

RESULTS

Those in the high diet quality tertile had a lower early pregnancy BMI [mean (M) = 23.48 kg/m2, SD = 3.38] compared with the middle (M = 27.35, SD = 6.01) and low (M = 27.49, SD = 6.99) diet quality tertiles (P < 0.01). There were no statistically significant associations between the HEI components or the total HEI score and the 4 alpha diversity measures. Differences in taxa and GO term enterotypes were found in participants with, but not limited to, a higher saturated fat component score (β = 1.35, P = 0.01), added sugar HEI component (β = 0.07, P < 0.001), and higher total dairy score (β = 1.58, P = 0.01).

CONCLUSIONS

Specific dietary components are associated with microbial composition and function in the second trimester of pregnancy. These findings provide a foundation for future testable hypotheses.

Associations of Appetitive Traits and Parental Feeding Style With Diet Quality During Early Childhood

BACKGROUND

Appetitive traits and parent feeding styles are associated with body mass index in children, yet their associations with child diet quality are unclear.

OBJECTIVE

The objective was to examine relations of appetitive traits and parental feeding style with diet quality in 3.5-year-old children.

DESIGN

The study was a secondary, cross-sectional analysis of data from Sprouts, a follow-up study of the Pregnancy Eating Attributes Study (PEAS). Birthing parents completed the Child Eating Behavior Questionnaire, Caregiver's Feeding Styles Questionnaire, and proxy 24-hour dietary recalls for their children from February 2019 to December 2020.

PARTICIPANTS/SETTING

Participants were 162 birthing parents (early pregnancy BMI ≥ 18.5 and absence of preexisting diabetes, any medical condition contraindicating study participation, self-reported eating disorder, or medications that could affect diet or weight) and their children living in North Carolina.

MAIN OUTCOME MEASURES

Healthy Eating Index-2015 (HEI-2015) total scores were calculated.

STATISTICAL ANALYSES PERFORMED

Path modeling was conducted using PROC CALIS with full information maximum likelihood (FIML) to account for missing data (< 2% of all data in dataset). Associations of child appetitive traits and parental feeding style with child HEI-2015 scores, adjusting for exclusive breastfeeding duration and household income-poverty ratio, were examined. Tests of simple effects were conducted in subsamples split by parental feeding style. Hypotheses were formulated during data collection.

RESULTS

A 1-standard deviation (SD) greater food fussiness was associated with a 2.4-point lower HEI-2015 total score (P = .02; 95% confidence interval [CI] [-4.32, -0.48]) in children. When parental feeding style was authoritarian, a 1-SD greater food responsiveness was associated with a 4.1-point higher HEI-2015 total score (P = .007; 95% CI [1.12, 7.01]) in children. When parental feeding style was authoritative, a 1-SD greater slowness in eating was associated with a 5.8-point lower HEI-2015 total score (P = .01; 95% CI [-10.26, -1.33]) in children.

CONCLUSIONS

Parental feeding style may modify the association of appetitive traits with diet quality in young children. Future research could determine whether matching parent feeding styles to child appetitive trait profiles improves child diet quality.

Hedonic hunger, ultra-processed food consumption, and the moderating effects of impulsivity in pregnant individuals with body mass index ≥ 25

Evidence suggests higher hedonic hunger (preoccupation with/desire to consume food for pleasure) is associated with greater ultra-processed food (UPF) consumption in non-pregnant individuals with higher, but not lower, self-report impulsivity or delay discounting. The current study tested the association between hedonic hunger and UPF consumption, and the moderating effects of self-report impulsivity and delay discounting, during pregnancy. Individuals (N = 220) with body mass index (BMI)≥25 completed the Power of Food Scale, 24-h dietary recalls, and Barratt Impulsiveness Scale-Version 11 in early-mid pregnancy. A subset enrolled in an ancillary study (n = 143) completed a Delay Discounting Task. Linear regression and moderation models covaried for age, gestational age, pre-pregnancy BMI, and socioeconomic status. The association between hedonic hunger and UPF consumption was nonsignificant (p = 0.47). Self-report impulsivity was not a significant moderator (p = 0.11), but delay discounting was (p = 0.01). Simple slopes analysis revealed a one-unit increase in hedonic hunger was associated with 7% lower UPF intake among participants with lower (M+1SD) delay discounting (p = 0.01) and 1% higher UPF intake among those with higher (M-1SD) delay discounting (p = 0.57). Findings contrast those from research with non-pregnant samples and indicate lower delay discounting may serve as a protective factor, associated with reduced UPF consumption at higher levels of hedonic hunger, during pregnancy.

Developmental trajectory of appetitive traits and their bidirectional relations with body mass index from infancy to early childhood

Appetitive traits, including food responsiveness, enjoyment of food, satiety responsiveness and slowness in eating, are associated with childhood body mass index. Change in appetitive traits from infancy to childhood and the direction of causality between appetitive traits and body mass index are unclear. The present study examined the developmental trajectory of appetitive traits and their bidirectional relations with body mass index, from infancy to early childhood. Mothers in the Pregnancy Eating Attributes Study and follow-up (n = 162) reported child appetitive traits using the Baby and Child Eating Behaviour Questionnaires at ages 6 months and 3.5 years, respectively. Standardized body mass index (zBMI) was calculated from child anthropometrics. Cross-lagged panel models estimated bidirectional relations between appetitive traits and zBMI. Food responsiveness, satiety responsiveness and slowness in eating increased from infancy to early childhood. In cross-lagged panel models, lower infant satiety responsiveness (B ± SE = -0.45 ± 0.19, p = .02) predicted greater child zBMI. Infant zBMI did not predict child appetitive traits (p-values >.36). From infancy to early childhood, appetitive traits may amplify. Appetitive traits, particularly satiety responsiveness, appear to influence body mass index during this period, suggesting early intervention targeting these traits may reduce childhood obesity.

Relationships of pregnancy and postpartum diet quality with offspring birth weight and weight status through 12 months

OBJECTIVE

This study examined relationships of maternal pregnancy and postpartum diet quality with infant birth size and weight status indicators through 12 months and tested whether breastfeeding duration modifies these associations.

METHODS

In the Pregnancy Eating Attributes Study (PEAS), dietary intake was assessed six times in 458 mothers who were followed from early pregnancy through 12 months postpartum (2014-2018). Logistic and linear mixed models estimated relationships of pregnancy and postpartum diet quality (Healthy Eating Index [HEI]) with offspring who were large-for-gestational-age (LGA) at birth, as well as BMI z score (BMIz) and weight-for-length z score (WFLz) at birth, 6 weeks, 6 months, and 12 months.

RESULTS

Pregnancy HEI was inversely related to LGA (odds ratio [OR] = 0.95, 95% CI: 0.92 to 0.98); HEI was also inversely related to WFLz (β = -0.01, 95% CI: -0.02 to -0.002) and BMIz (β = -0.009, 95% CI: -0.02 to -0.0009) from birth through 12 months. Postpartum HEI was inversely related to WFLz (β = -0.01, 95% CI: -0.02 to -0.0009) and BMIz (β = -0.008, 95% CI: -0.02 to 0.0007) in infants who were breastfed for at least 6 months, but not in those who were breastfed for a shorter duration.

CONCLUSIONS

Maternal diet quality during pregnancy (and during postpartum in mothers who breastfed for a longer duration) was inversely related to LGA and weight status indicators from birth through 12 months. Increasing maternal diet quality may have use for promoting healthy infant weight development.

Associations between Participation in the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) and Maternal Diet Quality

BACKGROUND

An objective of the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) is to improve maternal diet quality, but its effectiveness remains unclear. Better understanding how WIC participation shapes women's diet quality is crucial given that maternal diet plays a critical role in determining mothers' and children's short- and long-term overall health.

OBJECTIVES

This study aimed to compare the diet quality of WIC-participating women to WIC-eligible nonparticipating women and higher-income pregnant and postpartum women using a nationally representative sample.

METHODS

This was a cross-sectional analysis of the National Health and Nutrition Examination Survey (NHANES) 2011-2018 cycles. Women aged 20 to 44 with at least one 24-h recall and complete data on pregnancy and postpartum status and WIC participation were included (n = 626). Diet quality was evaluated using the Healthy Eating Index-2015 (HEI-2015). Multivariable Tukey-adjusted linear models were used to compare HEI-2015 total and component scores between groups. Models were adjusted for age, pregnancy and postpartum status, breastfeeding status, race and ethnicity, and food security.

RESULTS

Most women were postpartum and not pregnant (75%), nonbreastfeeding (60%), identified as non-Hispanic White (58%), and food secure (64%). WIC participants, WIC-eligible nonparticipants, and income-ineligible women had mean Total HEI-2015 scores of 52.7 (95% confidence interval [CI]: 50.6, 54.8), 54.2 (95% CI: 51.6, 56.7), and 55.0 (95% CI: 51.8, 58.2), respectively. There were no differences between groups for total and most component scores. Income-ineligible women had better Fatty Acids scores (5.7; 95% CI: 5.0, 6.4) than WIC participants (4.7; 95% CI: 4.1, 5.3; P < 0.05). WIC-eligible nonparticipants had better Refined Grains scores (6.0; 95% CI: 5.3, 6.6) than WIC participants (5.0; 95% CI: 4.4, 5.6; P < 0.05).

CONCLUSIONS

Overall diet quality was similar across WIC and income groups. Lower HEI-2015 component scores for WIC participants compared with WIC-eligible nonparticipants warrant further exploration. Research evaluating WIC's impact on maternal diet quality is needed to ensure continued support for low-income women's health.

Associations of parent feeding behaviors and early life food exposures with early childhood appetitive traits in an observational cohort study

PURPOSE

Child appetitive traits, eating styles that reflect responsiveness to external influences and internal hunger and satiety signals, are associated with eating behaviors and susceptibility to excess weight gain. However, relatively little is known about early life influences on child appetitive traits. This study investigated relations of early life maternal feeding behaviors and food exposures with appetitive traits at age 3.5 years.

METHODS

Participants of the Pregnancy Eating Attributes Study (PEAS) and follow-up study were enrolled in early pregnancy and followed prospectively. This analysis included data collected from baseline through child aged 3.5-years (n = 160). Child appetitive traits at age 3.5 years were measured using the Child Eating Behavior Questionnaire. Age at introduction to fruit, vegetables, discretionary sweets, and discretionary savory foods was assessed, along with intake frequency at infant ages 6, 9, and 12 months, and 2 years. Maternal feeding to soothe was assessed at child aged 3, 6, and 12 months. Maternal permissive feeding was assessed at child aged 2 years. Multiple linear regressions estimated relations of maternal feeding behaviors and infant food exposures with child appetitive traits at age 3.5 years, controlling for sociodemographics and breastfeeding duration.

RESULTS

Maternal feeding to soothe at 6 (r = 0.39, p < 0.001) and 12 months (r = 0.39, p < 0.001) was positively associated with permissive feeding at 2 years. Maternal feeding to soothe at 12 months and permissive feeding at 2 years were associated with greater child emotional overeating, emotional undereating, and desire to drink. Older age at introduction to fruit (β = 0.20±0.08, p = 0.01) and younger age at introduction to discretionary sweet foods (β = -0.07±0.04, p = 0.06) were associated with greater emotional overeating. Older age at introduction to vegetables (β = 0.22±0.11, p = 0.04) and less frequent feeding of fruit (β = -0.20±0.08, p = 0.01) were associated with greater food fussiness.

CONCLUSIONS

Associations of emotional eating with parent feeding behaviors and early life food exposures suggest the potential for interventions targeting early life feeding to have long-term impact on child appetitive traits and diet quality.

Prospective relations of maternal reward-related eating, pregnancy ultra-processed food intake and weight indicators, and feeding mode with infant appetitive traits

Background

Infant appetitive traits including eating rate, satiety responsiveness, food responsiveness, and enjoyment of food predict weight gain in infancy and early childhood. Although studies show a strong genetic influence on infant appetitive traits, the association of parent and infant appetite is understudied. Furthermore, little research examines the influence of maternal pregnancy dietary intake, weight indicators, and feeding mode on infant appetite. The present study investigated relations of maternal reward-related eating, pregnancy ultra-processed food intake and weight indicators, and feeding mode with infant appetitive traits.

Methods

Mothers in the Pregnancy Eating Attributes Study (458 mothers enrolled, 367 retained through delivery) completed self-report measures of reward-related eating, and principal component analysis yielded two components: (1) food preoccupation and responsiveness and (2) reinforcing value of food. Mothers completed 24-h dietary recalls across pregnancy, and the standardized NOVA (not an acronym) system categorized recalled foods based on processing level. Maternal anthropometrics were measured across pregnancy. At infant age 6 months, mothers reported on feeding mode and infant appetitive traits. Linear regressions were conducted predicting infant appetitive traits from household income-poverty ratio (step 1); maternal reward-related eating components (step 2); pregnancy ultra-processed food intake (% of energy intake), early pregnancy body mass index, and gestational weight gain (step 3); and exclusive breastfeeding duration (step 4).

Results

A 1-SD greater maternal food preoccupation and responsiveness was associated with 0.20-SD greater infant satiety responsiveness (p = .005). A 1-SD greater % energy intake from ultra-processed foods during pregnancy was associated with 0.16-SD lower infant satiety responsiveness (p = .031). A 1-SD longer exclusive breastfeeding duration was associated with 0.18-SD less infant food responsiveness (p = .014). Other associations of maternal reward-related eating, pregnancy ultra-processed food intake and weight indicators, and feeding mode with infant appetitive traits were non-significant.

Conclusions

Proximal early-life environmental factors including maternal pregnancy dietary intake and feeding mode may facilitate or protect against obesogenic infant appetitive traits, whereas infant appetite may not parallel maternal reward-related eating. Further investigation into the etiology of appetitive traits early in development, particularly during solid food introduction, may elucidate additional modifiable risk factors for child obesity.

Trial registration

Clinicaltrials.gov. Registration ID – NCT02217462. Date of registration – August 13, 2014.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12966-022-01334-9.

The Relationship between Addictive Eating and Dietary Intake: A Systematic Review

(1) Background: Research suggests that certain foods may have addictive effects; however, no reviews have systematically appraised studies in this area. The aims of this review were to determine the nutrients, foods and dietary patterns associated with addictive eating. (2) Methods: Published studies up to November 2020 were identified through searches of 6 electronic databases. Eligible studies included those in in children and adults that reported dietary intakes of individuals with 'food addiction'. (3) Results: Fifteen studies (n = 12 in adults and n = 3 in children/adolescents with Yale Food Addiction Scale defined 'food addiction') were included. Foods commonly associated with addictive eating were those high in a combination of fat and refined carbohydrates. Generally, intakes of energy, carbohydrates and fats were significantly higher in individuals with addictive eating compared to those without. (4) Conclusions: Due to the heterogeneity in study methodologies and outcomes across included studies, it is difficult to conclude if any specific foods, nutrients or dietary patterns facilitate an addictive process. Further research is needed to elucidate potential associations. However, present addictive eating treatment approaches could incorporate individualised dietary advice targeting foods high in fat and refined carbohydrates.