Relations of Pregnancy and Postpartum Diet Quality with Offspring Birth Weight and Weight Status Through 12 Months

Abstract

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 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 large for gestational age (LGA) at birth; and 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 (OR=0.95, 95%CI:0.92,0.98); HEI was also inversely related to WFLz (β=−0.01, 95%CI:−0.02,−0.002) and BMIz (β=−0.009, 95%CI:−0.02,−0.0009) from birth through 12 months. Postpartum HEI was inversely related to WFLz (β=−0.01, 95%CI:−0.02,−0.0009) and BMIz (β=−0.008, 95%CI:−0.02,0.0007) in infants breastfed for at least six months but not in those breastfed for shorter duration.

Conclusions:

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

INTRODUCTION

Rapid infant weight gain is a risk factor for child obesity development (1) as well as numerous chronic diseases and premature mortality (2). Given the persistence of obesity from childhood through adulthood (3) and the lack of effective treatments (4, 5), identifying early modifiable influences is a critical area of research (6).

Maternal pregnancy diet is hypothesized to influence infant growth and development through its effects on the intrauterine environment (7). Evidence suggests that lower maternal high-fat diet (8), higher protein diet (9), greater intake of fruit and pulses (10), healthier data-derived dietary patterns (11, 12), and higher scores on a priori diet quality indices (13, 14, 15) are associated with lower odds of macrosomia and multiple indicators of lower neonatal adiposity and weight status (e.g., skinfolds, fat mass, and size for gestational age). However, null associations have also been reported (16, 17). Fewer studies have investigated associations of pregnancy dietary intake with infant postnatal growth. One study in breastfeeding mothers found that maternal late-pregnancy diet quality (Healthy Eating Index-2015 at 37–40 weeks gestation) was inversely associated with weight-for-length z-scores (WFLz) from birth to 6 months (18). Similarly, two studies found associations of a healthier data-derived dietary pattern during pregnancy was inversely associated with child adiposity through approximately age 5 years, but most associations were attenuated after adjustment for confounders (11, 19). Most evidence on relations of maternal diet with infant weight outcomes is based on maternal pregnancy diet assessment at a single timepoint via food frequency questionnaires, which has lower validity than multiple 24-hour diet recalls or food records. Furthermore, dietary exposures of interest are inconsistent between studies and often data-driven, making implications for other populations difficult to interpret.

Maternal postpartum dietary intake is thought to influence infant growth via its influence on human milk composition (19, 20) or by influencing infant food exposures in the home food environment (21). Few studies have examined relations of postpartum maternal diet quality with weight status throughout infancy. Research in this area has focused on relations of specific components of maternal diet rather than overall diet quality with offspring adiposity at birth and in later childhood (3 years of age or later) (22, 23, 24). One study, which included only exclusively breastfeeding mothers, investigated the relationship of maternal postpartum diet quality with infant growth, finding that higher diet quality from pregnancy through three months postpartum was associated with lower infant WFLz and adiposity indicators from birth to six months (18). However, research is needed to test whether the relationship of maternal postpartum diet quality with infant weight status differs for infants who do not consume breastmilk exclusively, and whether relations of maternal diet quality with infant weight status persist beyond 6 months of age. Therefore, the purpose of this study was to investigate relations of maternal pregnancy and postpartum diet quality with infant weight development from birth to 12 months, and to examine whether breastfeeding modifies these associations.

METHODS

The Pregnancy Eating Attributes Study (PEAS) was a prospective cohort study of women enrolled in early pregnancy and followed through 12 months postpartum (25). The primary objective was to investigate relationships of reward-related eating, self-regulation, and the home food environment with pregnancy and postpartum dietary intake and weight change. Participants at two university-based obstetrics clinics in Chapel Hill, North Carolina were recruited between November 2014 and October 2016. Eligibility criteria included confirmed uncomplicated singleton pregnancy ≤ 12 weeks gestation, maternal age ≥18 and ≤ 45 years, willingness to participate and to provide informed consent for her participation and assent for baby’s participation, BMI ≥ 18 kg/m², ability to complete assessments in English, access to Internet with email, planning to deliver at the UNC Women’s Hospital and to remain in the geographical area through follow-up. Exclusion criteria included pre-existing diabetes, participant-reported eating disorders, and medical conditions contraindicating study participation (e.g., physical or mental health illness or medication affecting diet or weight). Additional details have been described elsewhere (26, 27). Participants attended study visits once per trimester (≤12 weeks, 16–22 weeks, and 28–32 weeks gestation), and three times postpartum (4–6 weeks, 6 months, and 12 months), and completed self-report measures online within each visit window. The Institutional Review Board of the University of North Carolina-Chapel Hill approved the protocol. Of 458 women enrolled, 367 were retained through pregnancy and 321 completed the 1-year postpartum assessment.

Outcome measures

Infant anthropometrics were obtained at birth, 6 weeks, 6 months, and 12 months of age at the postpartum study visits. Weight was measured on a calibrated infant scale (Tanita BD-585) to the nearest 10 g (with a clean, dry diaper and excess clothing removed) and length was measured to the nearest 0.1 cm on a recumbent infant length board with a stadiometer (Ellard Instrumentation PEB LB 35–107-x). Measures were obtained in duplicate; a third measure was taken if the two measures differed by more than 1 cm or 10 g, and the mean of the two closest measures was used as the final value. Age- and sex-specific BMI z-scores (BMIz) and weight-for-length z-scores (WFLz) were calculated according to World Health Organization (WHO) growth standards. These measures were selected since WFLz is the predominant international standard indicator of infant weight status used to monitor obesity risk in children, and BMIz has demonstrated higher positive predictive value for obesity at 2 years of age as compared with WFLz (28). Large-for-gestational age (LGA, > 90th percentile) at birth was also calculated (29). Infant anthropometric data were available for 348 infants, which comprised the analytic sample. Missing infant anthropometric data were present for n=34 at birth, n=36 at 6 weeks, n = 76 at 6 months, and n = 70 at 12 months. Fourteen infants (3.9%) had missing data on infant anthropometrics at all visits.

Exposure assessment

Mothers were asked to complete six 24-hour diet recalls, including one per trimester (≤15 weeks gestation, 16–27 weeks gestation, 28–36 weeks gestation), and at 6 weeks, 6 months, and 12 months postpartum. Participants completed recalls using the Automatic Self-Administered 24-hour Dietary Recall (ASA24), an online tool developed by the National Cancer Institute that uses an automated multiple pass method and has been validated against true intake and interviewer-administered recalls (30). The ASA24 prompts participants to indicate all foods consumed, including details of food preparation, brands, portion size, and additions. U.S. Department of Agriculture Food and Nutrient Database for Dietary Surveys numeric food codes are used to estimate macronutrient, micronutrient, food categories and USDA Food Patterns Equivalents Database food groups. Participants received written instructions on the ASA24, and research staff assisted participants who encountered problems and followed up with participants if outliers in amounts were noted during the interview. Staff at the Nutrition and Obesity Research Core at UNC – Chapel Hill identified and corrected implausible entries (e.g., food items with implausible energy, fat or amount consumed) and missing food or nutrient values and quantities. Dietary records indicating daily energy intake < 600 kcal (36 of 1883 records, 1.9%) were excluded from analyses. Records with daily energy intake > 4500 kcal (21 records, 1.1%) were reviewed individually and determined to reflect plausible intake.

Prenatal and postpartum diet quality were evaluated using the Healthy Eating Index-2015 (HEI), an a priori indicator of diet quality that measures conformance to the 2015 US Dietary Guidelines for Americans (31). The total score (HEI) has a maximum of 100 and is calculated by summing 13 component scores, including 9 “adequacy components” (total fruit, whole fruit, total vegetables, greens and beans, whole grains, dairy, total protein, seafood and plant proteins, fatty acids) and 4 “moderation components” (refined grains, sodium and added sugars and saturated fats) using density-based (e.g., per 1,000 kcal) scoring to enable comparisons across individuals with varying energy requirements. Food group intakes for calculating HEI component and total scores are quantified using the MyPyramid Equivalents Database, which disaggregates and translates foods consumed into the number of ounce or cup equivalents for 32 major groups and subgroups. Details regarding which foods contribute to each HEI component score have been described elsewhere (add citation 31). Subscale scores reflecting adherence to adequacy (max score = 60) and moderation (max score = 40) components were also calculated; for both subscales, higher scores indicate greater adherence to dietary guidelines. Prenatal and postpartum HEI scores were calculated for each participant from the combined pregnancy and postpartum diet recalls, respectively.

Sociodemographic characteristics and pregnancy-related factors

Early pregnancy weight was measured at enrollment by trained research staff using a standing scale and recorded to the nearest 0.1 kg. Mothers reported education, employment, marital status, household income, and household composition at baseline. The income-to-poverty ratio was calculated from reported household income and composition using 2016 poverty levels (32). Mothers self-reported race and ethnicity as mandated by the US National Institutes of Health (NIH), consistent with the Inclusion of Women, Minorities, and Children policy. Based on the NIH Policy on Reporting Race and Ethnicity Data, participants were categorized as Non-Hispanic American Indian or Alaska Native, Non-Hispanic Asian, Non-Hispanic Black or African American, Hispanic or Latino, Non-Hispanic Native Hawaiian or Other Pacific Islander, or Non-Hispanic White. Maternal age, parity, history of gestational diabetes diagnosis, current/history of smoking, final measured pregnancy weight, and delivery mode (vaginal vs cesarian) were obtained from the electronic medical record. Gestational weight gain was calculated as the difference between last measured medical pregnancy weight and weight at study enrollment, and classified according to the 2009 IOM guidelines as inadequate (below recommended), adequate (within recommended), or excessive (above recommended) (33). Mothers reported infant feeding mode at all postpartum visits using questions developed for the Infant Feeding Practices Study II (34); women who reported they were no longer breastfeeding were asked to indicate the infant’s age when they stopped, from which duration of any breastfeeding (defined as partial or exclusive breastfeeding) was calculated.

Statistics

Means ± SD and frequencies were used to summarize sample characteristics. Logistic regressions estimated odds of LGA associated with maternal pregnancy diet quality (total HEI, adequacy, and moderation scores). The minimally sufficient set of covariates for the models assessing the relationship of maternal diet quality with infant weight outcomes was determined to include maternal education and parity based on review of the literature. Variables hypothesized to lie along causal pathways were omitted to prevent overadjustment, which can bias estimates of the total association of maternal diet quality with infant weight outcomes. Linear mixed models with random intercepts using maximum likelihood estimation (using all information) and the independent correlation matrix estimated associations of pregnancy and postpartum diet quality with infant WFLz and BMIz, with separate models for each association. Child age was included as the time variable. The maximum likelihood estimation approach accounts for within-subject correlation of repeated measurements and allows for unbalanced data and unequal spacing (35). We considered including a random slope for child age consistent with growth models, but the random effect estimate was not statistically different from 0 and did not improve model fit (Akaike or Bayesian information criteria) over the random intercept model. Multiplicative interaction terms were used to test whether relations of postpartum diet quality with infant weight status differed by duration of any breastfeeding (≥ 6 months versus < 6 months, consistent with the Healthy People 2020 goal to increase the proportion of infants receiving any amount of breastmilk for 6 months). The margins postestimation command was used to estimate relations of postpartum diet quality with infant weight outcomes for each breastfeeding category. As a sensitivity analysis, we re-estimated all models excluding the 25 women with gestational diabetes diagnoses; coefficient estimates were essentially identical, so results from the full sample are presented. The P value level of statistical significance was .05, and all tests were 2-sided.

RESULTS

As shown in Table 1, mean age at baseline was 30 years. Over half the sample had an early pregnancy BMI over 25 kg/m². The sample was highly educated, with 73% having a bachelor’s degree or higher. The majority of participants were married or cohabitating and almost two-thirds were employed full-time. Just over half the sample was nulliparous. Fewer than 25% of mothers delivered via C-section, and average duration of any breastfeeding was nearly 9 months, with n=221 (72%) breastfeeding for at least 6 months. One third of the sample experienced gestational weight gain within the IOM guidelines. Infants were approximately half male, with average gestational age at delivery of 39 weeks, and average birth weight of 3.4 kg. Forty-one (12.0%) infant birth weights were LGA. Mean total HEI in pregnancy and postpartum was approximately 58 (Table 2). The 25^th – 75^th percentiles of total HEI was 49 – 67 in pregnancy and 49 – 68 in postpartum. Subscale scores indicate slightly higher adherence to adequacy (60% of maximum) than moderation (55% of maximum) diet components.

Table 1.

Sample characteristics (n = 348)

Participant	Characteristic	Mean or No.	SD or %
Mother	Baseline age, years	30.6	4.5
	Early pregnancy BMIa (No., %)
	Normal weight	168	48
	Overweight	95	27
	Obese	85	24
	Race and ethnicity (No., %)
	Non-Hispanic American Indian/Alaska Native	1	0.3
	Non-Hispanic Asian	16	5
	Non-Hispanic Black or African American	56	17
	Hispanic or Latino	25	7
	Non-Hispanic Native Hawaiian/Pacific Islander	1	0.3
	Other raceb	1	0.3
	Non-Hispanic White	239	71
	Education (No., %)
	Bachelor’s degree or higher	231	73
	Less than Bachelor’s degree	87	27
	Marital status (No., %)
	Married or cohabitating	289	91
	Single, divorced, or separated	29	9
	Employment (No., %)
	Full time	199	63
	Part time	47	15
	Student	17	5
	Not working	55	17
	Income-poverty ratioc	3.9	2.0
	Parity (No., %)
	Parous	147	42
	Nulliparous	201	58
	Delivery mode (No., %)
	Vaginal	269	77
	C-section	79	23
	Breastfeeding duration, months	8.7	4.6
	Gestational weight gaind (No., %)
	Inadequate	67	19
	Adequate	115	33
	Excessive	166	48
Child	Sex at birth (No., %)
	Female	177	51
	Male	171	49
	Gestational age at delivery, weeks	39.4	1.8
	Birth weight, kg	3.4	0.5

^aBaseline body mass index, calculated as weight (kg) divided by square of height (m²) (normal weight: 18.5 ≤ BMI < 25, overweight: 25 ≤ BMI < 30, obesity: 30 ≤ BMI)

^bSelf-identified as “Caucasian”.

^cRatio of household income to poverty levels accounting for household size and composition (greater values indicate higher income relative to poverty level).

^cGestational weight gain categories corresponding to 2009 Institute of Medicine BMI-specific recommendations.

Table 2.

Maternal pregnancy and postpartum diet quality in the Pregnancy Eating Attributes Study.

	Pregnancy (n=313)		Postpartum (n = 256)
	Mean	SD	Mean	SD
HEI total scorea	58.1	12.0	58.3	13.7
Adequacy scoreb	36.6	8.5	35.7	9.6
Moderation scorec	21.4	5.1	22.6	5.5

HEI: Healthy Eating Index – 2015

^aHEI total score (min = 0, max = 100) reflects adherence to all adequacy and moderation diet components.

^bHEI adequacy score (min= 0, max = 60) reflects adherence to fruit, vegetables, whole grains, dairy, protein, and fatty acid components.

^cHEI moderation score (min = 0, max = 40) reflects adherence to refined grains, sodium, added sugars, saturated fats components.

Each one-point increase in pregnancy total HEI was associated with 5% (95%CI: 2% - 8%, p = 0.003) lower odds of LGA (Table 3). Each one-point increase in pregnancy HEI adequacy score was associated with a 5% (95%CI: 0.002% - 0.8%0.998) decreases odds of LGA. A one-unit increase in the HEI moderation score was associated with 13% (95% CI: 5% - 20%) lower odds of LGA.

Table 3.

Associations^a of maternal pregnancy diet quality with odds of LGA (large-for-gestational age) in the Pregnancy Eating Attributes Study.

	LGAb
Pregnancy diet quality	OR (95% CI)	P-value
HEI total scorec	0.95 (0.92 – 0.98)	.003
HEI adequacy scored	0.95 (0.91 – 0.998)	.04
HEI moderation scoree	0.86 (0.79 – 0.94)	<.001

Abbreviations: LGA, large for gestational age; HEI, Healthy Eating Index – 2015.

^aEstimates from logistic regression models adjusted for maternal education and age.

^bBirth weight ≥ 90^th percentile for size-for-gestational age.

^cHEI total score (min = 0, max = 100) reflects adherence to all adequacy and moderation diet components.

^dHEI adequacy score (min= 0, max = 60) reflects adherence to fruit, vegetables, whole grains, dairy, protein, and fatty acid components.

^eHEI moderation score (min = 0, max = 40) reflects adherence to refined grains, sodium, added sugars, saturated fats components.

Higher pregnancy diet quality was associated with lower WFLz and BMIz from birth to 12 months (Table 4). A one-point increase in pregnancy HEI total score was associated with a 0.01 decrease in WFLz and 0.009 decrease in BMIz, and each one-point increase in pregnancy HEI moderation score was associated with a 0.04 lower WFLz and BMIz. Coefficient estimates were closest to zero for the adequacy score. Overall, higher postpartum diet quality was inversely associated with infant WFLz and BMIz, although all uncertainty estimates included the null association. However, there was a significant interaction with breastfeeding (p-interaction = 0.06 – 0.10). In infants breastfed for at least six months, each one-point higher maternal postpartum total HEI score was associated with 0.01 lower WFLz and 0.008 lower BMIz, and each one-point higher maternal postpartum HEI moderation score was associated with a 0.01 lower WFLz and 0.008 lower BMIz. The point estimates for the adequacy score were similar in magnitude, but standard errors were larger and uncertainty estimates included the null association. For all outcomes, larger coefficient estimates were observed for relations with the moderation score than for the adequacy or total HEI scores.

Table 4.

Associations^a of maternal pregnancy and postpartum diet quality with infant WFLz and BMIz in the Pregnancy Eating Attributes Study.

	WFLz		BMIz
Prenatal diet quality	β (95%CI)	P-value	β (95%CI)	P-value
HEI total scoreb	−0.01 (−0.02 – −0.002)	.02	−0.009 (−0.02 – −0.0009)	.03
HEI adequacy scorec	−0.006 (−0.02 – 0.007)	.41	−0.003 (−0.02 – 0.009)	.61
HEI moderation scored	−0.04 (−0.06 – 0.02)	<.001	−0.04 (−0.06 – −0.02)	<.001
Postnatal diet quality
HEI total scoree	−0.007 (−0.02, 0.001)	.09	−0.006 (−0.01 – 0.001)	.12
Breastfed < 6 mof	0.003 (−0.01 – 0.02)	.65	0.0002 (−0.01 – 0.01)	.98
Breastfed ≥ 6 mog	−0.01 (−0.02 – −0.0009)	.03	−0.008 (−0.02 – 0.0007)	.07
HEI adequacy scoree	−0.009 (−0.02 – 0.002)	.12	−0.007 (−0.02 – 0.003)	.18
Breastfed < 6 mof	0.003 (−0.02 – 0.02)	.77	−0.001 (−0.02 – 0.02)	.89
Breastfed ≥ 6 mog	−0.01 (−0.03 – 0.002)	.08	−0.009 (−0.02 – 0.004)	.17
HEI moderation scoree	−0.02 (−0.03 – 0.004)	.13	−0.01 (−0.03 – 0.004)	.13
Breastfed < 6 mof	0.02 (−0.02 – 0.05)	.47	0.007 (−0.03 – 0.04)	.70
Breastfed ≥ 6 mog	−0.02 (−0.05 - −0.003)	.03	−0.02 (−0.04 - −0.0009)	.04

Abbreviations: WFLz, z-score of sex-specific weight-for-length for age; BMIz, z-score of sex-specific body mass index for age; HEI, Healthy Eating Index-2015.

^aEstimates from linear mixed models adjusted for child age, maternal education, and parity.

^bHEI total score (min = 0, max = 100) reflects adherence to all adequacy and moderation diet components.

^cHEI adequacy score (min= 0, max = 60) reflects adherence to fruit, vegetables, whole grains, dairy, protein, and fatty acid components.

^dHEI moderation score (min = 0, max = 40) reflects adherence to refined grains, sodium, added sugars, saturated fats components.

^eEsimates for associations of postpartum diet quality with infant weight outcomes, not adjusted for breastfeeding.

^fEstimates for associations of postpartum diet quality with infant weight outcomes assuming all mothers breastfed for less than 6 months.

^gEstimates for associations of postpartum diet quality with infant weight outcomes assuming all mothers breastfed for at least 6 months.

DISCUSSION

In this prospective, observational study, higher maternal diet quality was associated with lower infant WFLz and BMIz throughout the first year of life. Associations were driven by the moderation components of the HEI, suggesting maternal pregnant and postpartum intake of refined grains, added sugars, fatty acids, sodium, and saturated fat are more strongly associated with infant weight outcomes than intake of adequacy components (i.e., fruit, vegetables, whole grains, dairy, and protein foods). Higher pregnancy diet quality was associated with lower LGA at birth and lower infant WFLz and BMIz from birth through age 12 months. In infants who received any breastmilk for at least six months, maternal postpartum diet quality was associated with lower WFLz and BMIz from birth through age 12 months, whereas the estimated associations were closer to zero and not statistically significant in infants breastfed for fewer than six months. The magnitudes of these associations were clinically meaningful. For example, holding all covariates constant, a one-unit change in total HEI in pregnancy, considered a very small difference in diet quality, was associated with 5% decreased odds of LGA, a 0.01 standard-deviation decrease in WFL, and a 0.009 standard deviation decrease in BMI. Since the difference between the 75^th and 25^th percentiles of HEI scores was 18 – 19 points, this suggests that, holding covariates constant, infants of mothers whose HEI was at the 75^th percentile would have a 90% decreased odds of LGA, 0.2 standard deviation lower WFL, and 0.16 standard deviation lower BMI as compared with infants of mothers with HEI at the 25^th percentile. The magnitude of associations of HEI moderation scores with infant weight outcomes were larger than those for total HEI.

The relationship of higher diet quality with lower odds of LGA is consistent with prior research showing relations of several aspects of maternal pregnancy diet with infant weight status and adiposity indicators (9, 10, 11, 12, 13, 14, 15). The relationship of higher maternal pregnancy diet quality, specifically moderation components, with weight status from birth through 12 months adds to the literature on the importance of the prenatal environment and maternal pregnancy nutrition for appropriate infant growth (36, 37). Estimates suggest that infants of mothers at the 75^th percentile for total HEI in pregnancy have 90% lower odds of LGA compared with infants of mothers at the 25^th percentile for total HEI in pregnancy. Additionally, these findings suggest the utility of assessing infant weight status after interventions targeting improved maternal adherence to the moderation components of the Dietary Guidelines for Americans during pregnancy.

This study adds to the sparse evidence on relations of postpartum maternal diet quality with infant growth (18). The finding that this relation was evident only in infants breastfed for longer duration has important implications. While previous work indicates that mothers with higher intake of foods that contribute to low overall diet quality introduce these foods earlier and more frequently to their infants (39), the null associations of postpartum diet quality with WFLz and BMIz from birth through 12 months in children of mothers who breastfed less than 6 months are not consistent with an influence via the household food environment; rather, an interpretation consistent with the findings is that relations of maternal postpartum diet on infant growth may be attributable mostly to the impact of maternal diet on human milk composition (40), which in turn influences infant growth (41). Several components of maternal intake (e.g., macronutrients, micronutrients, lipids, trace elements) influence human milk composition (42, 43), which have been related to differences in infant growth (41, 44). In contrast, very few studies have examined relations of overall maternal diet quality with human milk composition (45). While the total estimated association of breastfeeding (partial or exclusive breastfeeding for at least 6 months versus less than 6 months) with WFLz (difference of −0.3, 95%CI:−0.5 - −0.04) and BMIz (difference of −0.3, 95%CI: −0.05 - −0.03) was negative, the positive association of maternal postpartum diet quality with infant growth only in those receiving breastmilk for at least six months suggest that interventions targeting improved postpartum diet quality, especially via decreasing intake of moderation components, in mothers breastfeeding for at least six months may promote optimal infant growth trajectories.

Study strengths include the prospective study design, the large sample size, the repeated assessment of pregnancy and postpartum diet quality using 24-hour recalls throughout pregnancy and postpartum, repeated assessment of infant anthropometrics, and the assessment of multiple prospective confounders, all of which support the internal validity of the findings. Limitations include the single-site enrollment, which included only mothers with access to healthcare and limits generalizability to individuals in other geographic locations, and the observational study design, which limits causal inference. The sociodemographic characteristics of the sample are similar to the characteristics of the local geographic area, but included a higher proportion of highly educated mothers than is observed in the overall U.S. population. While anthropometric data were available for most infants (n=348), there were fewer observations of postpartum diet quality (n = 256) versus pregnancy diet quality (n = 313), which could have decreased statistical power for detecting postpartum associations. Additionally, while participants with postpartum dietary intake data did not differ from those with missing data on pregnancy HEI, maternal anthropometrics, or any outcome variable, those with data on postpartum diet quality were older (31.0±0.3 vs. 30.0±0.5 years), breastfed for longer (8.9±0.3 vs. 7.4±0.7 months), and were more highly educated (79% vs. 52% with at least a bachelor’s degree), which may have resulted in selection bias; however, this is unlikely to have impacted our findings given that there is little reason to hypothesize that the relationship of maternal diet quality with infant weight indicators would differ between those who dropped out versus those who remained in the study. Additionally, while confounding was addressed in these analyses, residual confounding attributable to any observational study design may weaken internal validity. Finally, although BMIz and WFLz are the international standard for monitoring child obesity risk, they do not directly reflect infant body composition or adiposity (i.e., body fat percentage as assessed via dual x-ray absorptiometry or air displacement plethysmography).

Taken together, findings from this study suggest that better alignment of maternal dietary intake with the Dietary Guidelines for Americans 2015 – 2020 during pregnancy, particularly for moderation components, was related to lower infant weight status at birth and throughout infancy. Higher postpartum maternal diet quality was also related to lower WFLz and BMIz from birth to 12 months in those who were breastfed for at least six months, suggesting a potential mechanism via breastmilk composition.

Study Importance Questions.

What is already known about this subject?

• Understanding influences on infant weight status is critical for preventing child obesity risk.

• Maternal pregnancy nutrition is an important influence on fetal growth.

• The relationship of pregnancy and postpartum nutrition with weight status throughout infancy is unclear.

What are the new findings in your manuscript?

• Better pregnancy diet quality was related to lower infant weight status.

• Better postpartum diet quality was related to lower infant weight status only in infants breastfed for at least 6 months.

How might your results change the direction of research or the focus of clinical practice?

Increasing maternal adherence to the Dietary Guidelines for Americans during pregnancy and postpartum may lead to more favorable offspring weight status throughout infancy.