Early introduction of complementary foods/drinks and milk feeding type in the Special Supplemental Nutrition Program for Women, Infants and Children (WIC)

Abstract

Infants younger than 4 months are not ready for complementary foods/drinks (any solid or liquid other than breast milk or infant formula). Almost half of US infants participate in the Special Supplemental Nutrition Program for Women, Infants and Children (WIC), which provides nutrition education and support to low-income families. We describe the prevalence of early introduction (<4 months) of complementary foods/drinks and examine the association of milk feeding type (fully breastfed, partially breastfed or fully formula fed) with early introduction of complementary foods/drinks. We used data from 3310 families in the longitudinal WIC Infant and Toddler Feeding Practices Study-2. We described the prevalence of early introduction of complementary foods/drinks and modeled the association of milk feeding type at Month 1 with early introduction of complementary foods/drinks using multi-variable logistic regression. Thirty-eight percent of infants were introduced early to complementary foods/drinks (<4 months). In adjusted models, infants who were fully formula fed or partially breastfed at Month 1 were 75% and 57%, respectively, more likely to be introduced early to complementary foods/drinks compared with fully breastfed infants. Almost two in five infants were given complementary foods/drinks early. Formula feeding at Month 1 was associated with higher odds of early introduction of complementary foods/drinks. There are opportunities to support families participating in WIC to prevent early introduction of complementary foods/drinks and promote child health.

INTRODUCTION

The introduction of complementary foods/drinks (any solid or liquid other than breast milk or infant formula) is an important milestone in infant nutrition (American Academy of Paediatrics Committee on Nutrition, 2019). The Dietary Guidelines for Americans (DGAs) and the American Academy of Paediatrics recommend the introduction of complementary foods/drinks to infants at about 6 months of age (American Academy of Paediatrics Committee on Nutrition, 2019; US Department of Agriculture and US Department of Health and Human Services, 2020). The early introduction (<4 months of age) of complementary foods/drinks is concerning as younger infants are developmentally not prepared because of gastrointestinal and motor immaturity (American Academy of Paediatrics Committee on Nutrition, 2019; Pérez-Escamilla et al., 2017). In addition, the early introduction of complementary foods/drinks prevents exclusively breastfed infants from reaching the recommended 6 months of exclusive breastfeeding (American Academy of Paediatrics Committee on Nutrition, 2019). Further, research suggests that early introduction of complementary foods/drinks may increase the risk of some chronic diseases, such as obesity (English et al., 2019).

Almost half of US infants participate in the Special Supplemental Nutrition Program for Women, Infants and Children (WIC) (US Department of Agriculture, 2022). WIC is a program administered by the US Department of Agriculture, and services are managed through state and local organizations (US Department of Agriculture Food and Nutrition Service, 2013). Nutritional education and support is provided by WIC, along with referrals to health and social services. Those enrolled in WIC also receive food packages to provide supplemental food. The infant's milk feeding type (fully breastfed, partially breastfed or fully formula fed) and age determine the food package types received (US Department of Agriculture, 2021). A recent systematic review noted that, following the 2009 revision of the WIC food package, breastfeeding exclusivity may have improved and receipt of breastfeeding support services may have improved breastfeeding initiation and duration among WIC participants (Gross et al., 2023). However, exclusive breastfeeding rates remain lower (Centers for Disease Control and Prevention, 2022b) and early introduction of complementary foods is higher (Chiang et al., 2020) among families with low income.

In nationwide samples, there are various reports of the prevalence of early introduction of complementary foods/drinks ranging from 16% to 40% (Barrera et al., 2018; Chiang et al., 2020; Clayton et al., 2013). To our knowledge, no previous studies have examined the prevalence of early introduction of complementary foods/drinks among a population comprised exclusively of WIC participants. Several studies have identified predictors of early introduction of complementary foods/drinks, which include low maternal education (Chiang et al., 2020; Clayton et al., 2013; Doub et al., 2015), low family income (Chiang et al., 2020; Clayton et al., 2013), younger maternal age (Chiang et al., 2020; Clayton et al., 2013; Doub et al., 2015), Non-Hispanic Black race (Barrera et al., 2018; Chiang et al., 2020) and no or shorter duration of breastfeeding (Barrera et al., 2018; Chaparro & Anderson, 2021; Chiang et al., 2020; Clayton et al., 2013; Doub et al., 2015; Klag et al., 2015; Kuo et al., 2011; Scott et al., 2009). Two studies of the general US population (Clayton et al., 2013; Kuo et al., 2011) identified WIC participation as a predictor of early introduction of complementary foods/drinks while two others found no association (Barrera et al., 2018; Deming et al., 2014).

Caregivers of infants provide myriad reasons for the early introduction of complementary foods/drinks, and some reasons vary by milk feeding type. In a nationwide sample (not specific to WIC), caregivers report giving complementary foods/drinks because the infant was old enough, the infant seemed hungry, they wanted to feed the infant something in addition to breast milk or infant formula, the infant wanted the food the caregiver ate, a doctor or other health care professional said the infant should begin eating solid food, or it would help the infant sleep longer at night (Clayton et al., 2013). Given these differing reasons observed by milk feeding type, in addition to previous studies which suggest no or shorter duration of breastfeeding is associated with early introduction of complementary foods/drinks, it may be important to explore the association of milk feeding type with early introduction of complementary foods/drinks among a WIC population.

Previous studies describing complementary food/drink introduction often include WIC participants, but none focus solely on a WIC population. Almost half of US infants participate in WIC (US Department of Agriculture, 2022), and WIC has the ability to reach participants with education and guidance on infant feeding, including the timing of introduction of complementary foods/drinks. The aims of this study are to describe the timing of introduction and prevalence of early introduction of complementary foods/drinks (<4 months) overall and by specific food group, as well as the association of milk feeding type at Month 1 with early introduction of complementary foods/drinks (<4 months) in a population of WIC participants.

METHODS Participants and recruitment

USDA's WIC program enrolled participants from July 1 through November 18, 2013, for an ongoing longitudinal study called the WIC Infant and Toddler Feeding Practices Study-2 (ITFPS-2). Briefly, ITFPS-2 was designed to provide information on feeding practices and nutrition outcomes among caregivers and children enrolled in WIC through the child's 6th birthday (Harrison et al., 2014; May et al., 2017). Sampling was conducted using a two-stage stratified approach. Sampling occurred within WIC sites projected to enroll ≥30 participants/month using a probability proportional to size sample design (May et al., 2017). A total of 80 WIC sites from 27 states were enrolled, representing 37% of WIC sites and 87% of WIC participants (May et al., 2017). Detailed study methodology has been published elsewhere (Harrison et al., 2014; May et al., 2017).

Within the 80 selected sites, individuals who were enrolling in WIC for the first time during their current pregnancy or were enrolling their newborn (<2.5 months old) were invited to participate if they were ≥16 years of age and spoke English or Spanish (May et al., 2017). Individuals completed an in-person screener to determine eligibility. Eligible individuals then completed follow-up surveys by phone; this analysis uses data from phone surveys conducted prenatally and every other month through their child's first 15 months, with additional surveys at 18 and 24 months (questions about complementary food/drink introduction were asked on all surveys between 1 month and 24 months).

Participants provided written informed consent during study screening and enrollment and were provided incentives for enrolling and completing each survey (May et al., 2017). The Centers for Disease Control and Prevention determined that this secondary analysis of de-identified data was not human subjects research and did not require institutional review board review. This study was registered at clinicaltrials.gov (NCT02031978).

There were 6775 caregivers invited to participate in the study; of those screened and eligible, 4367 enrolled (Figure 1) (May et al., 2017). We limited our analyses to participants who responded to Months 1 or 3 surveys and had complete year 1 complementary food/drink data, Month 1 milk feeding data, and covariate data collected prenatally. Final sample size was n = 3310 (weighted N = 388,506).

Measures

The main exposure of interest in this analysis was the type of milk feeding at Month 1. At the Month 1 survey, caregivers were asked, ‘Are you currently feeding breast milk either from the breast or from a bottle, formula, or both?’ and response choices included only breast milk, only formula and both breast milk and formula. Based on these responses, infants are described as fully breastfed, fully formula fed or partially breastfed, respectively; these three levels correspond with the WIC food packages available to infants aged 0–5 months (US Department of Agriculture, 2021). Ten percent of participants in our analytic sample were missing data on the milk feeding type item from the Month 1 survey; if milk feeding type data was missing, information on the WIC food package received at Month 1 was used as a proxy.

Early introduction of complementary foods/drinks was the main outcome of this study, defined as the introduction of any solid or liquid other than breast milk or infant formula before 4 months of age. Early introduction of complementary foods/drinks was calculated using information from 17 food and beverage groups, including 6 liquids (plain water, soft drinks, sweetened beverages, fruit juice, cow's milk, other drinks or liquids) and 11 solids (dairy products other than milk, infant cereal, other cereal, eggs, fruit, vegetables, beans, peanut butter, meats, salty snacks and sweets). Participants were asked in each interview in their child's first 2 years (Months 1, 3, 5, 7, 9, 11, 13, 15, 18 and 24) if their child had been given anything to eat other than breast milk or infant formula (yes/no). If yes, they were asked if the child had been given anything from the 17 food and beverage groups (yes/no). If yes to any of the food and beverage groups, they were then asked how old the child was when given the item. We used information on the timing of introduction to classify each food group as introduced early or not. Once it was reported that a child had been introduced to a food group, that food group was not asked about again. We created a composite variable for early introduction of liquids (‘yes’ if any of the six liquids were introduced early), early introduction of solids (‘yes’ if any of the 11 solid food items were introduced early) and early introduction of complementary foods/drinks (‘yes’ if any of the 17 food items were introduced early).

Caregiver characteristics included age at child's birth (16–19, 20–25 or ≥26 years), self-reported race/ethnicity (Non-Hispanic White, Non-Hispanic Black or African American [referred to as Non-Hispanic Black], Hispanic or another race/ethnicity), education status at birth (≤high school or >high school), marital status (married or unmarried) and parity (nulliparous or parous). Infant characteristics included preterm status (infant born <37 weeks or ≥37 weeks gestation) and type of delivery (vaginal or caesarean). Household characteristics included poverty level at the time of enrollment (≤75%, 76%–130% or >130% of the 2013 poverty guidelines [US Health and Human Services, 2013]).

We also examined the practice of adding infant cereal to bottles using two items asked at Months 1 and 3 which asked, ‘In the past two weeks, how often have you added baby cereal to your baby's bottle?’ If the caregiver responded with an answer other than ‘never’, they were asked ‘Why did you add baby cereal to your baby's bottle?’

Data analysis

Among our analytic population, we described the caregiver, infant and household sociodemographic characteristics of our sample overall and by early introduction of complementary foods/drinks. We assessed differences in proportions using χ² tests (p < 0.05). We estimated the cumulative prevalence of complementary food/drink introduction each month through 12 months of age. We described the early introduction of complementary foods/drinks overall and by food group. We also quantified the practice of adding infant cereal to bottles and caregiver reasoning for doing so. For a smaller subset of food and beverage groups that were commonly introduced early (defined as ≥100 participants introduced early), we described early introduction of complementary foods/drinks stratified by milk feeding type at Month 1. We used multi-variable logistic regression models to assess the association of milk feeding type and early introduction of complementary foods/drinks. We selected potential confounding variables a priori (from the previously described list of caregiver, infant and household characteristics) for consideration in model selection and used hierarchical backwards elimination to determine the final model (Kleinbaum & Klein, 2010). We adjusted the odds ratios (ORs) for caregiver race/ethnicity, age, education, marital status, parity and household poverty levels at time of enrollment. We chose the exposure reference group of fully breastfeeding based on the lowest prevalence of early introduction.

We used the survey procedures in SAS software (version 9.4, SAS Institute Inc., 2013) to account for the study's complex sample design. Survey weights were used so that results represented national estimates of the WIC population under study. Briefly, weights are assigned to each person sampled and are a measure of the number of people in the population represented by that sampled person (Westat, 2021). We present unweighted sample sizes and weighted estimates (percentages, ORs) throughout. Analyses were weighted using the Months 1 and 3 interview combined weight, which adjusts for differential probability of selection and non-response. We selected this weight because milk feeding type, a criterion for analytic eligibility, was collected at Month 1, and the early introduction of complementary food/drink would have occurred during this period.

RESULTS

Almost half of the study sample identified as Hispanic (46%) (Table 1). The majority of participants were 25 years or younger (52%), had a high school education or less (62%), were parous (58%), or were at ≤75% of the 2013 poverty guidelines when enrolling in WIC (62%). At 1 month of age, 31% of infants were fully breastfed, 30% were partially breastfed and 40% were fully formula fed. In bivariate analyses, there were demographic differences observed by early introduction of complementary foods/drinks by caregiver race/ethnicity, age, education, marital status, parity, milk feeding type at Month 1 and poverty level at enrollment. For instance, participants who were fully breastfeeding at Month 1 were less likely to introduce foods early.

Table 1 Demographic characteristics of caregiver-child dyads in the WIC Infant and Toddler Feeding Practices Study-2 (unweighted n = 3310), by early introduction of complementary foods/drinks (<4 months).^a

	Overall		Early introduction (<4 months) n = 1307		No early introduction (≥4 months) n = 2003
	n	%	n	%	n	%
Caregiver characteristic
Race			p = 0.02
Non-Hispanic White	1068	28.5	465	31.2	603	26.8
Non-Hispanic Black	809	20.1	346	21.7	463	19.1
Hispanic	1245	45.8	438	43.0	807	47.5
Other races/ethnicities	188	5.6	58	4.1	130	6.6
Age			p = 0.002
16–19 years	389	11.9	187	15.5	202	9.6
20–25 years	1381	40.4	578	42.0	803	39.5
26 years or older	1540	47.7	542	42.5	998	50.8
Education			p = 0.04
≤High school	2020	61.8	843	64.5	1177	60.1
>High school	1290	38.2	464	35.5	826	39.9
Marital status			p < 0.0001
Married	999	32.1	327	27.1	672	35.3
Unmarried	2311	67.9	980	72.9	1331	64.7
Parity			p = 0.008
Nulliparous	1400	42.3	617	48.3	783	38.5
Parous	1910	57.7	690	51.7	1220	61.5
Infant characteristic
Preterm statusb			p = 0.50
Preterm	326	9.8	139	10.7	187	9.3
Not preterm	2726	82.5	1071	81.5	1655	83.1
Missing	258	7.7	97	7.8	161	7.6
Delivery type			p = 0.69
Vaginal	2230	66.0	875	66.4	1355	65.8
Caesarean	1080	34.0	432	33.6	648	34.2
Milk feeding type at Month 1			p < 0.0001
Fully breastfed	1001	30.8	307	23.8	694	35.1
Partially breastfed	933	29.6	373	30.3	560	29.2
Fully formula fed	1376	39.6	627	45.8	749	35.7
Household characteristic
Poverty level at enrollment			p = 0.04
≤75% of 2013 poverty guidelines	2063	62.3	846	64.9	1217	60.7
76%–130% of 2013 poverty guidelines	906	26.9	348	26.3	558	27.3
130% of 2013 poverty guidelines	341	10.8	113	8.8	228	12.0

Abbreviations: n, number; WIC, Special Supplemental Nutrition Program for Women, Infants and Children.

a

P value determined by second-order χ² test. Estimates (%) are weighted to account for complex survey design while the sample sizes (n) are unweighted.

b

Preterm birth defined as <37 weeks gestation.

By 1 month of age, 9% of infants had been introduced to complementary foods/drinks (Figure 2). Thirty-eight percent of infants were introduced early to complementary foods/drinks (by 4 months of age). Over half (62%) of infants were introduced to complementary foods/drinks by 5 months of age, and by 7 months, almost all (97%) infants had been introduced to complementary foods/drinks. Among all the food and drink items examined, plain water was most commonly introduced early (24%) (Table 2). Seventeen percent of infants were introduced early to infant cereal. Of those infants introduced early to infant cereal, 45% had infant cereal added to their bottle; this practice was more common among partially breastfed and fully formula fed infants than fully breastfed infants, and the most common reason cited was to make the infant full (data not shown). Fruit juice was introduced early to 7% of infants. Six and five percent of infants were introduced early to fruits and vegetables, respectively. Less than 1 percent of infants were introduced early to soft drinks, cow's milk, sweets, other dairy products, meats, beans, salty snacks, eggs, other cereal or peanut butter.

Table 2 Prevalence of early introduction of complementary foods/drinks (<4 months) by food and drink group in the WIC Infant and Toddler Feeding Practices Study-2 (unweighted n = 3310).^a

Food and drink groups	Prevalence of early introduction (<4 months)
	n	%	(95% CI)
Any solid or liquid	1307	38.2	(35.3, 41.0)
Any liquid	937	28.2	(25.3, 31.2)
Plain water	804	24.3	(21.3, 27.3)
Fruit juice	248	7.0	(5.7, 8.3)
Other drinks or liquids	68	2.3	(1.7, 2.9)
Sweetened beverages	51	1.4	(0.8, 2.1)
Soft drinks	25	0.7	(0.4, 0.9)
Cow's milk	16	0.4	(0.1, 0.6)
Any solid	742	20.1	(17.2, 22.9)
Infant cereal	637	16.7	(13.8, 19.6)
Fruit	211	5.8	(4.8, 6.9)
Vegetables	166	4.6	(3.7, 5.5)
Sweets	30	0.7	(0.4, 1.1)
Dairy products other than milk	28	0.8	(0.5, 1.1)
Meats	26	0.7	(0.3, 1.1)
Beans	21	0.6	(0.3, 0.9)
Salty snacks	20	0.5	(0.2, 0.8)
Eggs	14	0.4	(0.2, 0.7)
Other cereal	13	0.4	(0.2, 0.6)
Peanut butter	11	0.3	(0.1, 0.6)

Abbreviations: CI, confidence interval; n, number; WIC, Special Supplemental Nutrition Program for Women, Infants and Children.

a

Estimates (%) are weighted to account for complex survey design while the sample sizes (n) are unweighted.

Forty-four percent of fully formula-fed infants and 39% of partially breastfed infants were introduced early to complementary foods/drinks, compared with 30% of fully breastfed infants (Table 3). In multi-variable logistic regression models, fully formula fed infants were 75% (adjusted OR: 1.75, 95% confidence interval [CI]: 1.36, 2.26) and partially breastfed infants were 57% (adjusted OR: 1.57, 95% CI: 1.18, 2.09) more likely to be introduced early to any complementary food/drink compared with fully breastfed infants. Fully formula fed and partially breastfed infants were more likely to be introduced early to infant cereal compared with fully breastfed infants. Fully formula-fed infants were also more likely to be introduced early to plain water than fully breastfed infants.

Table 3 Prevalence of early introduction of complementary foods/drinks (<4 months) by milk feeding type at Month 1 with adjusted^a ORs for the association of infant feeding type at Month 1 with early introduction of complementary foods/drinks in the WIC Infant and Toddler Feeding Practices Study-2 (unweighted n = 3310).^b,c

	Fully breastfed (n = 1001)			Partially breastfed (n = 933)			Fully formula fed (n = 1376)			Partially breastfed vs. fully breastfed		Fully formula fed vs. fully breastfed
	n	%	(95% CI)	n	%	(95% CI)	n	%	(95% CI)	aOR	(95% CI)	aOR	(95% CI)
Any solid or liquid	307	29.5	(26.1, 32.9)	373	39.1	(33.6, 44.7)	627	44.2	(39.5, 48.8)	1.57	(1.18, 2.09)	1.75	(1.36, 2.26)
Any liquid	229	22.2	(19.0, 25.5)	271	29.6	(23.0, 36.3)	437	31.8	(28.1, 35.6)	1.42	(1.02, 1.99)	1.57	(1.20, 2.07)
Plain water	188	18.6	(15.2, 21.9)	234	25.6	(18.8, 32.5)	382	27.8	(24.1, 31.5)	1.41	(0.97, 2.05)	1.62	(1.20, 2.20)
Fruit juice	56	5.2	(3.3, 7.1)	81	8.0	(6.1, 10.0)	111	7.6	(5.7, 9.6)	1.68	(1.02, 2.77)	1.32	(0.87, 2.00)
Any solid	150	14.0	(11.4, 16.7)	206	19.2	(15.7, 22.7)	386	25.4	(21.2, 29.6)	1.63	(1.25, 2.13)	1.87	(1.46, 2.40)
Infant cereal	117	10.5	(8.1, 12.9)	182	16.1	(12.8, 19.4)	338	22.0	(17.6, 26.3)	1.91	(1.43, 2.55)	2.16	(1.60, 2.93)
Fruit	37	3.8	(2.2, 5.4)	62	5.9	(4.0, 7.9)	112	7.3	(5.4, 9.2)	1.71	(0.90, 3.26)	1.77	(1.04, 3.04)
Vegetables	32	3.2	(1.8, 4.6)	51	4.9	(3.3, 6.4)	83	5.4	(3.8, 7.0)	1.58	(0.84, 2.95)	1.60	(0.91, 2.81)

Abbreviation: aOR, adjusted OR; CI, confidence interval; n, number; WIC, Special Supplemental Nutrition Program for Women, Infants and Children.

a

Adjusted for caregiver race/ethnicity, age, education, marital status, parity and household poverty level at time of enrollment.

b

Stratification by feeding type at Month 1 is shown only for food and drink groups with >100 infants with early introduction.

c

Estimates (% and aOR) are weighted to account for complex survey design while the sample sizes (n) are unweighted.

DISCUSSION

Among a cohort of infants enrolled in WIC, 38% were introduced early to complementary foods/drinks. The prevalence of early introduction of complementary foods/drinks differed by milk feeding type Month 1, with 44% of fully formula-fed infants being introduced early to complementary foods/drinks while 30% of fully breastfed infants were. These findings suggest that while early introduction of complementary foods/drinks is common, there are differences by milk feeding type in Month 1. Within the WIC program, the infant's milk feeding type determines the food package type received and might also affect the types of support offered to families (US Department of Agriculture, 2021). These findings suggest opportunities to support families participating in WIC to prevent early introduction of complementary foods/drinks and promote health.

In the first iteration of the WIC Infant Feeding Practices Study (WIC IFPS-1) in 1994, roughly 60% of caregivers introduced complementary foods before 4 months of age (Baydar et al., 1997). Twenty years later and following the 2009 revision of the WIC food package, our study using WIC ITFPS-2 data showed that a little less than 40% of caregivers introduced complementary foods/drinks before 4 months of age. Previous work has shown that WIC ITFPS-2 caregivers introduced foods from key food groups (infant cereal, fruits, vegetables and meats) at least 1 month later than in WIC IFPS-1 (Paolicelli et al., 2017). Our findings are encouraging when compared with the 1994 study, as they show a trend in infant feeding that is becoming more aligned with current recommendations (American Academy of Paediatrics Committee on Nutrition, 2019; US Department of Agriculture and US Department of Health and Human Services, 2020).

The prevalence of early introduction of complementary foods/drinks in this sample of WIC participants is 38%, which is higher than contemporary estimates reported in nationally representative samples of US children using 2016–2018 National Survey of Children's Health (NSCH) and 2009–2014 National Health and Nutrition Examination Surveys (NHANES) data (Barrera et al., 2018; Chiang et al., 2020). Both NSCH and NHANES collected data on infant feeding retrospectively, up to 5 years after infancy, while the current study of WIC participants used prospective data collection methods, which might account for some differences in the estimates of early introduction of complementary foods/drinks. The Infant Feeding Practices Study II, conducted in 2005–2007, collected data prospectively (like the present study) and found that 40% of infants were introduced early to complementary foods/drinks, very similar to the estimate of 38% in this study (Clayton et al., 2013).

Our examination of early introduction of complementary foods/drinks by specific food and beverage groups provides information on how food/beverage introduction practices may be contrary to current recommendations. For example, plain water was most commonly introduced early, as has been noted previously (Hornsby et al., 2021). Infants under 6 months of age who receive breast milk or infant formula as their main source of nutrition typically do not need additional water (World Health Organization, 2015). Water can fill up infant stomachs, which will then reduce the amount of nutrients they can receive at each feeding, which could slow growth and development. Further, extra water can disturb electrolyte balances, which can lead to seizures (Centers for Disease Control and Prevention, 1994). Another specific recommendation is related to cow's milk; children should not be introduced to cow's milk before 12 months of age (Centers for Disease Control and Prevention, 2022a). Before 12 months, cow's milk may put infants at risk for intestinal bleeding or overburden infants' kidneys because it has too many proteins and minerals (US Department of Agriculture and US Department of Health and Human Services, 2020; Ziegler, 2007). While less than half a percent of infants were introduced to cow's milk before 4 months, we found that over one-third of infants enrolled in WIC were introduced to cow's milk before 12 months old (data not shown). These results indicate that additional education and support is needed for WIC participants around the timing of introduction of beverages such as water and cow's milk.

Infant cereal was introduced early to roughly 1 in 6 infants enrolled in WIC and was the solid food group most commonly introduced early, as has been noted before (Hornsby et al., 2021). While there is no recommendation for introducing foods in a particular order, it is recommended to start with iron- and zinc-rich or fortified foods like infant cereal (US Department of Agriculture, 2019) once the infant is of an appropriate age. The habit of adding infant cereal to an infant's bottle is a long-standing one (Macknin, 1989) and was implicated in almost half of early infant cereal introductions in our present study. The addition of infant cereal to an infant's bottle is discouraged by the DGAs as it could increase the risk of choking and has not been shown to help infants sleep longer (US Department of Agriculture and US Department of Health and Human Services, 2020). One study found that almost one-third of low-income mothers report receiving advice from their social support network to add cereal to their infant's bottle (Ashida et al., 2016). Therefore, education and support for WIC participants could focus on increasing information provided prenatally or earlier in infancy that is supportive of recommendations on appropriate timing of infant cereal and complementary food/drink introduction. Moreover, families may need support in addressing some of the reasons foods are being introduced early, including understanding infant feeding cues and ways to help soothe infants having trouble falling asleep.

Previous work has identified the association of no breastfeeding initiation or short breastfeeding duration with the early introduction of complementary foods/drinks (Barrera et al., 2018; Chaparro & Anderson, 2021; Chiang et al., 2020; Clayton et al., 2013; Doub et al., 2015; Klag et al., 2015; Kuo et al., 2011; Scott et al., 2009), which our study also showed in bivariate and adjusted analyses of milk feeding type at Month 1 with early introduction of any complementary food/drink and specific items (e.g., infant cereal). Reasons for early introduction of complementary foods/drinks noted in previous studies varied by milk feeding type; parents of formula-fed infants were more likely to report that ‘friends or relatives said my infant should begin eating solid food’ and ‘a doctor or other health care professional said my infant should begin eating solid food’ than were parents of breastfed infants (Clayton et al., 2013). The present study's findings on the lower prevalence of early introduction of complementary foods/drinks among fully breastfed infants suggest that many caregivers who are fully breastfeeding may be aware of the recommendation to exclusively breastfeed for 6 months (Meek et al., 2022). Similar messaging about the duration of exclusive milk feeding could be used for caregivers who are formula feeding or partially breastfeeding to prevent the early introduction of complementary foods/drinks.

Previous work regarding the early introduction of complementary foods/drinks suggests misconceptions of infants being old enough to begin eating solid food as the top reason for early introduction (Clayton et al., 2013). Such knowledge gaps around infant feeding patterns, nutritional needs, developmental milestones and common myths could be addressed during prenatal or early post-partum WIC visits or health care appointments. The WIC program is one of the many sources of information that caregivers use in making decisions around infant feeding, and evidence suggests that clinicians are a trusted source of information on complementary feeding decisions (Clayton et al., 2013). Parents report high recollection of early childhood nutrition guidance from clinicians; however, certain topics, such as the appropriate timing of complementary food/drink introduction, could be prioritized, and some subpopulations may need additional focus to improve receipt of messages (McGowan et al., 2022). Families may also receive information from other sources, including family, friends, childcare providers (Centers for Disease Control and Prevention, 2022c) and other programs (such as the Maternal, Infant and Early Childhood Home Visiting Program or Title V [Health Resources and Services Administration, 2022a, 2022b]). Ensuring clear and consistent messaging on paediatric feeding recommendations is key to helping families make the best decisions around optimal infant feeding practices (Sutter et al., 2018).

The early introduction of complementary foods/drinks is intrinsically linked to exclusive breastfeeding. Thus, promoting exclusive breastfeeding, which has improved since the 2009 revision of the WIC food package (Gross et al., 2023), is an important strategy to reduce the early introduction of complementary foods/drinks. In addition to the aforementioned guidance from clinicians and WIC programming, existing public health programs such as the Baby Friendly Hospital Initiative (Kramer et al., 2001) and Continuity of Care in Breastfeeding Support (National Association of County and City Health Officials, & US Breastfeeding Committee, 2021) could be implemented in more hospitals and communities and augmented to include more specific guidance about the introduction of complementary foods/drinks.

To our knowledge, this is the first study to describe the early introduction of complementary foods/drinks in a sample comprised entirely of WIC participants. It is important to document early dietary practices in this ongoing cohort study of WIC participants for future studies to examine the association of early complementary feeding with long-term health consequences in this population. Strengths of this study include its prospective study design, which can minimize recall bias for both the exposure and outcome. Additionally, this study collected detailed data on the introduction of 17 different food and drink groups, which can assist in identifying foods and drinks that are typically introduced early. This study is generalizable to WIC participants who speak English or Spanish, and who are enrolled in WIC sites that enroll ≥30 participants per month, though results may not be generalizable to WIC enrollees at smaller sites, such as those in rural areas.

This study has four main limitations. First, data at some survey months were missing or incomplete; 590 participants (13.5%) of the original sample of 4367 were excluded due to missing data at the Months 1 and 3 surveys, and an additional 467 participants were excluded due to missing or implausible outcome data and missing exposure and covariate data (in total, 1057 participants (24.2%) of the original 4367-person sample were missing data) (Figure 1). Missing data due to incomplete surveys or loss to follow up could lead to bias. For example, if the participants lost to follow up are systematically different from those who participate, we may under- or overestimate the prevalence of early introduction of complementary foods/drinks. We are not overly concerned about this bias as we conducted a sensitivity analysis (data not shown) to assess patterns in the missing data and found no differences between the sample included in the present study and the overall ITFPS-2 sample. Second, age at introduction of complementary foods/drinks was estimated through calculation of when a particular food was introduced; therefore, if the food first introduced was not one of the foods asked about in the survey, the prevalence of early introduction of complementary foods/drinks may be underestimated. Third, despite adjustment for several important covariates, residual and unmeasured confounding may be a concern (e.g., post-natal employment). Fourth, these data were collected in 2013–2014 and may not reflect current practices. While these data were collected following the 2009 revision to the WIC food package (Gross et al., 2023), this study's findings around the introduction of potentially allergenic complementary foods such as peanut products and eggs may not reflect current practices, as there was a paradigm shift in recommendations beginning in 2015 (Du Toit et al., 2015; Greer et al., 2019).

CONCLUSION

The WIC program improves the health and nutritional status of low-income women, infants and children by improving infant feeding practices and diet (Ritchie et al., 2010; US Department of Agriculture, 2013; Whaley et al., 2012). The results of the present study show that almost 2 in 5 infants enrolled in WIC were introduced early to complementary foods/drinks, which indicates additional education and support are needed on this infant feeding topic. The variation in early introduction of complementary foods/drinks by milk feeding type also indicates the need to prioritize counseling and support of exclusive breastfeeding and optimal complementary feeding practices. Consistent messaging of infant feeding recommendations can help families and caregivers as they navigate food introduction and transition to the family diet. Future studies could examine barriers to appropriate timing of complementary food/drink introduction and the most effective ways to communicate complementary feeding guidance.

AUTHOR CONTRIBUTIONS

Kristin J. Marks conceptualized and designed the study, carried out the analyses, drafted the initial manuscript and reviewed and revised the manuscript. Jasmine Y. Nakayama conceptualized and designed the study, analyzed the data and critically reviewed the manuscript for important intellectual content. Ellen O. Boundy, Ruowei Li and Heather C. Hamner conceptualized and designed the study and reviewed and revised the manuscript. All authors read and approved the final manuscript as submitted and agreed to be accountable for all aspects of the work.

ACKNOWLEDGEMENTS

The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflict of interest.

DATA AVAILABILITY STATEMENT

The WIC ITFPS-2 data set is publicly available.