Women with Gestational Diabetes Mellitus Have Greater Formula Supplementation in the Hospital and at Home Despite Intention to Exclusively Breastfeed

Abstract

Background:

Women with gestational diabetes mellitus (GDM) have lower rates of exclusive breastfeeding compared with women without diabetes.

Objectives:

To assess associations between GDM and breastfeeding intentions and attitudes, formula supplementation, reasons for formula supplementation, and knowledge of type 2 diabetes mellitus (T2DM) risk reduction associated with breastfeeding among U.S. mothers.

Design/Methods:

Participants completed an online survey assessing infant feeding knowledge, attitudes, and practices; demographics; and pregnancy-related medical history. Multivariable logistic regression was used to estimate adjusted odds ratios for formula supplementation in the hospital and at home.

Results:

Of 871 respondents, a smaller proportion of women with GDM compared with women without diabetes intended to exclusively breastfeed. There were no differences between groups in attitudes toward public breastfeeding, attitudes toward breastfeeding beyond infancy, or actual duration of any breastfeeding. Approximately one in four participants believed that breastfeeding mothers may be less likely to develop T2DM, regardless of GDM status. Among those who intended to exclusively breastfeed, GDM was associated with higher odds of formula supplementation in the hospital (adjusted odds ratio [OR] 1.75, 95% confidence interval [CI] 0.97–3.18) and at home (adjusted OR 2.02, 95% CI 1.05–3.89). “Medical reasons,” which was reported as an important reason for formula supplementation, was reported more frequently by women with GDM.

Conclusions:

Women with GDM who intended to exclusively breastfeed had higher odds of in-hospital and at-home formula supplementation, cited medical reasons as a main reason for formula supplementation more often, and were largely unaware of T2DM risk reduction associated with breastfeeding.

Introduction

Gestational diabetes mellitus (GDM) is characterized by development of glucose intolerance during pregnancy.¹ GDM is associated with perinatal and postnatal morbidity, including preeclampsia,² preterm birth,³ neonatal hypoglycemia,⁴ and increased risk of type 2 diabetes mellitus (T2DM) for both mother and infant.^5,6 In 2020, 7.8 per 100 live births were affected by GDM in the United States, representing a 30% increase from 2016.⁷

Benefits of breastfeeding for infants include reduced risk of T2DM⁸ and obesity.⁹ For women who breastfeed, benefits include lower risk of cardiovascular disease,¹⁰ hypertension,¹¹ breast and ovarian cancers,^12,13 and T2DM.¹⁴ Given the protective association breastfeeding has with some health risks associated with GDM, it is recommended that women with GDM breastfeed their children.¹⁵ Evidence indicates that women with GDM have similar rates of breastfeeding initiation^16–19 but may have lower rates of exclusive breastfeeding (EBF)^{16–18,20,21} and shorter duration of any breastfeeding^17–19,22 compared with women with no diabetes mellitus (NDM).

We and others have investigated facilitators and barriers to breastfeeding for women with GDM and have described the high rates of formula supplementation in the hospital in this population.^20,23,24 Less described in the literature are the pattern of formula supplementation at home after hospital discharge, maternal perceptions of reasons for formula supplementation, and the extent to which differences in formula supplementation may be explained by differences in intention to EBF. Also, a critical question is whether women with GDM are informed of the reduction in risk of developing T2DM with breastfeeding. This question has not been addressed in a U.S. population and knowledge has not been compared between women with and without GDM.

Therefore, the purpose of this study was to address several gaps in current knowledge about breastfeeding outcomes and predictors in women with GDM by comparing infant feeding attitudes, intentions, and practices among women with GDM and women with NDM, with an emphasis on formula supplementation as a primary outcome.

Methods

Participants and study design

This cross-sectional study was completed in December 2023. Eligibility criteria included U.S. women aged 18 years or older with at least one child between the ages of 1 and 4 years. Potentially eligible participants were identified by Centiment, LLC (www.centiment.co), after which they received access to the 10–15-minute online survey, administered through Qualtrics, on their Centiment dashboard. Centiment recruits panels nationwide from a variety of sources, including social media sites such as Facebook and LinkedIn. Prior to answering any survey questions, participants viewed a consent form and provided their consent to participate in the study. Respondents received compensation through Centiment via PayPal, though they also had an option to donate their survey earnings to a nonprofit organization of their choice. Participants who were missing data for key variables or who indicated diagnosis of type 1 diabetes mellitus, T2DM, or multiple forms of diabetes were excluded from this analysis. The study was approved by the Fairfield University Institutional Review Board (protocol #4167).

Measures

All survey items pertaining to infant feeding intentions and behaviors, pregnancy, and birth asked the respondent to answer based on their experiences with their youngest child only.

Diagnosis of GDM in the most recent pregnancy was self-reported by participants, who selected “gestational diabetes” from a list of possible pregnancy and childbirth-related conditions that might apply to them. The two primary outcome measures were (1) whether the respondent’s youngest child received formula in the hospital and (2) whether the child received formula at home. In-hospital formula use was assessed by a single item in the survey, in which participants were asked if their baby received any formula in the hospital (“yes,” “no,” or “not sure”). Responses of “not sure” were set to missing. Formula use at home was likewise assessed through a single item asking participants if they ever fed their baby infant formula after leaving the hospital (“yes” or “no”).

Secondary outcome measures included duration of any breastfeeding, intention to breastfeed for the first 3 months of the infant’s life, intention to exclusively breastfeed for the first 3 months of the infant’s life, knowledge of the benefit of breastfeeding related to T2DM risk reduction, attitudes toward breastfeeding in front of other people, and attitudes toward breastfeeding for longer than 1 year.

To ascertain duration of any breastfeeding, we first asked participants whether they were still breastfeeding. For those who responded “no,” we also asked how old their child was when they stopped breastfeeding (<1, 1–2, 3–4, 5–6, 7–9, 10–12, or more than 12 months). Those who were still breastfeeding were considered to have breastfed more than 12 months. Intention to breastfeed and intention to exclusively breastfeed were assessed using the same single item in the survey, in which participants were asked how they intended to feed their infant for the first 3 months (“formula only,” “breast milk only,” “a combination of breast milk and formula,” and “I was not sure”). Those who responded “breast milk only” were considered to intend to exclusively breastfeed and those who responded either “breast milk only” or “a combination of breast milk and formula” were considered to intend to breastfeed. For knowledge of the T2DM-related benefit of breastfeeding, responses to the statement “If a mother breastfeeds her baby, she may be less likely to develop type 2 diabetes later in her life” were categorized as somewhat agree/strongly agree versus all other responses.

A score for attitudes toward breastfeeding in various settings that are not entirely private (for simplicity, “public breastfeeding”) was generated based on agreement on a 5-point Likert scale with four statements beginning with “It is appropriate for a woman to breastfeed…” and ending with “in a private home, with family or friends present, with a nursing cover,” “in a private home, with family or friends present, without a nursing cover,” “in a public space with a nursing cover,” and “in a public space without a nursing cover.” Higher scores indicated more agreement with these statements and therefore greater acceptance of public breastfeeding. The score ranged from 4 to 20. The score had acceptable internal consistency (Cronbach’s alpha 0.71).

A score for attitudes toward breastfeeding beyond infancy was generated in the same way, based on agreement on a 5-point Likert scale with three statements beginning with “It is appropriate for a woman to breastfeed…” and ending with “a toddler between the ages of 1 and 2 years,” “a child between the ages of 2 and 4 years,” and “a child older than age 4 years.” The score ranged from 3 to 15 and had good internal consistency (Cronbach’s alpha 0.80).

Other measures we assessed were race/ethnicity, age group, marital status, education, parity, participation in the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC; categorized as during or after pregnancy or never), annual household income, maternal prepregnancy weight and height (used to calculate body mass index [BMI]), and infant birth weight in pounds and ounces. Participants indicated whether each of the following applied to their most recent pregnancy and childbirth by checking all conditions that applied: C-section delivery, neonatal intensive care unit (NICU) admission, infant hypoglycemia, or infant preterm birth.

Statistical analysis

We calculated frequencies and percentages for categorical variables and means and standard deviations for continuous variables. We used medians and interquartile ranges to summarize breastfeeding attitudes scores. To identify potential confounders, we assessed bivariate associations between each candidate variable and GDM and between each candidate variable and each primary outcome measure (i.e., in-hospital formula and at-home formula) using Wilcoxon rank sum tests for ordinal variables and birth weight, which was not normally distributed, and chi-square tests or Fisher’s exact tests for nominal variables. We built separate multivariable logistic regression models for each primary outcome measure among all participants and for the subgroup of participants who intended to EBF. Possible confounders, defined as those associated with both GDM and the outcome measure dictated by directed acyclic graphs and at p < 0.10, were considered for inclusion in the multivariable models. We initially selected models to minimize Aikake’s information criterion. We then tested the addition of each removed covariate and other possible confounders not initially considered (income, education, and birth weight). The criterion for inclusion was a likelihood ratio test p value of less than or equal to 0.10 or a change in the estimated coefficient for GDM of 10% or more when the covariate was included. We retained maternal age and WIC participation, which was highly correlated with income (p < 0.01), in all models regardless of significance. The overall fit of logistic regression models was assessed using the Hosmer–Lemeshow test. Stata version 17.0 was used for all analyses, and two-tailed p values < 0.05 were considered statistically significant.

Results

A total of 958 individuals met eligibility criteria and provided consent to participate in the study. After excluding those with incomplete data for key variables (n = 97) and those who indicated diagnosis of type 1 diabetes (n = 12), T2DM (n = 18), or multiple forms of diabetes (n = 10), 871 participants comprised the analytic sample. Participants were 62.2% non-Hispanic White, 11.9% non-Hispanic Black, 15.7% Hispanic, and 10.1% other race or multiracial. Prevalence of GDM in the most recent pregnancy was 13.9% overall, 15.3% among non-Hispanic Whites, 5.8% among non-Hispanic Blacks, 16.1% among Hispanics, and 11.4% among all other racial and ethnic groups. Additional participant characteristics by GDM status are presented in Table 1. A larger proportion of participants with GDM participated in WIC during or after pregnancy and had higher BMIs compared with women without diabetes. Their infants had higher rates of NICU admission and hypoglycemia. There were no differences in age, education, marital status, parity, or household income. Infant birth weight and rate of infant preterm birth were slightly higher in the GDM group but not significantly so.

Table 1.

Sample Characteristics

Variable	No diabetes(n = 750) n (%)	Gestational diabetes(n = 121) n (%)	p
Race/ethnicity			0.05
Non-Hispanic White	459 (61.2)	83 (68.6)
Non-Hispanic Black	98 (13.7)	6 (5.0)
Hispanic	115 (15.3)	22 (18.2)
Other	78 (10.4)	10 (8.3)
Age			0.66
18–24	103 (13.7)	17 (14.1)
25–34	578 (77.1)	95 (78.5)
35–44	69 (9.2)	9 (7.4)
Education, n (%)			0.82
High school or less	262 (34.8)	38 (31.4)
Some college	266 (35.5)	54 (44.6)
College degree	222 (29.5)	29 (24.0)
Marital status			0.91
Married or living with a partner	579 (77.2)	94 (77.7)
Single, divorced, separated, or widowed	171 (22.7)	27 (22.3)
Parity			0.87
Primiparous	291 (38.8)	46 (38.0)
Multiparous	459 (61.2)	75 (62.0)
WIC participation			0.001
Yes	401 (53.5)	85 (70.3)
No	349 (46.5)	36 (29.8)
Annual household income^a			0.70
<$30,000	226 (30.1)	35 (28.9)
$30,000–<$50,000	155 (20.6)	26 (21.5)
$50,000–<$75,000	157 (21.0)	33 (27.3)
$75,000 or more	211 (28.2)	27 (22.3)
Prepregnancy body mass index			<0.01
<18.5	41 (5.5)	5 (4.1)
18.5–<25	300 (40.0)	32 (26.5)
25–<30	199 (26.5)	31 (25.6)
≥30	210 (27.9)	53 (43.8)
C-section delivery	267 (35.5)	38 (31.4)	0.37
NICU admission	93 (12.4)	23 (19.0)	<0.05
Infant hypoglycemia	39 (5.2)	13 (10.7)	0.02
Infant preterm birth	150 (20.0)	32 (26.5)	0.11
Infant birth weight (g), median 25th–75th percentile)^b	3231.8 (2835.0–3600.4)	3288.5 (2948.3–3572.0)	0.21

Income, no diabetes n = 749.

Birth weight, no diabetes n = 723, GDM n = 119.

GDM, gestational diabetes mellitus.

Approximately 83% of both GDM and NDM women intended to breastfeed for at least 3 months (Table 2). Women with GDM were somewhat less likely to intend to breastfeed exclusively for at least 3 months as compared with those without diabetes (45.5% versus 53.1%, p = 0.12). Infants of women with GDM more often received formula in the hospital and at home, whether or not the mother intended to exclusively breastfeed. There were no differences in attitudes toward public breastfeeding or attitudes toward breastfeeding beyond infancy by GDM status.

Table 2.

Infant Feeding Knowledge, Attitudes, and Intentions

All participants	No diabetes(n = 750) n (%)	Gestational diabetes(n = 121) n (%)	p
Intended to breastfeed for at least 3 months (including mixed feeding)	622 (82.9)	100 (82.6)	0.94
Intended to breastfeed exclusively for at least 3 months	398 (53.1)	55 (45.5)	0.12
Infant received formula in the hospital	459 (61.2)	86 (71.1)	0.04
Infant received formula at home	554 (73.9)	103 (85.1)	<0.01
Agreed or strongly agreed with the statement: if a mother breastfeeds her baby, she may be less likely to develop type 2 diabetes later in her life^a	190 (25.6)	33 (27.5)	0.67
Public breastfeeding attitudes score (median, 25th–75th percentiles)^b	16 (14–19)	16 (14–19)	0.69
Breastfeeding beyond infancy attitudes score (median, 25th–75th percentiles)^c	9 (7–11)	9 (6–11)	0.50

Among those who intended to breastfeed exclusively for at least 3 months	No diabetes(n = 398)	Gestational diabetes(n = 55)	p
Infant received formula in the hospital	161 (40.5)	32 (58.2)	0.01
Infant received formula at home	226 (56.8)	41 (74.6)	0.01

NDM n = 742, GDM n = 120.

NDM n = 746, GDM n = 120.

NDM n = 738, GDM n = 120.

GDM, gestational diabetes mellitus; NDM, no diabetes mellitus.

When comparing reasons for formula supplementation among participants who initiated breastfeeding, the most cited reasons were “I did not have enough milk or my baby wasn’t satisfied,” “my baby had trouble sucking or latching on,” and “breastfeeding was painful or uncomfortable” (Table 3). The only reason that was reported significantly more frequently in the GDM group as compared with the NDM group was “medical reasons (e.g., illness or medication).”

Table 3.

Reasons for Formula Supplementation among Participants Who Ever Breastfed and Ever Introduced Formula

Reason	No diabetes(n = 453)n (%)	Gestational diabetes(n = 86)n (%)	p
Breastfeeding was painful or uncomfortable	118 (26.1)	19 (22.1)	0.44
My baby had trouble sucking or latching on	177 (39.1)	38 (44.2)	0.38
Medical reasons	52 (11.5)	19 (22.1)	<0.01
I did not have enough milk or my baby wasn’t satisfied	255 (56.3)	41 (47.7)	0.14
My baby was not gaining enough weight	91 (20.1)	22 (25.6)	0.25
Breastfeeding was too inconvenient	35 (7.7)	6 (7.0)	0.81
I could not or did not want to breastfeed after going back to work	35 (7.7)	5 (5.8)	0.54
My baby lost interest or was old enough that breastfeeding was no longer important	16 (3.5)	4 (4.7)	0.54^a
I wanted someone else to feed my baby	35 (7.7)	8 (9.3)	0.62
Breastfeeding took too much time and/or effort	44 (9.7)	6 (7.0)	0.42
Other	37 (8.2)	5 (5.8)	0.46

Fisher’s exact test.

In unadjusted and age-adjusted analyses, GDM was associated with significantly increased odds of formula supplementation in the hospital and at home among all participants and in the subgroup with intention to exclusively breastfeed (Table 4). These associations were somewhat attenuated in the fully adjusted models, but all remained significant except for formula association in the hospital among all participants. Among those who intended to exclusively breastfeed, GDM was associated with 85% increased odds of formula supplementation in the hospital (adjusted odds ratio [aOR] 1.75, 95% confidence interval [CI] 0.97–3.18) and a two-fold increase in odds of formula supplementation at home (aOR 2.02, 95% CI 1.05–3.89).

Table 4.

Odds Ratios for the Associations between Gestational Diabetes Mellitus and Formula Supplementation

	Unadjusted	Age-adjusted	Fully adjusted^a
All participants (n = 871)
In-hospital formula supplementation	1.56 (1.02–2.37)	1.55 (1.02–2.36)	1.25 (0.80–1.93)^b
At-home formula supplementation	2.02 (1.20–3.43)	2.01 (1.19–3.41)	1.78 (1.04–3.04)
Participants who intended to exclusively Breastfeed (n = 453)
In-hospital formula supplementation	2.05 (1.16–3.63)	2.02 (1.14–3.59)	1.75 (0.97–3.18)
At-home formula supplementation	2.22 (1.18–4.22)	2.25 (1.18–4.30)	2.02 (1.05–3.89)

Adjusted for maternal age, WIC participation, preterm birth, NICU admission.

Further adjusted for pre-pregnancy BMI.

BMI, body mass index; NICU, neonatal intensive care unit.

Discussion

We identified several important differences between women with GDM and women with NDM in terms of infant feeding intentions and practices. Although there was no difference in intention to breastfeed, inclusive of mixed feeding, between groups, there was a nonsignificant but potentially meaningful difference in intention to exclusively breastfeed (45.5% among GDM women versus 53.1% among NDM women). These findings add to the limited existing research on breastfeeding intention in GDM, which has had conflicting results. In one previous study, there was no association between GDM and any breastfeeding intention.²¹ At least one study has suggested reduced intention to EBF among women with GDM,²⁴ whereas another found no association.²⁰

In unadjusted analyses, GDM was associated with increased odds of both in-hospital and at-home formula use among all participants and among those who intended to EBF. After adjusting for possible confounders, these relationships were somewhat attenuated, rendering the associations between GDM and in-hospital formula use not statistically significant. However, among those who intended to EBF, the magnitude of association between GDM and in-hospital formula use (aOR 1.75) was not negligible and, because of the modest size of this subgroup (n = 453), we cannot exclude the possibility of a type II error. GDM remained significantly associated with increased odds of at-home formula use, regardless of EBF intention. Across the multivariable models, GDM was more strongly associated with at-home formula use than with in-hospital formula use and was more strongly associated with formula use among those who intended to EBF than among the entire sample. These findings suggest that differences in formula use between GDM and NDM women are not due to differences in EBF intention. Furthermore, higher rates of formula supplementation among women with GDM who intend to EBF are not confined to the hospital stay, when infant hypoglycemia and other acute problems around the time of birth may lead to formula supplementation. However, because in-hospital formula supplementation is associated with lower likelihood of sustaining EBF over time,²⁵ efforts to reduce formula supplementation in the hospital may also reduce formula supplementation at home.

Social norms regarding breastfeeding are also potential predictors of breastfeeding outcomes.^26–28 For example, not wanting to breastfeed in public has been cited by 15–18% of U.S mothers as a reason for stopping breastfeeding before 6 months²⁶ and comfort breastfeeding in public has been associated with 63% increased odds of intending to EBF.²⁷ Although there is little quantitative research on attitude toward breastfeeding beyond infancy or its influence on breastfeeding outcomes, some qualitative research suggests that women frequently perceive and are negatively affected by the disapproval of others when breastfeeding longer than 1 year.^29,30 This was the first study to examine whether attitudes toward public breastfeeding or breastfeeding beyond infancy differ by GDM status. We found no differences between groups in these attitudes. Because these factors did not differ between groups, it is unlikely that they contributed to observed differences in infant feeding practices between GDM and NDM participants.

Approximately one in four participants agreed that breastfeeding is protective against T2DM and there was no difference in agreement between GDM and NDM women. We found only one previous study examining knowledge of T2DM risk reduction with breastfeeding among women with GDM.³¹ In that study, based in China, most participants (67 − 92%) knew of health risks of GDM (i.e., pregnancy-induced hypertension, macrosomia, and infant hypoglycemia). However, just 46.0% responded correctly to a question about whether breastfeeding could reduce the likelihood of T2DM, obesity, and cardiovascular disease. In another study in a general population of women with young children in Poland, Bednarek and colleagues³² found that 17.3% believed breastfeeding reduced the risk of T2DM for the mother. Neither of these previous studies compared knowledge among GDM versus NDM women. It is important for women with GDM to be informed of this benefit because they are at increased risk of T2DM⁵ and breastfeeding knowledge, inclusive of knowledge of maternal health benefits, may influence EBF rates.³³

Among women with GDM who initiated breastfeeding, 22.1% reported giving their infant formula for medical reasons and this was the only reason that differed by GDM status. Although other researchers have investigated reasons for breastfeeding cessation^34,35 and reasons for formula supplementation in the hospital,^36,37 we are not aware of any existing studies that have compared reasons for formula supplementation among GDM and NDM women. We found that the most common reasons for formula supplementation overall were perceived insufficient milk, infant difficulty sucking or latching on, pain or discomfort, and inadequate infant weight gain. These reasons are largely similar to the most common reasons for early breastfeeding cessation according to a 2019 review of 10 studies, which were perceived inadequate milk supply and breast or nipple pain.³⁵ Reasons for in-hospital formula supplementation appear to differ depending on whether hospital staff³⁶ or mothers³⁷ are asked. In their study of U.S. hospital staff, Bookhart et al.³⁶ found that the most common reasons for formula supplementation in the hospital were infant hypoglycemia (46.1%), infant weight loss (36.5%), request of mother (34.8%), and jaundice (27.1%). In another study of mothers in Croatia, the most common reported reasons for supplementation were insufficient milk (49.8%), infant crying (35.5%), and C-section (11.5%).³⁷ The authors of that study estimated that 24.6% of the reasons were medically acceptable. We do not have further information to describe why a significant proportion of breastfeeding women with GDM in our study believed they had medical reasons for feeding their infants formula. Although the question about reasons for formula use was placed directly after the question about formula use at home, some women may have responded with in-hospital supplementation in mind. In that case, the “medical reason” may have been infant hypoglycemia. However, alternatives to infant formula supplementation, such as oral dextrose gel,³⁸ colostrum expressed pre- or postbirth,³⁹ or donor milk,⁴⁰ could be considered in breastfed neonates experiencing clinically significant hypoglycemia. It is also possible that mothers whose infants experienced hypoglycemia in the early postnatal period could be sensitized to perceived medical concerns in the postpartum period, leading them to supplement with formula more quickly than those whose infants were not labeled medically problematic in the hospital. Alternatively, participants with GDM may have believed they had medical reasons for introducing formula at home that were entirely unrelated to the hospital stay. Further research is needed to understand what the specific medical reasons are and whether they are evidence-based.

Our findings must be considered in the context of the study’s limitations. Because the study was cross-sectional and participants were 1–4 years postpartum, there is a chance of both recall bias and nondifferential misclassification of GDM and/or infant feeding outcomes. For example, those who breastfed more or less than they originally planned may have misreported their prenatal breastfeeding intentions (recall bias) and recall of formula use and reasons for using formula may have been inaccurate overall (nondifferential misclassification). Measurement of GDM also may have been subject to misclassification as it was self-reported. Some respondents (n = 10), who were excluded from the study, indicated that they had had both GDM and one or more other types of diabetes, suggesting some lack of understanding about these diagnoses or error in reporting them. In addition, actual breastfeeding behavior may have influenced participants’ responses to questions about breastfeeding benefits, attitudes toward public breastfeeding, and attitudes toward breastfeeding beyond infancy. If that were the case, it would be expected that women with GDM, who were more likely to use formula, may have underreported their knowledge of breastfeeding benefits and their support for breastfeeding under various circumstances. We found no difference between groups in these variables, suggesting that less favorable breastfeeding knowledge and attitudes are unlikely to explain differences in formula supplementation by GDM status. Finally, all participants were members of an opinion panel and likely differ in some ways from the general population. For example, the prevalence of GDM in our sample was 13.9%, which is higher than the U.S. average of 7.8%.⁷ Women with GDM may have been more likely to be part of the opinion panel that we recruited from and/or to participate in the study. In our sample, prevalence of GDM was highest among Hispanic participants (16.1%) followed by non-Hispanic White participants (15.3%) and lowest among non-Hispanic Black participants (5.8%). In contrast, national estimates suggest that GDM incidence is 7.0% and 6.5% for non-Hispanic White and non-Hispanic Black mothers, respectively.⁷ This suggests that non-Hispanic White women may have been more likely than non-Hispanic Black women in our sample to be diagnosed with GDM or more likely to report their diagnosis.

Despite these limitations, this study contributes important new evidence to the medical and scientific understanding of infant feeding patterns and determinants among women with GDM. To our knowledge, this was the first study to compare women with GDM and women without diabetes in their awareness of the potential for breastfeeding to reduce the risk of T2DM, reasons for formula supplementation, and attitudes toward public breastfeeding and breastfeeding beyond infancy. We obtained a reasonably diverse sample, with representation of individuals with a range of household incomes and education levels. Our analyses were adjusted for several possible confounders.

Conclusions

We found that women with GDM had similar breastfeeding duration as women without diabetes but were more likely to report formula use, both in the hospital and at home. The association between GDM and formula supplementation appeared to be stronger among those who intended to exclusively breastfeed. Most women with GDM were not aware of the potential T2DM risk reduction associated with breastfeeding, and they were no more likely than women without diabetes to know about this benefit. We observed no differences in attitude toward public breastfeeding or attitude toward breastfeeding beyond infancy by GDM status. More women with GDM than women without diabetes reported introducing formula for medical reasons. Future research should focus on identifying ways to reduce in-hospital formula supplementation of infants born to women with GDM and to support continued EBF at home in this population.

Footnotes

Acknowledgments

The authors acknowledge and thank Dr. Bridget Hussain who provided statistical support.

Authors’ Contributions

K.N.D. conceived of and designed the study and collected data, analyzed data, wrote the initial draft of the article, and reviewed and edited the article. L.J. contributed to the initial draft of the article and reviewed and edited the article. S.N.T. contributed to the design of the study and reviewed and edited the article.

Disclosure Statement

The authors have no disclosures to make.

Funding Information

This study was supported by Fairfield University through the Elizabeth DeCamp McInerny Endowed Chair of Health Sciences.

References

Geno

, Cervinski

, Nerenz

, et al. Chapter 15—Pregnancy and the Fetus. In: Handbook of Diagnostic Endocrinology, (Third Edition). ( Winter

, Holmquist

, Sokoll

eds.) Academic Press; 2021; pp. 543–579; doi: 10.1016/B978-0-12-818277-2.00015-7

Östlund

, Haglund

, Hanson

. Gestational diabetes and preeclampsia. Eur J Obstet Gynecol Reprod Biol, 2004; 113(1):12–16; doi: 10.1016/j.ejogrb.2003.07.001

Hedderson

, Ferrara

, Sacks

. Gestational diabetes mellitus and lesser degrees of pregnancy hyperglycemia: Association with increased risk of spontaneous preterm birth. Obstet Gynecol, 2003; 102(4):850–856; doi: 10.1016/S0029-7844(03)00661-6

Zhao

, Liu

, Zhou

, et al. Identifying risk effectors involved in neonatal hypoglycemia occurrence. Biosci Rep, 2020; 40(3):BSR20192589; doi: 10.1042/BSR20192589

Ben‐Haroush

, Yogev

, Hod

. Epidemiology of gestational diabetes mellitus and its association with type 2 diabetes. Diabet Med, 2004; 21(2):103–113; doi: 10.1046/j.1464-5491.2003.00985.x

Bellamy

, Casas

J-P

, Hingorani

, et al. Type 2 diabetes mellitus after gestational diabetes: A systematic review and meta-analysis. Lancet, 2009; 373(9677):1773–1779; doi: 10.1016/S0140-6736(09)60731-5

Gregory

CWE

, Danielle

. Trends and characteristics in gestational diabetes: United States, 2016–2020. Natl Vital Stat Rep, 2022; 71(3):1–15; doi: 10.15620/cdc:118018

Horta

, De Lima

. Breastfeeding and type 2 diabetes: Systematic review and meta-analysis. Curr Diab Rep, 2019; 19(1):1; doi: 10.1007/s11892-019-1121-x

Marseglia

, Manti

, D’Angelo

, et al. Obesity and breastfeeding: The strength of association. Women Birth, 2015; 28(2):81–86; doi: 10.1016/j.wombi.2014.12.007

10.

Nguyen

, Jin

, Ding

. Breastfeeding and maternal cardiovascular risk factors and outcomes: A systematic review. PLoS One, 2017; 12(11):e0187923; doi: 10.1371/journal.pone.0187923

11.

, Wang

, Tang

, et al. Association between duration of breastfeeding and maternal hypertension: A systematic review and meta-analysis. Breastfeed Med, 2018; 13(5):318–326; doi: 10.1089/bfm.2017.0180

12.

Anonymous. Breast cancer and breastfeeding: Collaborative reanalysis of individual data from 47 epidemiological studies in 30 countries, including 50 302 women with breast cancer and 96 973 women without the disease. Lancet, 2002; 360(9328):187–195; doi: 10.1016/S0140-6736(02)09454-0

13.

Babic

, Sasamoto

, Rosner

, et al. Association between breastfeeding and ovarian cancer risk. JAMA Oncol, 2020; 6(6):e200421; doi: 10.1001/jamaoncol.2020.0421

14.

Pinho-Gomes

A-C

, Morelli

, Jones

, et al. Association of lactation with maternal risk of type 2 diabetes: A systematic review and meta-analysis of observational studies. Diabetes Obes Metab, 2021; 23(8):1902–1916; doi: 10.1111/dom.14417

15.

Much

, Beyerlein

, Roßbauer

, et al. Beneficial effects of breastfeeding in women with gestational diabetes mellitus. Mol Metab, 2014; 3(3):284–292; doi: 10.1016/j.molmet.2014.01.002

16.

Oza-Frank

, Moreland

, McNamara

, et al. Early lactation and infant feeding practices differ by maternal gestational diabetes history. J Hum Lact, 2016; 32(4):658–665; doi: 10.1177/0890334416663196

17.

Oza-Frank

, Chertok

, Bartley

. Differences in breast-feeding initiation and continuation by maternal diabetes status. Public Health Nutr, 2015; 18(4):727–735; doi: 10.1017/S1368980014000792

18.

Nguyen

PTH

, Pham

, Chu

, et al. Gestational diabetes and breastfeeding outcomes: A systematic review. Asia Pac J Public Health, 2019; 31(3):183–198; doi: 10.1177/1010539519833497

19.

Nguyen

PTH

, Binns

, Nguyen

, et al. Gestational diabetes mellitus reduces breastfeeding duration: A prospective cohort study. Breastfeed Med, 2019; 14(1):39–45; doi: 10.1089/bfm.2018.0112

20.

Haile

, Oza-Frank

, Azulay Chertok

, et al. Association between history of gestational diabetes and exclusive breastfeeding at hospital discharge. J Hum Lact, 2016; 32(3):NP36–NP43; doi: 10.1177/0890334415618936

21.

Finkelstein

, Keely

, Feig

, et al. Breastfeeding in women with diabetes: Lower rates despite greater rewards. A population‐based study. Diabet Med, 2013; 30(9):1094–1101; doi: 10.1111/dme.12238

22.

Baerug

, Sletner

, Laake

, et al. Recent gestational diabetes was associated with mothers stopping predominant breastfeeding earlier in a multi‐ethnic population. Acta Paediatr, 2018; 107(6):1028–1035; doi: 10.1111/apa.14274

23.

Doughty

, Abeyaratne

, Merriam

, et al. Self-efficacy and outcomes in women with diabetes: A prospective comparative study. Breastfeed Med, 2023; 18(4):307–314; doi: 10.1089/bfm.2022.0298

24.

Loewenberg Weisband

, Rausch

, Kachoria

, et al. Hospital supplementation differentially impacts the association between breastfeeding intention and duration among women with and without gestational diabetes mellitus history. Breastfeed Med, 2017; 12(6):338–344; doi: 10.1089/bfm.2017.0019

25.

Whipps

MDM

, Yoshikawa

, Demirci

, et al. Estimating the impact of in-hospital infant formula supplementation on breastfeeding success. Breastfeed Med, 2021; 16(7):530–538; doi: 10.1089/bfm.2020.0194

26.

, Fein

, Chen

, et al. Why mothers stop breastfeeding: Mothers’ self-reported reasons for stopping during the first year. Pediatrics, 2008; 122(Supplement_2):S69–S76; doi: 10.1542/peds.2008-1315i

27.

Stuebe

, Bonuck

. What predicts intent to breastfeed exclusively? Breastfeeding knowledge, attitudes, and beliefs in a diverse urban population. Breastfeed Med, 2011; 6(6):413–420; doi: 10.1089/bfm.2010.0088

28.

Scott

, Kwok

, Synnott

, et al. A comparison of maternal attitudes to breastfeeding in public and the association with breastfeeding duration in four European countries: Results of a cohort study. Birth, 2015; 42(1):78–85; doi: 10.1111/birt.12138

29.

Thompson

, Topping

, Jones

. ‘Surely you’re not still breastfeeding’: A qualitative exploration of women’s experiences of breastfeeding beyond infancy in the UK. BMJ Open, 2020; 10(5):e035199; doi: 10.1136/bmjopen-2019-035199

30.

Jackson

, Hallam

. ‘It’s quite a taboo subject’: An investigation of mother’s experiences of breastfeeding beyond infancy and the challenges they face. Women Health, 2021; 61(6):572–580; doi: 10.1080/03630242.2021.1938790

31.

Qian

, Duan

, Lin

, et al. Decision-making process of breastfeeding behavior in mothers with gestational diabetes mellitus based on health belief model. BMC Pregnancy Childbirth, 2023; 23(1):242; doi: 10.1186/s12884-023-05527-3

32.

Bednarek

, Bodys-Cupak

, Serwin

, et al. Mothers’ attitudes towards breastfeeding in terms of health safety and professional lactation education: A national survey of women. J Multidiscip Healthc, 2023; 16:3273–3286; doi: 10.2147/JMDH.S431576

33.

Zhang

, Zhu

, Zhang

, et al. What factors influence exclusive breastfeeding based on the theory of planned behaviour. Midwifery, 2018; 62:177–182; doi: 10.1016/j.midw.2018.04.006

34.

Odom

, Li

, Scanlon

, et al. Reasons for earlier than desired cessation of breastfeeding. Pediatr, 2013; 131(3):e726–e732; doi: 10.1542/peds.2012–1295

35.

Morrison

, Gentry

, Anderson

. Mothers’ reasons for early breastfeeding cessation. MCN Am J Matern Child Nurs, 2019; 44(6):325–330; doi: 10.1097/NMC.0000000000000566

36.

Bookhart

, Anstey

, Kramer

, et al. A nation-wide study on the common reasons for infant formula supplementation among healthy, term, breastfed infants in US hospitals. Matern Child Nutr, 2022; 18(2):e13294; doi: 10.1111/mcn.13294

37.

Boban

, Zakarija-Grković

. In-hospital formula supplementation of healthy newborns: Practices, reasons, and their medical justification. Breastfeed Med, 2016; 11:448–454; doi: 10.1089/bfm.2016.0039

38.

Harris

, Weston

, Signal

, et al. Dextrose gel for neonatal hypoglycaemia (The Sugar Babies Study): A randomised, double-blind, placebo-controlled trial. Lancet, 2013; 382(9910):2077–2083.

39.

Tozier

. Colostrum versus formula supplementation for glucose stabilization in newborns of diabetic mothers. J Obstet Gynecol Neonatal Nurs, 2013; 42(6):619–628.

40.

Ponnapakkam

, Rees

, Gallup

, et al. Supplementation-based hypoglycemia guidelines including donor breast milk reduce NICU admission. J Perinatol, 2021; 41(8):2088–2094.