Contribution of Maternal Obesity to Medically Indicated and Elective Formula Supplementation in a Baby-Friendly Hospital

Abstract

Objective:

Determine if maternal obesity increases use of medically indicated or elective formula in the context of a Baby-Friendly Hospital with high prevalence of obesity.

Study Design:

We conducted a secondary analysis of mothers who initiated breastfeeding of their term, singleton infant born at a Baby-Friendly community hospital in 2016. We defined medically indicated as formula given per physician order; and elective as formula given per maternal request. We used multinomial logistic regression to determine the odds ratio (OR) and 95% confidence interval (95% CI) for medically indicated and elective formula (each versus exclusive breastfeeding) by obesity status. We adjusted for available covariates and mediating conditions that may be exacerbated by obesity.

Results:

A total of 1,245 mothers met inclusion criteria, of which 41% were obese. Exclusive breastfeeding, medically indicated formula, and elective formula were 84% versus 70%, 5% versus 12%, and 11% versus 18%, in nonobese versus obese women, respectively. After adjusting for covariates, obesity significantly increased the risk for medically indicated (OR 2.6 [95% CI 1.7–4.1]) and elective (OR 2.0 [95% CI 1.5–2.8]) formula. After additionally adjusting for conditions exacerbated by obesity, the risk of medically indicated formula was attenuated by 48% (OR 1.7 [95% CI 1.02–2.7]), and there was little attenuation of the risk of elective formula (OR 1.8 [95% CI 1.3–2.6]).

Conclusions:

In a setting with high obesity prevalence and strong support for exclusive breastfeeding, obesity accounted for 36% of medically indicated formula and 21% of elective formula use. In this era of globally increasing maternal obesity prevalence, there is an urgent need to develop successful strategies for supporting breastfeeding that goes above and beyond standard Baby-Friendly approaches.

Introduction

Reducing the proportion of infants who receive formula supplementation within the first 2 days of life and increasing the proportion of infants who are exclusively breastfed through 3 and 6 months postpartum are among the United States Healthy People 2020 goals.¹ In addition, the Joint Commission, which accredits and certifies health care organizations in the United States, requires birth hospitals to track exclusive breastfeeding rates.² This Joint Commission mandate is consistent with research demonstrating that in-hospital formula use increases the risk of breastfeeding discontinuation.³ The Baby-Friendly Hospital Initiative (BFHI) was developed to provide a framework for hospitals to optimally support breastfeeding, including avoidance of unnecessary formula supplementation.⁴ Multiple studies have demonstrated an association between implementation of the BFHI and improved breastfeeding rates.^5–7

Maternal obesity is associated with shortened exclusive breastfeeding duration^8–10; thus, settings with high maternal obesity rates may have difficulty achieving high exclusive breastfeeding rates, even when best practices are followed to optimize breastfeeding promotion and support during the maternity stay. Furthermore, it is unknown to what extent any difference in exclusive breastfeeding between obese and nonobese mothers who initiate breastfeeding can be attributed to maternal choice to supplement with formula versus medically indicated formula use.

Our objectives were to determine if maternal obesity contributes significantly to medically indicated or elective formula use among mothers who initiate breastfeeding in the context of a certified Baby-Friendly hospital.

Materials and Methods

Setting

The setting is a community hospital in southwestern Ohio that serves the eastern Cincinnati metropolitan area. It is a family birthing center with a Level 2 special care nursery and has been certified Baby-Friendly⁴ since 2003. During the year of data collection for this study, 56% of maternity clients were privately insured, 42% were publically insured, and 2% were self-pay. All newborn care at the study hospital is managed by a network of physicians under the umbrella of Cincinnati Children's Hospital Medical Center, and this network strives to follow consistent protocols in the care of newborns, including medical indications for formula use. Examples of long-standing Baby-Friendly policies at the study hospital include the following: (1) mandatory training in the hospital's breastfeeding policy and its implementation for all birth center nursing staff and physicians, (2) immediate initiation of skin-to-skin contact for all stable newborns birthed vaginally or by Caesarean delivery, (3) facilitating timely initiation of breast pumping whenever newborn is admitted to the special care nursery or transferred out to higher level care, (4) requiring a physician's order for medically indicated formula supplementation, (5) all newborns room-in with their mother unless admitted to the special care nursery or transferred out to higher level care, and (6) all mothers are shown how to breastfeed, and all families are educated on the importance of breastfeeding exclusively, breastfeeding frequently in response to early feeding cues, avoiding pacifiers, and avoiding formula supplements unless medically indicated. As part of continuous quality improvement, all study hospital staff are regularly updated on the hospital's exclusive breastfeeding rate and the distribution of formula use reasons. The study hospital also maintains an outpatient breastfeeding follow-up clinic that is staffed with an International Board Certified Lactation Consultant (IBCLC).

Study design

We analyzed deidentified data obtained from a prospectively maintained continuous quality improvement dataset that captured all babies born between January 1, 2016, and December 31, 2016. at the study hospital. The Institutional Review Boards of the University of Cincinnati, Cincinnati Children's Hospital Medical Center, and Jewish Hospital (which serves as the review board for the study hospital), approved this study as nonhuman subjects research.

Source dataset and key variables used in analysis

The source dataset was prospectively built as supervised by a single nurse assigned to collect quality improvement data on the Family Birth Center unit at the study hospital. On a daily basis, the unit clerk would generate a list of mother–baby pairs discharged from the maternity unit on the previous day. Select fixed fields from the medical record were recorded on a quality improvement database. On a regular basis, the nurse would verify fixed field data extraction and then review the newborn Intake and Output Flowsheet to assess feeding status during the birth hospitalization. Upon encountering the first documentation of formula use for a given newborn, the nurse would review the physician note and nurse note. As per hospital policy, physician recommendation to supplement with formula must be documented with the reason specified, either directly in the physician note or in a nurse note where the nurse summarizes the physician recommendation and reason. In situations where the infant required more calories or volume than they received from direct breastfeeding, mothers were assisted in expressing their breast milk before using formula supplement. Hospital policy also stipulates that whenever maternal request for formula is not medically indicated, maternity nurses and/or IBCLCs provide education on advantages of exclusive breast milk and encourage exclusive breast milk feeding, including offering assistance to mothers in providing expressed breast milk when appropriate, and document this activity. Maternal reason for requesting formula was not recorded in the quality improvement database. The nurse manager was alerted to any lapse in documentation of formula use. An audit determined that 98% of newborns had formula use reason documented in their medical record.

Using the above process, the quality improvement nurse coded infant feeding category during the birth hospitalization using the following categories: (a) breastfed exclusively throughout entire maternity stay; (b) breastfed with formula supplement as per mother's request; (c) initiated breastfeeding, but switched to formula as per mother's request; (d) formula fed exclusively from birth; (e) given formula supplement as per physician order due to excessive weight loss; (f) given formula supplement as per physician order due to hyperbilirubinemia, with or without excess weight loss; (g) given formula supplement as per physician order due to hypoglycemia; (h) given formula supplement as per physician order due to inadequate urine output, or (i) given formula supplement as per physician order due to other medically indicated reason. For the purposes of our analyses, we consolidated the above infant feeding codes into four categories: (1) exclusively breastfed throughout entire maternity stay (original category a), (2) breastfed with medically indicated formula supplement (original categories e–i), (3) breastfed with elective formula use during the maternity stay (original categories b or c), and (4) exclusively formula fed from birth (original category d).

Maternal weight and height in the source data set were extracted from fixed fields in the maternal maternity record and based on maternal report to the admitting nurse of weight at last prenatal visit and height. We estimated early postpartum maternal weight as follows: (reported weight from last prenatal visit – [infant birth weight × 2]), consistent with the components of gestational weight gain,¹¹ and as done previously.¹² We then calculated maternal body mass index (BMI, estimated maternal weight/reported maternal height, kg/m²) and categorized participants as underweight, normal weight, overweight, obese class I, obese class II, or obese class III according to World Health Organization BMI criteria.¹³

We excluded mother-infant dyads with missing data, data values that were biologically implausible, infants that were born premature (born before 37 completed weeks), infants that required transfer out of the birth hospital to a regional hospital with Level 3 or Level 4 neonatal care, infant did not survive to discharge, multiple gestation, planned adoption, or planned foster care placement.

Statistical analysis

First, we examined the distribution of maternal characteristics, labor and delivery characteristics, and infant characteristics within each of the four consolidated infant feeding categories. We used chi-square analysis and multiple variable regression to test for significant differences in breastfeeding initiation versus exclusive formula feeding by obesity status. For the remainder of the analysis, we excluded mothers who failed to initiate breastfeeding (i.e., they exclusively fed formula from birth).

We then used chi-square analysis to test for significant differences in the distribution of study variables between the feeding categories of exclusively breastfed, medically indicated formula supplement, and elective formula use. Next, we used multinomial logistic regression analysis to estimate the odds ratio (OR) and 95% confidence interval (95% CI) in a series of models with obesity as the primary exposure and either medically indicated formula supplement or elective formula use as the outcomes (versus exclusive breastfeeding in both model series). In Model 1, we determined the unadjusted odds of formula use by obesity status. In Model 2, we added available demographic covariates to Model 1 (maternal age, parity, race/ethnicity, infant sex, and gestational age). In Model 3, we added mediating conditions known to be exacerbated by obesity (a priori factors available in the dataset were hypertension in pregnancy, diabetes in pregnancy, labor induction or augmentation, cesarean delivery, low and high infant birthweight, and Apgar scores at 1 and 5 minutes).¹⁴ The dataset included information on newborn admission to study hospital's special care nursery, but this was not included in the logistic regression models due to admission not always temporally preceding formula use. The extent to which conditions that are exacerbated by obesity attenuated the relationship between obesity and formula use was calculated using the difference method, as follows: (100 × [(ln OR for obesity exposure in Model 2 – ln OR for obesity exposure in Model 3)/ln OR for obesity exposure in Model 2]), which represents the percentage of obesity's contribution to formula use that is mediated through conditions exacerbated by obesity.¹⁵ We also calculated the population attributable risk of obesity exposure on medically indicated and elective formula use as follows: (incidence of formula outcome in overall study population – incidence in nonobese group)/(incidence in overall study population), which is a measure of the proportion of formula use in the study population that is attributable to obesity, given the current prevalence of obesity in the study population.¹⁶

Results

Exclusions, including failure to initiate breastfeeding

A total of 1,833 mothers gave birth at the study hospital in 2016. Of the 1,833 mother-infant dyads, we excluded 74 due to missing or implausible maternal anthropometry data, 102 due to preterm birth, 25 requiring newborn transfer out of the study hospital to a regional Level 3 or Level 4 nursery, 18 fetal demise, 25 multiple gestations, and 1 case of an infant born to a surrogate mother, resulting in 1,588 mothers–baby pairs meeting inclusion criteria.

Overall, 22% of mothers failed to initiate breastfeeding. Obese and nonobese mothers failed to initiate breastfeeding at similar rates (23% and 20%, respectively, p = 0.17). Demographic covariates significantly associated with failing to initiate breastfeeding were younger maternal age and multiparity. Obesity persisted in not being significantly associated with failure to initiate breastfeeding in a model adjusted for all demographic covariates (maternal age, ethnicity, primiparity, and infant sex and gestational age, OR 1.19 [95% CI 0.93–1.52]).

Medically indicated formula supplementation among obese and nonobese mothers who initiated breastfeeding

The remainder of the analysis focused on the 1,245 mother-infant dyads who initiated breastfeeding. Distribution of study variables stratified by breastfeeding category is shown in Table 1. Of the 99 newborns who received medically indicated formula, the indicated reason was hypoglycemia in 43%, hyperbilirubinemia (with or without excess weight loss) in 25%, excess weight loss in 21%, low urine output in 8%, and other medical reason in 8%. In bivariate analysis, medically indicated formula use was 2.4-fold higher in obese versus nonobese women (12% versus 5%, respectively, p < 0.0001). In logistic regression analysis (Table 2), there was negligible change in the significantly higher odds of medically indicated formula supplementation in obese versus nonobese women in Model 1 (OR 2.5 [95% CI 1.7–3.9]) compared to after adjusting for demographic covariates (Model 2, OR 2.6 [95% CI 1.7–4.1]). In Model 3, the conditions exacerbated by obesity explained 48% of the effect of obesity on medically indicated formula supplementation; however, the odds of medically indicated formula supplementation persisted in being significantly higher in obese versus nonobese women (Model 3, OR 1.7 [95% CI 1.02–2.7]).

Table 1.

Characteristics of Sample Who Initiated Breastfeeding, Stratified by Infant Feeding Status During the Maternity Stay (N = 1,245)

Variable	Exclusively breastfed, n = 970 (78%)	Medically indicated formula, n = 99 (8%)	Elective formula use, n = 176 (14%)
Maternal characteristics, n (%)
Maternal age, years^**
<24	229 (24)	21 (21)	72 (41)
24–30	371 (38)	33 (33)	54 (31)
>30	370 (38)	45 (45)	50 (28)
BMI category^**^,a
Normal	250 (26)	15 (15)	34 (19)
Overweight	363 (37)	25 (25)	47 (27)
Obese class I	190 (20)	29 (29)	48 (27)
Obese class II	107 (11)	17 (17)	21 (12)
Obese class III	60 (6)	13 (13)	26 (15)
Obesity^**
Not obese	613 (63)	40 (40)	81 (46)
Obese	357 (37)	59 (60)	95 (54)
Parity^**
Multiparous	547 (56)	37 (37)	94 (53)
Primiparous	423 (44)	62 (63)	82 (47)
Diabetes in pregnancy^**
No	914 (94)	82 (83)	165 (94)
Yes	56 (6)	17 (17)	11 (6)
Hypertension in pregnancy^**
No	904 (93)	77 (78)	151 (86)
Yes	66 (7)	22 (22)	25 (14)
Labor induction/augmented^**
No	689 (71)	56 (57)	100 (57)
Yes	281 (29)	43 (43)	76 (43)
Delivery mode^**
Vaginal	721 (74)	55 (56)	120 (68)
Cesarean	249 (26)	44 (44)	56 (32)
Infant characteristics, n (%)
Infant sex^*
Female	469 (48)	35 (35)	75 (43)
Male	501 (52)	64 (65)	101 (57)
Race/ethnicity^*
African American	11 (1)	2 (2)	6 (3)
Asian, Indian	10 (1)	2 (2)	6 (3)
Biracial	12 (1)	0 (0)	1 (0.6)
Hispanic	2 (0.2)	0 (0)	0 (0)
White, non-Hispanic	923 (95)	95 (96)	156 (89)
Other	12 (1)	0 (0)	7 (4)
Gestational weeks^days,^**
Early term, 37⁰–38⁶	182 (19)	35 (35)	46 (26)
Full term, 39⁰–40⁶	712 (73)	52 (53)	113 (64)
Post term, ≥41⁰	76 (8)	12 (12)	17 (10)
Birthweight category^**
Low, <2,500 g	4 (0.4)	4 (4)	2 (1)
Normal, 2,500–3,999 g	851 (88)	65 (66)	160 (91)
Large, ≥4,000 g	115 (12)	30 (30)	14 (8)
Apgar at 1 minute
<8	119 (12)	18 (18)	24 (14)
≥8	851 (88)	81 (82)	152 (86)
Apgar at 5 minutes
<9	72 (7)	12 (12)	13 (7)
≥9	898 (93)	87 (88)	163 (93)
Special care nursery^**
Not admitted	948 (98)	60 (61)	164 (93)
Admitted	22 (2)	39 (39)	12 (7)

p < 0.05, ^**p < 0.005, Chi-square statistic.

BMI categories: normal, 18.5–24.9 kg/m²; overweight, 25.0–29.9 kg/m²; obese class I, 30.0–34.9 kg/m²; obese class II, 35.0–39.9 kg/m²; obese class III, ≥40.0 kg/m²; none were underweight (<18.5 kg/m²).

BMI, body mass index.

Table 2.

Odds of Medically Indicated or Elective Formula Supplementation Versus Exclusive Breastfeeding During the Maternity Stay Among Mothers Who Initiated Breastfeeding (N = 1,245)

Risk factor group versus referent group ^a	Supplement category	OR (95% CI) ^b
Risk factor group versus referent group ^a	Supplement category	Model 1	Model 2	Model 3
Obese BMI versus normal/overweight^c	Indicated	2.5 (1.7–3.9)^**	2.6 (1.7–4.1)^**	1.7 (1.02–2.7)^*
Obese BMI versus normal/overweight^c	Elective	2.0 (1.5–2.8)^**	2.0 (1.5–2.8)^**	1.8 (1.3–2.6)^**
Maternal age 18–23 years versus ≥30 years	Indicated	—	0.5 (0.3–0.9)^*	0.7 (0.4–1.2)
Maternal age 18–23 years versus ≥30 years	Elective	—	2.3 (1.5–3.6)^**	2.3 (1.5–3.5)^**
Maternal age 24–29 years versus ≥30 years	Indicated	—	0.6 (0.3–0.9)^*	0.7 (0.4–1.2)
Maternal age 24–29 years versus ≥30 years	Elective	—	1.1 (0.7–1.6)	1.1 (0.7–1.6)
Primiparous versus multiparous	Indicated	—	2.7 (1.7–4.3)^**	2.6 (1.6–4.3)^**
Primiparous versus multiparous	Elective	—	0.9 (0.7–1.3)	0.9 (0.6–1.2)
Infant non-white versus white, non-Hispanic^a	Indicated	—	0.9 (0.3–2.5)	0.7 (0.2–2.1)
Infant non-white versus white, non-Hispanic^a	Elective	—	2.5 (1.4–4.5)^**	2.5 (1.4–4.4)^**
Infant male versus female	Indicated	—	1.9 (1.2–3.0)^*	1.8 (1.1–2.8)^*
Infant male versus female	Elective	—	1.3 (0.9–1.8)	1.3 (0.9–1.8)
Early term versus full-term gestation^a	Indicated	—	2.7 (1.7–4.3)^**	3.2 (1.9–5.5)^**
Early term versus full-term gestation^a	Elective	—	1.6 (1.1–2.4)^*	1.7 (1.1–2.5)^*
Post-term versus full-term gestation^a	Indicated	—	1.8 (0.9–3.6)	1.7 (0.8–2.6)
Post-term versus full-term gestation^a	Elective	—	1.5 (0.8–2.7)	1.4 (0.8–2.6)
Diabetes in pregnancy versus none	Indicated	—	—	2.7 (1.4–5.2)^**
Diabetes in pregnancy versus none	Elective	—	—	0.9 (0.4–1.8)
Hypertension in pregnancy versus none	Indicated	—	—	2.0 (1.1–3.9)^*
Hypertension in pregnancy versus none	Elective	—	—	1.3 (0.7–2.3)
Labor induction versus none	Indicated	—	—	1.7 (1.02–2.8)^*
Labor induction versus none	Elective	—	—	1.7 (1.2–2.5)^**
Cesarean delivery versus vaginal	Indicated	—	—	1.9 (1.2–3.0)^*
Cesarean delivery versus vaginal	Elective	—	—	1.3 (0.9–2.0)
Low birthweight versus normal^a	Indicated	—	—	7.0 (1.5–33.1)^*
Low birthweight versus normal^a	Elective	—	—	2.2 (0.4–12.8)
Large birthweight versus normal^a	Indicated	—	—	3.8 (2.2–6.5)^**
Large birthweight versus normal^a	Elective	—	—	0.7 (0.4–1.2)
Apgar at 1 minute <8 versus 8–10	Indicated	—	—	1.0 (0.5–1.9)
Apgar at 1 minute <8 versus 8–10	Elective	—	—	1.0 (0.6–1.7)
Apgar at 5 minutes <9 versus 9–10	Indicated	—	—	0.7 (0.3–1.6)
Apgar at 5 minutes <9 versus 9–10	Elective	—	—	0.9 (0.4–1.8)

p < 0.05, ^**p < 0.005.

See Table 1 for group definitions

Logistic regression OR for given supplement category versus exclusive breastfeeding; Model 1: obesity only, Model 2: obesity + covariates, Model 3: obesity + covariates + mediators.

Obese, ≥30.0 kg/m²; normal/overweight, 18.5–29.9 kg/m²; none were underweight (<18.5 kg/m²).

CI, confidence interval; OR, odds ratio.

Elective formula use among obese and nonobese mothers who initiated breastfeeding

In bivariate analysis, elective formula use was 1.6-fold higher in obese versus nonobese women (18% versus 11%, respectively, p < 0.0001). As shown in Table 2, the odds of elective formula use were significantly higher in obese mothers compared to nonobese mothers in Model 1 (OR 2.0 [95% CI 1.5–2.8]), with no change after adjusting for demographic covariates (Model 2, OR 2.0 [95% CI 1.5–2.8]), and little attenuation after the addition of conditions exacerbated by obesity (Model 3, 12% attenuation (OR 1.8 [95% CI 1.3–2.6]).

Population attributable risk of obesity on formula use

Overall, 78% of the mothers who initiated breastfeeding succeeded in breastfeeding exclusively during the maternity stay, 8% supplemented for medically indicated reasons, and 14% elected to use formula. In this study population, obesity prevalence was 41%. As a result, 36% of medically indicated formula (PAR = (8–5)/8) and 21% of elective formula (PAR = (14–11)/14) could be directly or indirectly attributed to obesity.

Discussion

In this population of mothers who gave birth at a certified Baby-Friendly community hospital, obesity was significantly associated with both medically indicated and elective formula use. Among the 1,245 mothers who initiated breastfeeding, 78% exclusively breastfed during the maternity stay. According to the Centers for Disease Control's most recent Breastfeeding Report Card, 83% of mothers in the United States who initiated breastfeeding, did so exclusively during the first 2 days postpartum (whereas our data reflect the entire maternity stay).¹ Quite strikingly, we observed a 14 percentage point disparity in exclusive breastfeeding during the maternity stay in women without obesity (84%) compared to women with obesity (70%). Both medically indicated and elective formula use contributed significantly to this disparity. Adjustment for demographic covariates did not weaken the magnitude of risk of obesity on formula use for either medically indicated or elective reasons; thus, our calculated 36% population attributable risk of obesity on medically indicated formula use and 21% population attribute risk of obesity on elective formula use are unlikely to be due to confounding factors.

Conditions exacerbated by obesity explained almost half (48%) of the increased risk of medically indicated formula use in women who are obese. The conditions that are both exacerbated by obesity and were found to be significant risk factors for medically indicated formula were hypertension, diabetes, labor induction/augmentation, Cesarean delivery, and high birthweight. In contrast, labor induction/augmentation was the only exacerbated condition significantly predictive of elective formula use. Further research is needed to identify how to best support breastfeeding mothers with induced and/or augmented labor, as they may experience a more stressful or complicated pregnancy and postpartum recovery, thus making breastfeeding more difficult.¹⁷

The most common medical reason for medically indicated formula supplementation was newborn hypoglycemia, and 63% of medically indicated formula use due to hypoglycemia was attributable to newborns of obese mothers. The study hospital has since implemented a protocol for newborn hypoglycemia that incorporates use of oral dextrose gel to reduce the need for formula to treat this condition. One study found that using dextrose gel, in addition to feeding breast milk, was more effective in reversing neonatal hypoglycemia than breastfeeding alone, did not require an admission to a special care nursery, and still supported exclusive breastfeeding.¹⁸ This strategy may narrow the disparity in medically indicated formula use in infants of obese versus nonobese mothers. Of the study infants given medically indicated formula due to excess weight loss, 73% were born to obese mothers, which is consistent with a study by Dewey et al.,¹⁹ who reported higher maternal BMI to be a significant risk factor for excess newborn weight loss. This may reflect inadequate milk transfer as obesity is a significant risk factor for lactation difficulties,^20,21 including delayed lactogenesis,^22,23 poor latch due to large breast size,^21,24,25 lower prolactin response to suckling,²⁶ and maternal report of insufficient milk.²⁷ In the latter study, a mediation analysis found that “insufficient milk” significantly mediated the association between obesity and lower exclusive breastfeeding rates at 1 and 2 months postpartum. Furthermore, gestational diabetes and hypertension are indicators of poorer maternal metabolic health, which is emerging as a risk factor for delayed lactogenesis²⁸ and persistent low milk production.^29,30

Our study is not without limitations. We estimated early postpartum BMI status based on an equation that relied on maternal report of her last prenatal weight rather than directly weighing mothers in the early postpartum. Our approach may introduce random error or bias; and there may have been misreporting of height and weight by the mother, most likely resulting in underclassification of obesity.³¹ Second, we do not know the reasons for elective formula use or the timing of medically indicated formula use. For some newborns, medically indicated formula may have been a transient situation in the immediate postnatal period, such as due to hypoglycemia, and for others, the need for supplemental formula may have persisted beyond hospital discharge. Furthermore, the reason for formula use may have evolved over the birth hospitalization, and our data only reflect the reason for first use of formula. Also, we cannot determine if health care provider practices, either in their support for exclusive breastfeeding or determination of medically indicated formula need, were biased by maternal obesity status. Further research is needed of this important potential modifier of formula use by obesity status. Finally, data were collected from a single site, and thus may not be generalizable to sites with differing maternity care practices or formula supplementation protocols. Nonetheless, robust designation of medically indicated versus elective formula in the study dataset is supported by strong agreement between study independent variables and known risk factors for medically indicated versus elective formula use. For example, primiparity is a known cause of delayed lactogenesis,¹⁹ and we found a very strong association between primiparity and medically indicated, but not elective, formula use. In contrast, young maternal age is a known risk factor for weaker breastfeeding intentions³²; we found a very strong association between younger maternal age and elective formula use, but younger maternal age was protective against medically indicated use.

Conclusions

In summary, we believe that this may be the first study to examine prospectively documented medically indicated formula use based on physician's order in a setting with strong protocols for formula supplementation. We observed the risk of medically indicated formula use to be 2.4-fold higher in mothers who were obese compared to nonobese. This finding, combined with obesity being very common in the study hospital setting (41% obese), resulted in obesity contributing to 36% of the overall incidence of medically indicated formula use, and contributing to 21% of the overall incidence of elective formula use, despite strong maternity staff support for exclusive breastfeeding in the study setting. A greater understanding of the underlying drivers contributing to elective and medically indicated formula use in women who are obese is urgently needed. Furthermore, greater clarity is needed to determine if the breastfeeding support needs of obese women who supplement for medically indicated reasons are different from the needs of women who supplement for elective reasons. It is possible that an important proportion of the obese mothers who are supplementing for “elective” reasons is perceiving a subclinical deficit in their milk production and/or milk transfer, thus anticipating an imminent medical indication for formula supplementation. For example, a previous analysis reported biochemical evidence of significantly less progress toward copious milk production at 7 days postpartum in exclusively breastfeeding mothers, who had insufficient milk supply concerns even though there were no clinical sign of insufficient milk.³³ In conclusion, even though BFHI is an essential component of breastfeeding support for all new mothers, there is a critical need to identify which obese women will need tailored interventions that go above and beyond BFHI, and what the components of these tailored interventions should specifically be.

Footnotes

Acknowledgments

We want to thank Mercy Anderson Hospital for making these data available, the Ruth E. Rosevear Endowment for making this research study possible, and Erin Wagner, MS, at the University of Cincinnati for helping contribute to the development of the protocol and for her helpful insight on this study.

Disclosure Statement

The authors declare no conflicts of interest. No external funding was received in the development of this article. The senior author (L.A.N.R.) is supported by the Ruth Rosevear Endowment, which has no influence on the conduct or results reported for this article.

References

Centers for Disease Control. Breastfeeding Report Card. 2018. Available at www.cdc.gov/breastfeeding/data/reportcard.htm (accessed September 4, 2018).

Feldman-Winter

, Douglass-Bright

, Bartick

, et al. The new mandate from the joint commission on the perinatal care core measure of exclusive breast milk feeding: Implications for practice and implementation in the United States. J Hum Lact, 2013; 29:291–295.

Chantry

, Dewey

, Peerson

, et al. In-hospital formula use increases early breastfeeding cessation among first-time mothers intending to exclusively breastfeed. J Pediatr, 2014; 164:1339.e5–1345.e5.

WHO/UNICEF (Joint Statement). Protecting, Promoting, and Supporting Breastfeeding: The Special Role of Maternity Services. Geneva: World Health Organization, 1989.

Declercq

, Labbok

, Sakala

, et al. Hospital practices and women's likelihood of fulfilling their intention to exclusively breastfeed. Am J Public Health, 2009; 99:929–935.

Murray EK Ricketts

, Dellaport

. Hospital practices that increase breastfeeding duration: Results from a population-based study. Birth, 2007; 34:202–211.

Perez-Escamilla

, Martinez

, Segura-Perez

. Impact of the Baby-Friendly Hospital Initiative on breastfeeding and child health outcomes: A systematic review. Matern Child Nutr, 2016; 12:402–417.

, Jewell

, Grummer-Strawn

. Maternal obesity and breast-feeding practices. Am J Clin Nutr, 2003; 77:931–936.

Baker

, Michaelsen

, Sorensen

, et al. High prepregnant body mass index is associated with early termination of full and any breastfeeding in Danish women. Am J Clin Nutr, 2007; 86:404–411.

10.

Winkvist

, Brantsaeter

, Brandhagen

, et al. Maternal prepregnant body mass index and gestational weight gain are associated with initiation and duration of breastfeeding among Norwegian mothers. J Nutr, 2015; 145:1263–1270.

11.

Institute of Medicine (U.S.) subcommittee on Nutrition During Pregnancy. Nutrition During Pregnancy. Washington, DC: National Academy Press, 1990.

12.

Nommsen-Rivers

, Mastergeorge

, Hansen

, et al. Doula care, early breastfeeding outcomes, and breastfeeding status at 6 weeks postpartum among low-income primiparae. J Obstet Gynecol Neonatal Nurs, 2009; 38:157–173.

13.

WHO Consultation on Obesity. Obesity: Preventing and managing the global epidemic. World Health Organ Tech Rep Ser, 2000; 894:i–xii, 1–253.

14.

Catalano

, Shankar

. Obesity and pregnancy: Mechanisms of short term and long term adverse consequences for mother and child. BMJ, 2017; 356:j1.

15.

Vanderweele

, Vansteelandt

. Odds ratios for mediation analysis for a dichotomous outcome. Am J Epidemiol, 2010; 172:1339–1348.

16.

Szklo

, Nieto

Epidemiology: Beyond the Basics. Gaithersburg, MD: Aspen Publishers, Inc., 2000.

17.

Chen

, Nommsen-Rivers

, Dewey

, et al. Stress during labor and delivery and early lactation performance. Am J Clin Nutr, 1998; 68:335–344.

18.

Bennett

, Fagan

, Chaharbakhshi

, et al. Implementing a protocol using glucose gel to treat neonatal hypoglycemia. Nurs Womens Health, 2016; 20:64–74.

19.

Dewey

, Nommsen-Rivers

, Heinig

, et al. Risk factors for suboptimal infant breastfeeding behavior, delayed onset of lactation, and excess neonatal weight loss. Pediatrics, 2003; 112:607–619.

20.

Kair

, Colaizy

. Obese mothers have lower odds of experiencing pro-breastfeeding hospital practices than mothers of normal weight: CDC Pregnancy Risk Assessment Monitoring System (PRAMS), 2004–2008. Matern Child Health J, 2016; 20:593–601.

21.

Garner

, McKenzie

, Devine

, et al. Obese women experience multiple challenges with breastfeeding that are either unique or exacerbated by their obesity: Discoveries from a longitudinal, qualitative study. Matern Child Nutr, 2017; 13.

22.

Nommsen-Rivers

, Chantry

, Peerson

, et al. Delayed onset of lactogenesis among first-time mothers is related to maternal obesity and factors associated with ineffective breastfeeding. Am J Clin Nutr, 2010; 92:574–584.

23.

Rasmussen

, Hilson

, Kjolhede

. Obesity may impair lactogenesis II. J Nutr, 2001; 131:3009S–3011S.

24.

Katz

, Nilsson

, Rasmussen

. Danish health care providers' perception of breastfeeding difficulty experienced by women who are obese, have large breasts, or both. J Hum Lact, 2010; 26:138–147.

25.

Hilson

, Rasmussen

, Kjolhede

. Excessive weight gain during pregnancy is associated with earlier termination of breast-feeding among White women. J Nutr, 2006; 136:140–146.

26.

Rasmussen

, Kjolhede

. Prepregnant overweight and obesity diminish the prolactin response to suckling in the first week postpartum. Pediatrics, 2004; 113:e465–e471.

27.

O'Sullivan

, Perrine

, Rasmussen

. Early breastfeeding problems mediate the negative association between maternal obesity and exclusive breastfeeding at 1 and 2 months postpartum. J Nutr, 2015; 145:2369–2378.

28.

Nommsen-Rivers

, Dolan

, Huang

. Timing of stage II lactogenesis is predicted by antenatal metabolic health in a cohort of primiparas. Breastfeed Med, 2012; 7:43–49.

29.

Riddle

, Nommsen-Rivers

. A case control study of diabetes during pregnancy and low milk supply. Breastfeed Med, 2016; 11:80–85.

30.

Nommsen-Rivers

, Thompson

, Ward

, et al. Metabolic syndrome severity score identifies mothers with severely low milk output. In: Abstracts from The Academy of Breastfeeding Medicine 22(nd) Annual International Meeting Atlanta, Georgia, November 9–12, 2017, Vol. 12: Breastfeeding Medicine, 2017, pp. S1–S34.

31.

Russell

, Gillespie

, Satya

, et al. Assessing the accuracy of pregnant women in recalling pre-pregnancy weight and gestational weight gain. J Obstet Gynaecol Can, 2013; 35:802–809.

32.

Nommsen-Rivers

, Chantry

, Cohen

, et al. Comfort with the idea of formula feeding helps explain ethnic disparity in breastfeeding intentions among expectant first-time mothers. Breastfeed Med, 2010; 5:25–33.

33.

Murase

, Wagner EA

J C

, hantry

, et al. The relation between breast milk sodium to potassium ratio and maternal report of a milk supply concern. J Pediatr, 2017; 181:294.e3–297.e3.