Changes in mothers’ intended duration of breastfeeding from the prenatal to neonatal periods

Abstract

Background:

Although previous studies suggest that the intentions of mothers to breastfeed during pregnancy strongly predict actual breastfeeding practice, no studies have examined the changes in the intentions of mothers to breastfeed from the prenatal to neonatal periods. The purpose of this study was to examine changes in intended breastfeeding duration from the prenatal to neonatal periods, their association with actual duration, and predictors for shortened duration.

Methods:

The Infant Feeding Practices Study II was a longitudinal study of mothers in the United States. Changes to intended breastfeeding duration were calculated as the difference from prenatal to neonatal reports (months); we compared this change to actual breastfeeding duration. By using multivariable logistic regression, we identified maternal characteristics associated with a shortened breastfeeding intention.

Results:

Of 1,780 women, 43.7% had no change to intended breastfeeding duration, 35.0% had a shorter intended duration, and 21.3% had a longer intended duration. Mothers with shortened intended duration also had shorter actual duration (P <.001). Women of Hispanic ethnicity, with a prepregnancy body mass index of ≥30 kg/m², who were primiparous, and who smoked prenatally, had increased odds of shortening their breastfeeding intention from prenatal to neonatal reports. A maternal age of ≥35 years was associated with decreased odds of shortened breastfeeding intention.

Discussion:

Approximately one in three women shorten their intended breastfeeding duration during the early postpartum period, which negatively affects the actual duration of their breastfeeding. Women may need additional support during the early postpartum period to meet their prenatal breastfeeding intentions.

Introduction

More than half of women decide how long they intend to breastfeed their infant during the prenatal period.^{1, 2} There are several factors that contribute to a woman’s breastfeeding plans, including her knowledge and attitudes towards breastfeeding,³ employment status,^{4, 5} social support,⁶ parity,⁵ maternal confidence,^{2, 5} and her comfort with breastfeeding.⁷ A mother’s intention to breastfeed is a strong predictor of her actual breastfeeding practices^{8, 9} because women who prenatally intend to breastfeed are more likely to initiate and continue breastfeeding.^{1, 10} Despite these plans, many mothers do not achieve their prenatal breastfeeding intentions^{1, 9, 11–13} and do not breastfeed for as long as they want.¹⁴

A national study among mothers in the United States (US) who had given birth to singletons between 2011–2012 has shown that most women (81%) intend to breastfeed, either exclusively or in combination with formula, at the end of their pregnancy; however, only 76% were breastfeeding at 1 week after birth,¹⁵ suggesting mothers may be changing their breastfeeding intentions between the prenatal and neonatal periods. However, no studies are available, so far, that specifically address changes to prenatal breastfeeding intentions, or how such changes may affect actual breastfeeding duration, or the factors associated with the changes of maternal breastfeeding intention. Thus, the purpose of this study is to understand mothers’ changes to their intended breastfeeding durations from the prenatal to neonatal periods. In addition, we examined how these changes were associated with actual breastfeeding durations and identified maternal characteristics associated with shortening intentions.

Methods

The Infant Feeding Practices Study II (IFPS II) was a longitudinal cohort study of mothers in the US of healthy full-term or near full-term singletons. The study was conducted from 2005 to 2007 by the US Food and Drug Administration (USFDA) in partnership with the Centers for Disease Control and Prevention. A detailed description of IFPS II methodology was published in 2008.¹⁶ Institutional Review Board approval was obtained from the USFDA.

Women were recruited during their third trimester of pregnancy after being identified from a nationally distributed consumer opinion panel of more than 500,000 US households. Eligibility criteria included that mothers be aged 18 years or older, and that both mother and infant were without any medical conditions that would affect feeding. The infant had to be born after 35 weeks of gestation, weigh 5 or more pounds, and not have stayed in intensive care for 3 or more days. Women were mailed one prenatal and 10 postpartum surveys almost monthly after birth. More than 3,000 (n = 3,033) women completed both the prenatal (surveyed at approximately 7 months of gestational age) and neonatal (surveyed at approximately 1 month after birth) questionnaires of IFPS II, 2,524 (83%) of whom intended to breastfeed. We excluded the following: (a) 362 women who had stopped breastfeeding before the neonatal survey because they were not asked about their breastfeeding intention on the neonatal survey; (b) 120 women who did not answer the question on intended breastfeeding duration on the prenatal survey; and (c) 68 women who did not answer the question on intended duration of breastfeeding on the neonatal survey. Therefore, 550 women were excluded because of missing data on breastfeeding intention. We also excluded 194 who were missing demographic information, leaving 1,780 mothers who intended to breastfeed and who were still breastfeeding at one month postpartum in the final analytical sample.

Women were asked on the prenatal and neonatal surveys: “How old do you think your baby will be when you completely stop breastfeeding?” (in months). Change in breastfeeding intention was calculated as the difference in intention reported on neonatal and prenatal surveys (i.e., neonatal intention minus prenatal intention). We classified change of intended breastfeeding duration as shorter neonatal intention, no change, and longer neonatal intention.

Actual breastfeeding duration was estimated by the question, “How old was this child when you completely stopped breastfeeding and pumping milk for him or her?” (in weeks). To calculate actual breastfeeding duration for children who were still breastfeeding at their last completed IFPS II questionnaire (n = 753), we linked IFPS II to its Year 6 Follow-up Study (Y6FU) in which mothers were asked to recall their breastfeeding duration. Detailed methods of Y6FU have been described elsewhere.¹⁷ Because of interactions between prenatal intention and its changes on actual breastfeeding duration, the examination of the association of actual breastfeeding duration with whether breastfeeding intention was shortened, lengthened, or constant was stratified by prenatal intended duration of more than 6 months (> 26 weeks) versus 6 months or less (≤ 26 weeks).

Covariates included maternal age (18–24, 25–29, 30–34, ≥35 years), race/ethnicity (non-Hispanic white, non-Hispanic black, Hispanic, Asian/Pacific Islander/Other, unknown), poverty-income ratio (<185%, 185%–349%, ≥350%), maternal education (≤ high school, 1–3 years of college, ≥ college graduate), prepregnancy body mass index (BMI) (<18.5, 18.5–24.9, 25.0–29.9, ≥30 kg/m²), primiparous (yes/no), prenatal smoking status (yes/no), participation in the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) (yes/no), cesarean delivery (yes/no), and marital status (yes/no).

Analyses were conducted in SAS 9.3 (SAS Institute, Cary, NC). Because breastfeeding duration was not normally distributed, the nonparametric Wilcoxon signed rank test was used to compare the mean ranks of actual breastfeeding duration (the dependent variable) by changes in intended breastfeeding duration (the predictor variable). Bivariate associations were assessed by using Pearson’s chi-squared tests. Multivariable logistic regression was used to assess the odds of shortened intended breastfeeding duration versus no change or longer intended duration, by maternal sociodemographic characteristics.

Results

The median intended duration of breastfeeding on the prenatal survey was 11.0 months (range 1–48 months) and 10.0 months (range <1– 48 months) on the neonatal survey. Comparing prenatal with neonatal surveys, 43.7% of mothers did not change their intended duration of breastfeeding, 35.0% shortened their intention, and 21.3% lengthened their intention (Table 1). Among mothers whose neonatal intention was shorter, 44.5% intended to breastfeed for 1–2 months shorter, 27.1% for 3–4 months shorter, and 28.4% for longer than 5 months shorter. Among mothers whose neonatal intended duration of breastfeeding was longer, 49.9% intended to breastfeed for 1–2 months longer, 24.5% for 3–4 months longer, and 25.6% for more than 5 months longer.

Table 1.

Change to Intended Duration of Breastfeeding Between the Prenatal and Neonatal Reports, IFPS II, 2005–2007 (N = 1,780)

	n (%)
No change	778 (43.7)
Neonatal intention shorter than prenatal, by:	623 (35.0)
1 to 2 months	277 (44.5)
3 to 4 months	169 (27.1)
>5 months	177 (28.4)
Neonatal intention longer than prenatal, by:	379 (21.3)
1 to 2 months	189 (49.9)
3 to 4 months	93 (24.5)
>5 months	97 (25.6)

Note. IFPS II refers to the Infant Feeding Practices Study II.

Among women who had a prenatal breastfeeding intention of 6 or less months (approximately ≤26 weeks), mean actual breastfeeding duration was 3.4 months among those with a shorter intended breastfeeding duration, 7.1 months among those with no change in intention, and 8.0 months among those with a longer intention (P <.001) (Table 2). Among women who had a prenatal breastfeeding intention of more than 6 months (approximately >26 weeks), mean actual breastfeeding duration was 10.3 months among those with a shortened breastfeeding intention, 13.5 months for no change, and 13.6 months for longer intention (P <.001).

Table 2.

Mean Actual Breastfeeding Duration by Prenatal and Neonatal Intention (N = 1,780), IFPS II, 2005–2007

Prenatal intended breastfeeding duration	Change in intended breastfeeding duration at neonatal report	n	Mean prenatal intention (months)	Mean neonatal intention (months)	Mean actual breastfeeding duration (months)	P-valuea
≤6 months	Shorter	151	5.0	2.7	3.4	<.001
(≤26 weeks)	No change	184	5.3	5.3	7.1
	Longer	149	4.2	8.0	8.0
>6 months	Shorter	472	12.2	8.3	10.3	<.001
(>26 weeks)	No change	594	12.9	12.9	13.5
	Longer	230	11.6	14.7	13.6

^aP-value was obtained by Wilcoxon signed rank test comparing the mean ranks of actual breastfeeding duration by change in intended breastfeeding duration at neonatal report.

In bivariate analysis, the percentage of mothers with a shorter breastfeeding intention varied significantly by prepregnancy BMI (P = 0.01) and maternal age, race/ethnicity, parity, prenatal smoking, WIC participation, and marital status (P < 0.01) (data not shown). In the multivariable analysis adjusted for sociodemographic factors (Table 3), odds of shortened intended breastfeeding duration from the prenatal to neonatal reports was increased among Hispanic women, women with a prepregnancy BMI of 30 kg/m² or higher, women who were primiparous, and women who smoked prenatally. Maternal age of 35 years or older was associated with decreased odds of shortened intended breastfeeding duration.

Table 3.

Odds of Shorter Intended Duration of Breastfeeding, by Demographic Characteristics, IFPS II, 2005–2007 (N = 1,780)

	Total	Shorter Intentionb
	n (%a)	n (%c)	aORd (95% CI)
Total	1,780 (100)	623 (35.0)
Maternal age (year)
18–24	309 (17.4)	134 (43.4)	Reference
25–29	633 (35.6)	217 (34.3)	0.77 (0.57–1.05)
30–34	539 (30.3)	186 (34.5)	0.83 (0.59–1.17)
≥35	299 (16.8)	86 (28.8)	0.64 (0.43–0.94)
Race/ethnicity
Non-Hispanic white	1,502 (84.4)	501 (33.4)	Reference
Non-Hispanic black	72 (4.0)	30 (41.7)	1.29 (0.78–2.13)
Hispanic	112 (6.3)	52 (46.4)	1.60 (1.08–2.39)
Asian/Pacific Islander/Other	94 (5.3)	40 (42.6)	1.48 (0.96–2.28)
Poverty-income ratio
<185%	661 (37.1)	241 (36.5)	Reference
185%–349%	671 (37.7)	223 (33.2)	1.02 (0.79–1.33)
≥350%	448 (25.2)	159 (35.5)	1.11 (0.80–1.54)
Maternal education
≤ High school	269 (15.1)	96 (35.7)	0.89 (0.63–1.25)
1–3 year(s) of college	704 (39.6)	265 (37.6)	1.09 (0.85–1.39)
≥ College graduate	807 (45.3)	262 (32.5)	Reference
Prepregnancy BMI (kg/m2)
<18.5	70 (3.9)	19 (27.1)	0.69 (0.40–1.20)
18.5–24.9	848 (47.6)	290 (34.2)	Reference
25.0–29.9	468 (26.3)	151 (32.3)	0.96 (0.75–1.22)
≥30	394 (22.1)	163 (41.4)	1.43 (1.10–1.84)
Multiparous
Yes	1,295 (72.8)	418 (32.3)	Reference
No	485 (27.2)	205 (42.3)	1.39 (1.09–1.78)
Prenatal smoker
Yes	99 (5.6)	53 (53.5)	1.69 (1.10–2.60)
No	1,681 (94.4)	592 (35.2)	Reference
WIC participation
Yes	576 (32.4)	228 (39.6)	1.10 (0.84–1.45)
No	1,204 (67.6)	395 (32.8)	Reference
Cesarean delivery
Yes	476 (26.7)	170 (35.7)	1.00 (0.79–1.26)
No	1,304 (73.3)	453 (34.7)	Reference
Married
Yes	1,491 (83.8)	495 (33.2)	Reference
No	289 (16.2)	128 (44.3)	1.23 (0.92–1.65)

IFPS II refers to the Infant Feeding Practices Study II. BMI is body mass index. WIC refers to the Special Supplemental Nutrition Program for Women, Infants, and Children.

^aColumn percentage

^bA neonatal breastfeeding intention that was shorter than prenatal breastfeeding intention.

^cRow percentage

^dAdjusting for listed sociodemographic characteristics

Discussion

Many women (43.7%) adhere to their prenatal breastfeeding intentions, and slightly more than one-fifth (21.3%) had longer neonatal than prenatal breastfeeding intentions. However, approximately one in three mothers in our study shortened their intended breastfeeding duration after birth (35.0%). Even though previous reports already show that many mothers are not meeting their prenatal breastfeeding intentions compared to their actual breastfeeding duration,^{9, 11, 12} and women who meet their breastfeeding intentions are more often satisfied with how long they breastfeed,¹⁸ there are no reports available that have examined the changes of maternal breastfeeding intentions from late pregnancy to shortly after birth. Our study demonstrates that change to intended duration of breastfeeding is significantly associated with actual breastfeeding duration, as suggested by other studies of the relationship between prenatal breastfeeding intentions and actual breastfeeding durations.^{8, 9} Within each prenatal intention (≤6 and >6 months), mothers who shortened their intended duration of breastfeeding after birth also had the shortest mean breastfeeding duration compared to those who did not change or increased their intended breastfeeding duration. These findings highlight the importance of breastfeeding support during the neonatal period.

Further, Hispanic women, women with a prepregnancy BMI of 30 kg/m² or higher, women who were primiparous, and women who smoked prenatally were at increased risk of shortening their intended duration of breastfeeding. Women with these demographic characteristics often have lower breastfeeding rates¹⁹ and, therefore, may need extra support, such as access to professional and peer breastfeeding support, to meet their prenatal breastfeeding intentions.

Our findings highlight the importance of supporting breastfeeding mothers and infants during the late prenatal and early postpartum periods. Health care providers, including both obstetric²⁰ and pediatric²¹ providers, play a key role in discussing infant feeding plans with mothers during the prenatal period, as well as supporting breastfeeding after birth. Further, evidence-based maternity care practices, such as those outlined in the Ten Steps to Successful Breastfeeding,²² provide an optimal environment for the establishment of breastfeeding during the early postpartum period. These practices have been shown to positively affect the initiation,²³ actual duration,^{24, 25} and exclusivity^{23, 26} of breastfeeding.

Although evidence-based practices supportive of breastfeeding are important, community breastfeeding support, especially during the early neonatal period, is also critical. The American Academy of Pediatrics²⁷ and American College of Obstetrics and Gynecology²⁸ have both published guidelines for office-based clinical care providers, outlining how providers can best support breastfeeding mothers being seen in their offices. Additional efforts, such as strengthening programs providing mother-to-mother support, improving peer counseling, and others, as outlined in the Surgeon General’s Call to Action to Support Breastfeeding,²⁹ may also help mothers successfully attain their prenatal breastfeeding goals after birth.

This study has several limitations and strengths. The IFPS II study, which was derived from a large sample of pregnant women drawn from a consumer opinion mail panel, was the largest longitudinal study of infant feeding in the United States. Despite this scope, when compared with other nationally representative samples, women included in IFPS II tended to have a higher socioeconomic status and were more likely to breastfeed and to breastfeed for a longer duration.¹⁶ Because women who stopped breastfeeding before the neonatal survey were not asked the neonatal intention question, the loss of this information during follow-up may result in the underestimation of negative changes to breastfeeding intention. All survey information was based on maternal report; as such, recall bias may have been introduced if mothers incorrectly reported information. Attempts to limit recall bias were made, however, by surveying women shortly after recruitment (prenatal survey) and within the first month after birth (neonatal survey), except for women who were still breastfeeding at the end of the initial study period who reported breastfeeding duration during the 6-year follow-up study.

Conclusions

Even though the majority of mothers in this study intended to breastfeed prenatally (83%), approximately one-third shortened their intended duration of breastfeeding between late pregnancy and a few weeks after giving birth. This decrease in intended breastfeeding duration was, in turn, associated with a shortened actual duration of breastfeeding. Further studies to identify the environmental factors for the changes on intention before and after birth are needed to support mothers in meeting their own prenatal breastfeeding intention. Efforts to support breastfeeding mothers both prenatally and during the neonatal period may prevent mothers from shortening their intended and actual durations of breastfeeding.