Abstract
Abstract
Objective:
To examine women's perceptions about support from pediatric primary care providers (PCPs), family, and friends for breastfeeding beyond 12 months, which is an increasing common practice.
Study Design:
Women who breastfed at least one child beyond 12 months completed an online questionnaire distributed via La Leche League USA (2013). Questionnaire content focused on sources of support for breastfeeding beyond 12 months, support ratings, and participant characteristics. Bivariate statistics and multivariable log-binomial regression compared ratings of support across sources, by PCP sex, and with breastfeeding duration.
Results:
Of 48,379 eligible U.S. women, about half discussed their decision to breastfeed beyond infancy with their child's PCP. In contrast, almost all (91.4%) did so with their spouse, partner, or significant other. Women were consistently more comfortable discussing their decision to breastfeed for more than a year with their family and closest friend than they were with their child's PCP (all p < 0.001). Three-fourths of PCPs were rated as supportive, but 11.1% were somewhat or very unsupportive. Female pediatricians received similar ratings as males (adjusted risk ratio = 1.01, 95% confidence interval: 1.00, 1.03). Thirty-eight percent of women who reported their PCP was unsupportive changed PCPs.
Conclusion:
Family and PCP support is likely to be important for the growing proportion of U.S. mother–child dyads who are breastfeeding beyond 12 months. Many, but not all, women rated their child's PCP as supportive, and lack of support was a reason women reported for changing PCPs. Evidence-based interventions in primary care to support breastfeeding beyond infancy are needed.
Introduction
C
Support from pediatric care providers, family members, and peers has been shown to be helpful in promoting breastfeeding initiation and extending breastfeeding duration.4–7 Given the evidence for benefit, the U.S. Preventive Services Task Force recently articulated an updated recommendation in support of primary care interventions to support breastfeeding. 8 However, much of the evidence has come from studies that focused on women at high risk of breastfeeding, discontinuation, and on achieving short-term breastfeeding duration goals of less than 12 months.9,10
Effective primary care or social support-based interventions to support breastfeeding beyond infancy would differ from interventions aimed at initiation or short-term duration because the barriers to continuing breastfeeding change across infancy. For instance, infants' sleep patterns and diet composition change, and the pressures of employment outside the home can increasingly interfere with sustaining a milk supply.
There remains a need for large observational studies to identify which sources of support are important for breastfeeding beyond infancy, and such a study would inform interventions tailored to mother–child dyads who wish to continue breastfeeding. The limited available evidence suggests that as breastfeeding duration approaches 12 months, support from family members declines. 11 Recent studies examining whether pediatrician support also declines are scant. However, numerous studies have documented that social stigma is a commonly articulated concern by women who breastfeed beyond infancy. 12
The present study examined the support for breastfeeding beyond infancy that mothers received from their child's primary care provider (PCP) compared with support received from their family and friends, in a large national sample of U.S. women who breastfed at least one child beyond 12 months of age. The objective was to identify opportunities for enhanced support for the increasingly common practice of breastfeeding beyond infancy, specifically interventions provided in pediatric primary care settings.
Materials and Methods
Participant recruitment and data collection
Study participants were recruited by distributing an invitation via e-mail lists, online support groups and chat rooms, list-serves, and social media platforms (e.g., Facebook) maintained by La Leche League USA. La Leche League USA is a large national organization focused on providing peer support for breastfeeding women. The invitation included a brief study description and a weblink to an online questionnaire, “Breastfeeding Beyond One Year of Age” (http://breastfeedingtoddlers.org). U.S. women aged 18 years and older who had breastfed at least one child beyond 12 months of age were eligible.
Participants indicated their consent to participate using a checkbox on the first screen of the survey. The study questionnaire consisted of demographic questions, a breastfeeding history for each child the woman breastfed, and a section focused on the woman's breastfeeding support experiences during the time she was breastfeeding her oldest child who was breastfed for more than 12 months. Specifically, women were asked whether they discussed their decision to breastfeed for more than 12 months with their spouse, partner, or significant other; their mother; their closest friend; and their child's pediatric PCP. For each listed individual with whom she discussed her decision, she was asked to rate how comfortable she felt discussing her decision and how supportive each individual was. Women who indicated that they felt unsupported by their child's PCP were asked if they changed PCPs as a result. Data contained no identifiers, and no incentive was provided. This study was reviewed and approved by the Institutional Review Board of the North Shore-LIJ Health System.
Statistical analysis
Univariate statistics were used to describe the study sample and calculate the proportions of respondents who discussed their decision to breastfeed for more than 12 months with their child's PCP and family and closest friend, as well as their degree of comfort and support received from each of these people. McNemar's test was used to compare the likelihood of discussing the decision to breastfeed for more than 12 months with each individual compared with the child's PCP. The degree of comfort in discussing and support received from each individual was compared with comfort and support from the child's PCP using Wilcoxon signed-rank tests. These comfort and support variables were also compared for female PCPs versus male PCPs, stratified by provider type (physician, nonphysician) because of differences in the proportions of females versus males practicing within each provider type. These sex differences were analyzed using log-binomial regression to produce risk ratios adjusted for confounders chosen a priori based on the literature (maternal education, race, ethnicity, region of residence, parity, and age). As an indicator of whether support from the listed individuals resulted in longer breastfeeding duration, women who breastfed their oldest child the longest (beyond 24 months of age) were compared with women who breastfed for a shorter duration (13–24 months of age) on their median ratings of support. A continuous composite support rating was calculated by assigning a value ranging −3 (very unsupportive) to 3 (very supportive) for each individual with whom the participant discussed her decision to breastfeed for more than 12 months, then summing the values across all individuals with whom she discussed her decision. Median composite support ratings were compared between the longest duration group and the shorter duration group using Wilcoxon rank-sum tests. Among women who felt that their child's PCP was somewhat or very unsupportive, those who changed to a different PCP were compared with those who did not change to a new PCP in terms of demographics and breastfeeding history using chi-square tests.
Results
Study participants
A total of 86,667 surveys were received based on unique Internet Protocol addresses. Women (n = 12,041) who were breastfeeding group leaders or clinically credentialed professionals were excluded because their personal breastfeeding experiences may not be generalizable. Women whose breastfeeding experiences were in the distant past were excluded by omitting 3,202 women whose oldest child was older than the age of 25 years. Women younger than 18 years of age were excluded (n = 20). To focus this analysis on the experiences of U.S. women whose child saw a PCP practicing conventional medicine, we excluded 21,243 respondents from other countries and 1,782 who saw alternative PCPs such as naturopaths, leaving a final sample of 48,379 eligible women for analysis.
More than half of participants were 30–39 years of age (Table 1). For 72.8% of participants, their youngest child was younger than 3 years of age at the time of the study. One-third were still breastfeeding their oldest or only child, and 9.8% had breastfed three or more children for more than 12 months. Of those who had stopped breastfeeding their oldest child by the time of the study, breastfeeding for much longer than 12 months was not uncommon: 19.1% of the sample breastfed beyond 24 months of age. Most women were non-Hispanic white, with over half having obtained at least a Bachelor's degree.
Two hundred sixty participants were missing the data for age, 347 for race and ethnicity, 71 for region, and 133 for education.
The vast majority (91.4%) of women discussed their decision to breastfeed for more than 12 months with their spouse, partner, or significant other, while far fewer discussed their decision with their child's PCP (49.2%), their own mother (72.5%), or their closest friend (69.7%) (all McNemar's p < 0.001). Overall, 1.5% of women did not discuss their decision with any of the listed individuals, and 28.7% discussed it with all of them. Of the women who did not discuss their decision with their child's PCP, the vast majority (82.8%) gave the reason that they did not need additional support from the PCP. The next most common reason was that they were afraid or knew that the PCP would not support their decision (7.8%).
Perceived comfort in discussing with and support received by pediatric PCPs
Of women who discussed their decision to breastfeed, at least two-thirds felt very comfortable discussing with each of the listed individuals (their child's PCP, the participant's spouse, partner, or significant other; the participant's mother, the participant's closest friend) (Table 2). However, ratings were highest for the spouse, partner, or significant other and lowest for the child's PCP (p < .001). Approximately 36% felt less than “very comfortable” discussing with the child's PCP. Ratings of supportiveness were lower than for comfort for all listed individuals. Participants' spouse, partner, or significant other had the highest ratings of supportiveness, while the child's PCP and the participant's mother had the lowest rankings (59.4% of participants rated their child's PCP as “very supportive.”).
Twenty-nine participants were missing the ratings of comfort for the child's primary care provider, 21 for participant's spouse, partner, or significant other; 14 for participant's mother; 8 for participant's closest friend. Sixty-two participants were missing the ratings of support for the child's primary care provider, 11 for participant's spouse, partner, or significant other; 22 for participant's mother; 47 for participant's closest friend. p-values are for Wilcoxon signed-rank tests comparing each family member or friend to the child's primary care provider, among those who discussed their decision with both individuals in the comparison.
Whether the child's PCP was female versus male made only a modest difference within most PCP types (Table 3). For instance, female general practitioners or family doctors were 5% more likely to be rated very supportive or somewhat supportive compared with male general practitioners or family doctors, after controlling for the participant's education, race, ethnicity, region, parity, and age (adjusted risk ratio (RR) = 1.05, 95% confidence interval [CI]: 1.02, 1.08). Smaller or no similar differences were observed for pediatricians and other provider types.
Twenty-nine were missing the data for how comfortable they felt, 62 for how supportive their provider was. Adjusted models included maternal education, race, ethnicity, region of residence, parity, age. Models for Physician Assistant are unadjusted due to sparse data.
CI, confidence interval; RR, risk ratio.
For women who discussed their decision to breastfeed for more than 12 months with only two of the four individuals listed as sources of support (e.g., discussed with child's PCP and spouse, partner or significant other, but not with her mother or closest friend), composite ratings of the support received from the two individuals were not associated with breastfeeding duration (longer than 24 months compared with 13–24 months) (Table 4).
Sample restricted to women who breastfed all their children for more than 12 months, had stopped breastfeeding their oldest child, and whose children were all >12 months of age at the time of the study.
Women who breastfed for only 13–24 months reported lower composite ratings of support (median = 8.0, interquartile range [IQR] = 4.0) compared with women who breastfed for more than 24 months (median = 9.0, IQR = 6.0), among women who discussed their decision with and rated all four listed individuals. Thus, higher composite ratings of support were associated with shorter breastfeeding duration in this group who discussed their decision with all listed individuals.
Of the 2,465 women who discussed breastfeeding with their child's PCP and reported that the PCP was somewhat or very unsupportive, approximately one-third (37.6%) changed providers as a result (Table 5). Women younger than 30 years of age and women with some college or a 2-year degree were represented in greater proportion among those who changed providers. The sex of the unsupportive provider was not associated with the risk of switching (RR for female versus male = 1.07, CI: 0.97, 1.19). Women were no more likely to leave a physician PCP than a nonphysician PCP (RR = 0.92, CI: 0.68, 1.25).
Thirteen participants were missing the data for age, 15 for race, 15 for ethnicity, 10 for region, and 7 for education.
Discussion
In this large national sample of women who breastfed their child beyond 12 months of age, only about half discussed their decision to continue breastfeeding beyond 12 months with their child's PCP. In contrast, almost all discussed their decision with their spouse, partner, or significant other, and more than two-thirds discussed it with their mother or closest friend.
Overall, women were most likely to rate their spouse, partner, or significant other as supportive, with smaller proportions indicating supportiveness from their closest friend, the child's PCP, and their own mother. Of women who discussed their decision with their child's PCP, most felt that the provider was supportive, but ∼11% reported that the provider was somewhat or very unsupportive. Female PCPs did not receive notably higher ratings of comfort and support compared with males. Of those who reported that their PCP was unsupportive, about 38% changed PCPs. Women who breastfed the longest (>24 months) reported slightly lower overall supportiveness when they discussed their decision with most or all of the sources of support listed, compared with women who breastfed for 13–24 months.
Women did not discuss their decision to breastfeed beyond 12 months with their child's PCP for varied reasons, most commonly because they felt that they did not need the provider's support, at least in late infancy. Because these women were likely to be successful with breastfeeding, they may have had little need to bring up the topic during office visits. Also, while PCPs often inquire about infant feeding at early well-baby visits, it is often of less concern later, so PCPs may not have initiated a discussion.
Nevertheless, about 8% of women who did not discuss their decision with their child's PCP reported that it was because that they were afraid that the provider would not be supportive. Numerous studies have reported that many women who breastfeed beyond infancy resort to “closet nursing” to conceal their behavior, including sometimes from their healthcare provider. 13 The proportion of women in the present study (49.2%) who discussed their decision with their child's PCP is much higher than reported in a study from the early 1990s, which reported only 11% of a sample of Minnesota women who breastfed beyond 12 months received support from a healthcare provider. 14 Thus, while many PCPs and women are not discussing breastfeeding beyond 12 months, it may be that the proportion that does is increasing.
While it is reassuring that most women who did discuss their decision to breastfeed beyond infancy felt comfortable discussing it with their child's PCP and felt supported in their decision, women were more likely to feel comfortable discussing with the other family and friend sources of support than they were with the PCP. In fact, only the participants' mothers were rated lower in supportiveness than PCPs. Unfortunately, 22.9% of women felt neutral or unsupported by their child's PCP, and some of those who did not discuss their decision with their child's PCP avoided the discussion out of fear that they would not be supported. Thus, there remains room for improvement in the breastfeeding support that pediatric PCPs provide, especially when considering that more than one-third of dissatisfied participants left their pediatric PCP because of the lack of support.
This conclusion applies to both male and female PCPs, as provider sex did not appear to play a large role in women's ratings of the support they received. This contrasts the findings of a survey of 875 pediatricians, in which the 64% who had breastfeeding experience (either themselves or their spouses) were stronger proponents of breastfeeding and were much more likely to counsel new mothers about infant feeding and teach breastfeeding, although the study did not focus on the sex of the providers. 15 The present study lacked data on the breastfeeding experience of PCPs to compare directly to these findings.
The extent to which quality pediatric PCP support for breastfeeding beyond infancy results in longer duration of breastfeeding remains unknown. For instance, the Infant Feeding Practices Study II found that women who reported having pediatric PCP support at 3 months for breastfeeding were significantly more likely to be breastfeeding at 6 months, but the focus was on early infancy. 16
Pediatric PCP support may be just as important as breastfeeding continues beyond infancy because new barriers to breastfeeding arise as children mature (e.g., changes in diet and sleep patterns, behavior). PCP support may be increasingly important if other individuals like spouses and employers become less supportive as breastfeeding continues. Surprisingly, this study found that women who breastfed the longest (>24 months) rated the overall support they received from all sources no better, and in some cases, worse than women who breastfed for only 13–24 months, suggesting that better support did not encourage longer duration in this population. These findings could be interpreted to indicate that women breastfeeding into the late toddlerhood/preschool years experienced more negative reactions and stigmatizing attitudes from people in a position to be supportive, compared with women who stopped breastfeeding by age 2.
This is compatible with findings from the early 1990s that the older the breastfeeding child the greater the social stigma reported by the mother. 17 Many healthcare professionals continue to maintain negative views of breastfeeding beyond infancy: 61% of a convenience sample of New York providers in 2012 found it acceptable to breastfeed a 1 to 2-year-old child, while only 22% found it acceptable for 3 to 4-year-old children. 18
The present study may not be generalizable to all women and children who breastfeed beyond infancy because the sample was recruited via channels maintained by La Leche League USA. However, the general population of women and children who breastfeed for long durations is difficult to access through population-based sampling because they comprise a small proportion of total population. Support groups are an ideal way to enlist their participation in research. We did, however, exclude women who were themselves breastfeeding educators to increase generalizability.
Also, the sample reflected all regions of the United States and included a moderate proportion of women who were members of racial or ethnic minority groups (13.3% of the sample); these are advantages over prior geographically limited or less diverse samples. Some participants were recalling their breastfeeding experiences from several years in the past, which may have introduced measurement error into our data; however, most participants had young children at the time of the study (72.8% had at least one child younger than the age of 3 years), which limited the length of recall required. Finally, our sample was very large compared to most studies about breastfeeding, particularly for research focused on women and children who breastfed beyond infancy.
Most studies about breastfeeding beyond infancy have involved very small samples or have been limited to qualitative analyses. The large sample enabled multivariable statistical analyses to control for important sociodemographic confounders such as education, race, region, and parity, as these factors may underlie breastfeeding practices and perceptions about sources of support. The large sample also facilitated analyses focused on important subsamples such as families who changed PCPs because of dissatisfaction with breastfeeding support, a novel aspect of this research.
Conclusion
The proportion of U.S. infants continuing to breastfeed to 12 months of age and potentially beyond is rapidly increasing—a 50% increase over just 9 years, according to the CDC.2,3 As a result, pediatric healthcare providers can be expected to be in a position to support an increasing number of mothers and children who want to continue breastfeeding beyond infancy.
Evidence-based interventions in pediatric primary care specific to support for breastfeeding after infancy are lacking. However, it is likely that pediatrician support would be influential in helping mothers meet their long-term breastfeeding goals. Finally, improved PCP support for these families may have the additional benefit of retaining families in the provider's practice, as breastfeeding support appears to be an important factor for some families in remaining with the same provider.
Footnotes
Acknowledgments
We thank La Leche League USA for their help in disseminating our questionnaire. The authors have no consultantships, stock ownership, equity interests, arrangements regarding patents, and other vested interests related to this research to report. This work was presented at the 2015 Pediatric Academic Societies meeting.
Disclosure Statement
No competing financial interests exist.