The Impact of the Professional Qualifications of the Prenatal Care Provider on Breastfeeding Duration

Abstract

Background:

A prenatal commitment to breastfeed is a strong predictor for breastfeeding success. Prenatal care providers have the opportunity to educate and promote breastfeeding. However, differences in education and training between healthcare providers such as physicians and midwives may result in differing breastfeeding outcomes. This study explores whether breastfeeding initiation and duration differ by prenatal care provider.

Materials and Methods:

Longitudinal data from the Infant Feeding Practices Survey II were analyzed (N = 2,832 women). Prenatal care providers were categorized as obstetrician, family/other physician, and midwife/nurse-midwife. Breastfeeding initiation was dichotomized (yes; no). Breastfeeding duration and exclusive breastfeeding duration were reported in weeks. Logistic regression was used to investigate the relationship between prenatal care provider and breastfeeding initiation. Cox proportional hazard models provided crude and adjusted hazard ratios and 95% confidence limits to determine the relationship between type of prenatal care provider and breastfeeding duration.

Results:

After adjusting for confounders, women who received care from a midwife were 68% less likely to never breastfed than women whose prenatal care was provided by an obstetrician. Women whose prenatal care was provided by a midwife had 14% lower risk of discontinuing breastfeeding and 23% lower risk of discontinuing exclusive breastfeeding. No significant association was found between women whose prenatal care was provided by a family physician or other type of physician and breastfeeding initiation and duration.

Conclusion:

Findings highlight the importance of prenatal care providers on breastfeeding duration. Future studies should examine factors (i.e., training, patient–provider interaction) that contribute to differences in breastfeeding outcomes by type of prenatal care provider.

Introduction

Prenatal care is a vital component of maternal and child health. Research has shown that prenatal care can help prevent pregnancy complications and poor birth outcomes.¹ Literature has also demonstrated that prenatal classes are an important factor in breastfeeding duration. For example, a cross-sectional study reported the odds of breastfeeding were 2.7 times higher among women who attended prenatal classes than among women who did not attend.² Furthermore, a qualitative study found that bottle-feeding mothers would have felt encouraged to breastfeed if more breastfeeding information was provided in the prenatal class and from TV, magazines, and books.³

The prenatal care provider's knowledge, commitment, and communication skills are important aspects of prenatal care that can influence breastfeeding duration. However, a growing body of literature affords insight into physicians' deteriorating attitudes and commitment to breastfeeding support.^4–6 For instance, a study that surveyed practicing obstetrics and gynecologists (ob/gyn), pediatricians, and family physicians reported a significant deficit in the knowledge of breastfeeding benefits and clinical management.⁵ On the contrary, research has shown that midwives have greater knowledge of breastfeeding benefits and higher self-confidence when managing breastfeeding problems than physicians.⁷ The differences observed could be caused by the lack of breastfeeding training and education physicians receive⁸ compared with midwives whose core competencies include breastfeeding support and management.⁹

One aspect that prenatal care providers can directly influence is breastfeeding intention—a correlate of breastfeeding duration.¹⁰ Research has demonstrated that a prenatal commitment to breastfeeding is strongly associated with breastfeeding initiation and duration.¹¹ Therefore, prenatal care providers have a unique opportunity to influence breastfeeding intentions and subsequent breastfeeding behaviors.¹² This study is guided by the Theory of Planned Behavior¹³—which predicts an individual's behavior based on his or her intentions. This theory states that an intention to engage in a behavior (i.e., breastfeeding) is a function of (1) attitudes, (2) subjective or social norms, and (3) perceived behavioral control.¹³ Research has stated that a maternal prenatal intention to breastfeed is the strongest predictor of breastfeeding duration and intiation.¹¹

In light of recent trends, understanding the influence of prenatal care providers on breastfeeding duration is essential to improve the low rate of breastfeeding. In 2013, only one in five (22.3%) mothers in the United States exclusively breastfeed for 6 months¹⁴—the recommended duration according to the American Academy of Pediatrics.¹⁵ Owing to widespread concern about the low breastfeeding rates, the increase of women utilizing nontraditional providers warrants investigation into prenatal care providers and their impact on breastfeeding duration. Estimates from 2009 to 2010 state that only one out of seven (14.1%) routine prenatal care visits was with a non-ob/gyn provider, with the highest estimates being among women aged 15–19 years (20.5%).¹⁶

Although some research suggests that midwives have better knowledge and ability to provide breastfeeding support,⁷ it is unknown whether breastfeeding duration is influenced by the type of prenatal provider (e.g., physicians and midwives). Understanding breastfeeding duration by prenatal care provider will provide insight into the effectiveness of healthcare providers in breastfeeding promotion, and where additional training is warranted. Guided by the Theory of Planned Behavior, this study aims to examine whether type of prenatal care provider is associated with (1) breastfeeding initiation, (2) breastfeeding duration (breast milk in combination with other supplements), and (3) exclusive breastfeeding duration.

Materials and Methods

The Infant Feeding Practices Survey (IFPS) II is a large size longitudinal study conducted in the United States between May 2005 and June 2007 by the Food and Drug Administration and Centers for Disease Control and Prevention. The IFPS II collected information about infant feeding, including patterns of breastfeeding, formula feeding, solid food intake, other complementary food and liquid intake, and feeding practices in the first year of the infants' life.¹⁷

To be included in the IFPS II, study participants were at least 18 years old at the time of the prenatal survey and had good maternal and child health at birth. Good maternal and child health at birth was defined as, “neither the mother nor the infant could have a medical condition at birth that would affect feeding and that the infant had to have been born after at least 35 weeks' gestation, weigh at least 2.3 kilograms, be a singleton, and not have stayed in the intensive care for >3 days.”¹⁸ Additional information on IFPS methodology¹⁸ and questionnaires¹⁷ can be found elsewhere. This study also excluded women who received prenatal care from multiple providers (n = 321), did not receive prenatal care from a health professional (n = 27), or received care from a healthcare provider that was not specified (n = 84), leaving 2,832 women for analysis. This study was approved as exempt by the Virginia Commonwealth University Institutional Review Board.

The exposure variable, type of prenatal care provider (obstetrician; family doctor, general practitioner, internist, or other physician; a midwife or nurse-midwife), was based on the survey question, “Who provides your prenatal care?” that was asked during the prenatal survey. Breastfeeding initiation (yes; no) was based on the survey question, “Did you ever breastfeed this baby (or feed this baby your pumped milk)?” Breastfeeding duration and exclusive breastfeeding duration measured the number of weeks the infant was breastfed among women who initiated breastfeeding. Breastfeeding duration included any combination of infant feeding that included breast milk and was based on two survey questions. If mothers responded “yes” to the question, “Did you ever breastfeed this baby (or feed this baby your pumped milk)?” they were asked the following, “Have you completely stopped breastfeeding and pumping milk for your baby?” This question was asked every month postpartum until breastfeeding cessation. If women responded “yes,” they were asked, “How old was your baby when you completely stopped breastfeeding and pumping milk?” Participants report the number of days they breastfed if the infant was younger than 2 weeks old or in the number of weeks if the infant was at least 2 weeks old. If mothers were still breastfeeding at the time of the last questionnaire (12 months postpartum) (N = 917), the following survey question was asked at the 6-year follow-up to determine breastfeeding duration, “How old was your 6-year-old when the following happened? He or she stopped being fed breast milk, including pumped breast milk.”

The variable utilized in IFPS to distinguish between “breastfeeding duration” and “exclusive breastfeeding duration” was prespecified by IFPS II using questionnaire items regarding exclusive breastfeeding for the hospital stay and after discharge. Exclusive breastfeeding was defined as the infant receiving breast milk and no other food or drink—which is consistent with the definition from the American Academy of Pediatrics.¹⁵ Exclusive breastfeeding for the hospital stay was determined using two questions: “While you were in the hospital or birth center, was your baby fed water, formula, or sugar water at any time” and “When you left the hospital or birth center, how were you feeding your baby?” Women who did not feed their baby water, formula, or sugar water at any time in the hospital and who were only feeding their baby breast milk were categorized as “exclusive breastfeeding.” Exclusive breastfeeding after discharge was defined using the food frequency checklist that asked, “In the past 7 days, how often was your baby fed each food listed below? Include feedings by everyone who feeds the baby and include snacks and night-time feeding.” Mothers were provided a list of food that the infant could consume and filled in columns asking the frequency of feeding per day or per week.¹⁷ If women only reported feeding their infant breast milk, they were categorized as “exclusive.” This question was asked approximately every month postpartum.

Potential confounders identified in the literature and available in the data set were examined. These include marital status (married, not married), maternal race/ethnicity (non-Hispanic [NH] white, NH black, Hispanic, NH other), maternal age (continuous), maternal education (less than high school, high school graduate, 1–3 years of college, college graduate), household income (<$20,000, $20,000–49,999, ≥$50,000), prepregnancy body mass index (BMI) (underweight (<18.5 kg/m²), normal weight (18.5–24.9 kg/m²), overweight (25.0–29.9 kg/m²), obese (30.0+ kg/m²), prenatal health insurance or healthcare plan (yes; no), prenatal participation in the Special Supplemental Nutrition Program from the Women, Infants, and Children (WIC) program (yes, no), mode of delivery (vaginal, cesarean section), breastfeeding intention (breastfeed only, formula feed only, both breast and formula feed, do not know), month expected to completely stop breastfeeding (continuous), age when baby was first fed formula (never fed formula, ≤1 day old, 2–6 days old, 7–13 days old, 14–20 days old, >20 days old), and average daily number of cigarettes smoked during the prenatal period (continuous).

All baseline characteristics were summarized using percentages and means with standard errors (SEs). A chi-square test was used to compare groups of women based on their prenatal care provider. Logistic regression was used to investigate the relationship between prenatal care provider and breastfeeding initiation. Cox proportional hazard models provided crude and adjusted hazard ratios (HRs) and 95% confidence limits (CL) to determine the relationship between type of prenatal care provider and breastfeeding duration. All survival analyses were modeled separately for time to any breastfeeding cessation and time to exclusive breastfeeding cessation.

A Kaplan–Meier curve was constructed to illustrate differences between prenatal care provider and time to breastfeeding cessation. The proportional hazard assumption was verified using the Kolmogorov-type supremum test, which showed no significant deviation from the assumption (p = 0.087). A Cox proportional hazard models obtained crude and adjusted HRs and 95% CL. Based on the literature, breastfeeding intention and intended duration of breastfeeding were tested as effect modifiers and found to be significant (p-value = 0.0033; 0.0042, respectively); therefore, they were not considered as potential confounders. A stratified analysis by breastfeeding intention was not conducted due to small cell sizes. Only potential confounders that resulted in at least a 10% change in the crude estimate were included in the final parsimonious model.¹⁹ Because all potential confounders did not meet the 10% change rule for exclusive breastfeeding duration, a fully adjusted model was fit. SAS version 9.4 statistical software (SAS, Cary, NC) was used for all analyses.

Results

The majority of study participants were NH white (84.6%), had at least some college education (79.3%), and intended to breastfeed during the prenatal period (59.6%). The mean breastfeeding duration was 6.5 months (25.8 weeks; SE = 0.4), whereas the mean exclusive breastfeeding duration was about 3 months (12.1 weeks; SE = 0.3). The majority of participants' prenatal care was provided by an obstetrician (83.8%). Table 1 displays the study characteristics by type of prenatal care provider. There was a statistically significant association between age, education, income, prepregnancy BMI, health insurance, prenatal WIC, mode of delivery, breastfeeding duration, breastfeeding intention, and type of prenatal care provider.

Table 1.

Distribution of Maternal Characteristics by Prenatal Care Provider

Characteristic	Total (N = 2,832)	Obstetrician (N = 2,373)	Family or other MD (N = 214)	Midwife (N = 245)	Chi-square (p)
Mean age (SE)	28.8 (0.1)	29.0 (0.1)	27.3 (0.3)	28.1 (0.3)	<0.0001
Marital status (%)					0.0623
Not married	20.8	20.1	26.9	22.7
Maternal race (%)					0.1861
White, NH	84.6	84.1	85.6	88.4
Black, NH	4.8	4.9	6.7	2.5
Hispanic	6.2	6.3	5.8	5.4
Other	4.4	4.7	1.9	3.7
Maternal education (%)					<0.0001
Less than high school	3.2	2.7	5.3	6.1
High school	17.5	16.7	27.0	17.5
1–3 Years of college	40.0	39.7	42.3	40.8
College graduate	39.3	40.9	25.4	35.5
Income (%)					<0.0001
<$20,000	13.7	12.7	21.0	16.7
$20,000–$49,999	42.9	41.7	49.1	49.0
≥$50,000	43.5	45.6	29.9	34.3
Prepregnancy BMI (%)					0.0484
Underweight	4.7	4.7	2.8	5.4
Normal weight	46.0	46.0	40.6	50.6
Overweight	25.7	25.0	33.5	25.3
Obese	23.7	24.2	23.1	18.7
Health insurance (%)					<0.0001
No	4.6	3.8	8.4	9.4
Prenatal WIC (%)					0.0001
Yes	29.8	28.2	40.7	35.1
Mode of delivery (%)					<0.0001
Vaginally—not induced	38.1	34.5	47.7	63.7
Vaginally—induced	33.7	34.9	31.8	23.7
Planned cesarean section	16.3	17.9	10.8	5.3
Unplanned or emergency cesarean section	12.0	12.6	9.8	7.4
Mean breastfeeding duration in weeks (SE)	25.8 (0.4)	25.4 (0.4)	23.9 (1.4)	30.9 (1.4)	0.0007
Mean exclusive breastfeeding duration in weeks (SE)	12.1 (0.3)	11.8 (0.3)	11.4 (1.0)	14.4 (0.8)	0.0024
Breastfeeding intention					<0.0001
Breastfeed only	59.6	58.9	49.1	76.1
Formula feed only	13.3	13.7	17.3	6.2
Both breastfeed and formula feed	23.5	23.7	30.4	15.6
Unsure	3.6	3.8	3.3	2.1
Mean daily cigarettes smoked (SE)	0.9 (0.1)	0.9 (0.1)	1.1 (0.2)	0.9 (0.2)	0.7338

Not all percentages sum to 100% due to rounding.

Bold indicates statistical significance.

BMI, body mass index; SE, standard error; WIC, women, infants, and children.

Table 2 shows results from the logistic regression analysis investigating breastfeeding initiation and type of prenatal care provider. Compared with women whose prenatal care was provided by an obstetrician, women who received care from a midwife were 67% less likely to never breastfeed (crude odds ratio = 0.33; 95% confidence interval [CI] = 0.19–0.58). After adjusting for maternal education, women who received care from a midwife were 68% less likely to never breastfeed (adjusted odds ratio = 0.32; 95% CI = 0.18–0.57). No statistically significant association was observed among family/other physicians.

Table 2.

Association Between Prenatal Care Provider and Breastfeeding Initiation

	OR (95% CI)
	Unadjusted	Adjusted ^a
Never breastfed
Family/other physician^b	1.19 (0.82–1.72)	0.90 (0.59–1.37)
Midwife/nurse midwife^b	0.33 (0.19–0.58)^***	0.32 (0.18–0.57)^**

Adjusted for maternal education.

Compared with mothers who received prenatal care from an obstetrician.

p ≤ 0.01; ^***p ≤ 0.001.

CI, confidence interval; OR, odds ratio.

The survival curves (not provided) showed a divergence at 5 weeks between the prenatal providers and breastfeeding duration. After 5 weeks, women whose prenatal care was provided by a physician (obstetrician or family doctor/other physician) had a lower probability of breastfeeding relative to women whose prenatal care was provided by a midwife. Moreover, the mean breastfeeding duration among women with an obstetrician or a family doctor was 25 and 24 weeks, respectively, compared with a mean duration of 31 weeks for women whose prenatal care was provided by a midwife.

The unadjusted analysis showed that women whose prenatal care was provided by a midwife had 21% lower risk (HR = 0.79, 95% CL = 0.69–0.90) of discontinuing breastfeeding compared with women whose prenatal care was provided by an obstetrician. After adjusting for maternal age and age when baby was first fed formula, the estimate remained significant but was slightly attenuated. Compared with women whose prenatal care was provided by an obstetrician, women whose prenatal care was provided by a midwife had 14% lower risk (HR = 0.86; 95% CL = 0.75–0.99) of discontinuing breastfeeding. No significant difference in breastfeeding discontinuation was found among women whose prenatal care was provided by a family physician or other type of physician (Table 3).

Table 3.

Proportional Hazards of Breastfeeding Duration by Prenatal Care Provider

	HR (95% CL)
	Unadjusted	Adjusted
Any breastfeeding^a
Family/other physician^b	1.07 (0.91–1.24)	0.97 (0.82–1.14)
Midwife/nurse midwife^b	0.79 (0.69–0.90)^**	0.86 (0.75–0.99)^*
Exclusive breastfeeding^c
Family/other physician^b	1.04 (0.83–1.29)	1.01 (0.80–1.28)
Midwife/nurse midwife^b	0.72 (0.60–0.85)^**	0.77 (0.64–0.93)^**

Adjusted for maternal age and age when baby was first fed formula.

Compared with mothers who received prenatal care from an obstetrician.

Adjusted for age, marital status, education, race/ethnicity, income, prepregnancy BMI, insurance status, prenatal WIC participation, mode of delivery, and smoking during pregnancy.

p ≤ 0.05; ^**p ≤ 0.01.

HR, hazard ratio.

Analysis examining exclusive breastfeeding duration also demonstrated that women whose prenatal care was provided by a physician had a higher probability of breastfeeding cessation. Specifically, women whose prenatal care was provided by a midwife had 28% lower risk (HR = 0.72, 95% CL = 0.60–0.85) of discontinuing breastfeeding than women whose prenatal care was provided by an obstetrician. After fully adjusting for age, marital status, education, race/ethnicity, income, prepregnancy BMI, insurance status, prenatal WIC participation, mode of delivery, and smoking during pregnancy, women whose prenatal care was provided by a midwife had 23% lower risk (HR = 0.77; 95% CL = 0.64–93) of discontinuing breastfeeding than women whose prenatal care was provided by an obstetrician. No significant association was found between women whose prenatal care was provided by a family physician or other type of physician and breastfeeding duration (Table 3).

Discussion

This study found a relationship between type of prenatal care provider and breastfeeding initiation and duration. Specifically, women whose prenatal care was provided by a midwife or nurse midwife had a lower risk of never breastfeeding and breastfeeding cessation than women whose prenatal care was provided by an obstetrician. No statistically significant relationship was observed between women who received prenatal care from a family physician and breastfeeding duration.

Results from this study may be explained by healthcare providers' breastfeeding support—which is associated with a decreased likelihood of breastfeeding cessation.²⁰ For example, a retrospective study stated that women whose prenatal care was provided by a midwife reported better communication than those provided by other types of physicians.²¹ Also, 9 out of 10 midwives reported encouraging mothers to breastfeed more if they were concerned with insufficient milk supply⁷—one of the major reasons women prematurely cease breastfeeding.²² Midwives also report that breastfeeding support is an important component of their role as a healthcare provider.⁷ Clinician support provided during the prenatal period could reduce stress that is often present during the transition to parenthood. Research has demonstrated that stress interferes with lactation²³ and the main hormone responsible for milk ejection (i.e., milk let-down), oxytocin²⁴; however, the effect of stress has been shown to be mitigated by social support.²⁵

Findings from this study may also be explained by current hospital policies. Owing to the current U.S. healthcare system, hospital policies may push physicians to see an increased number of patients, limiting the availability of time with each patient. Midwives may not be constrained by the same policies since expectations between healthcare professionals vary. These time differences could result in midwives providing adequate breastfeeding support and education, whereas physicians do not have the time to provide this care.

Although findings from this study cannot be corroborated by research conducted in the United States, results from a cross-sectional analysis conducted in Canadian provinces and territories were found similar. Specifically, women whose prenatal care provider was a midwife or family doctor were significantly more likely to breastfeed a longer duration.²⁶ Although this study showed lower risk of discontinuation among midwife than among obstetricians and gynecologists physicians, similar effects were not observed among family physicians. This may be due to the differing healthcare systems between the United States and Canada.²⁷ Specifically, family physicians are integral to the universal healthcare program of Canada. One main component of universal healthcare is primary prevention—which often occurs at the family physician visit. Within this healthcare system, family physicians may also provide other specialty services usually offered by specialized providers to reduce costs associated with specialist.^28,29 Because the United States does not place a large focus on family physicians,³⁰ they may not be incentivized to provide specialty services and focus on primary prevention.

This study has a number of strengths. To the authors' knowledge, this is the first study to investigate the association between prenatal care providers and breastfeeding duration. Utilizing IFPS II, a prospective longitudinal study, allowed temporality to be established. The IFPS II also used a standardized data collection protocol (consumer opinion panel) that minimized the potential for information bias. Lastly, all survey questions used in each questionnaire were extensively tested³¹—which increases the likelihood that the questions are valid and reliable indicators for the concepts of interest.³²

Despite its strengths, this study is not without limitations. Because IFPS II used a consumer opinion panel to identify participants, the study population disproportionately represents women who are NH white, are of higher socioeconomic status, can read English, and have stable mailing addresses. Therefore, results from the IFPS II are not generalizable to the U.S. population of pregnant women and new mothers.¹⁸ Furthermore, breastfeeding duration is self-reported, which may be prone to social desirability bias and nondifferential misclassification, as women wanting to be seen as being caring of their babies may have overreported their breastfeeding duration; however, research has shown that maternal self-report of breastfeeding duration is a reliable measurement.³³ Lastly, potential confounding factors that could affect estimates including high-risk pregnancies, desired level of obstetric intervention (e.g., women who desire a low intervention birth might be more likely to seek midwifery and breastfeed a longer duration), type of facility where mother received prenatal care, number of prenatal care visits, self-efficacy, and alcohol/substance use were not available in the data set and could not be assessed.

Conclusion

Overall, this study highlights the stark differences in breastfeeding outcomes by type of prenatal care provider. The moderating role of breastfeeding intention should be explored in the relationship between prenatal care provider and breastfeeding duration in a larger sample. Furthermore, research is needed to explore factors, such as breastfeeding education/training and hospital and medical school policies on breastfeeding practices that may potentially explain these differences.

Footnotes

Disclosure Statement

No competing financial interests exist.

References

National Institutes of Health. What is prenatal care and why is it important? Updated. 2013. Available at https://www.nichd.nih.gov/health/topics/pregnancy/conditioninfo/pages/prenatal-care.aspx (accessed October 27, 2016).

Giugliani

, Caiaffa

, Vogelhut

, et al. Effect of breastfeeding support from different sources on mothers' decisions to breastfeed. J Hum Lact, 1994; 10:157–161.

Arora

, McJunkin

, Wehrer

, et al. Major factors influencing breastfeeding rates: Mother's perception of father's attitude and milk supply. Pediatrics, 2000; 106:E67.

Feldman-Winter

, Schanler

, O'Connor

, et al. Pediatricians and the promotion and support of breastfeeding. Arch Pediatr Adolesc Med, 2008; 162:1142–1149.

Freed

, Clark

, Sorenson

, et al. National assessment of physicians' breast-feeding knowledge, attitudes, training, and experience. JAMA, 1995; 273:472–476.

Philipp

, Merewood

, O'Brien

. Physicians and breastfeeding promotion in the United States: A call for action. Pediatrics, 2001; 107:584–587.

Hellings

, Howe

. Assessment of breastfeeding knowledge of nurse practitioners and nurse‐midwives. J Midwifery Womens Health, 2000; 45:264–270.

Dusdieker

, Dungy

, Losch

. Prenatal office practices regarding infant feeding choices. Clin Pediatr (Phila), 2006; 45:841–845.

American College of Nurse-Midwives. Core Competencies for Basic Midwifery Practice, 2012. Available at: www.midwife.org/ACNM/files/ACNMLibraryData/UPLOADFILENAME/000000000050/Core%20Comptencies%20Dec%202012.pdf (accessed December 7, 2017).

10.

Brand

, Kothari

, Stark

. Factors related to breastfeeding discontinuation between hospital discharge and 2 weeks postpartum. J Perinatal Educ, 2011; 20:36–44.

11.

Donath

, Amir

. Relationship between prenatal infant feeding intention and initiation and duration of breastfeeding: A cohort study. Acta Paediatr, 2003; 92:352–356.

12.

Stuebe

, Bonuck

. What predicts intent to breastfeed exclusively? breastfeeding knowledge, attitudes, and beliefs in a diverse urban population. Breastfeed Med, 2011; 6:413–420.

13.

Ajzen

. From intentions to actions: A theory of planned behavior. In: Action Control. Kuhl

, Beckmann

, eds. Springer, Berlin, Heidelberg, 1985, pp. 11–39.

14.

Centers for Disease Control and Prevention. Breastfeeding report card. Updated. 2016. Available at https://www.cdc.gov/breastfeeding/data/reportcard.htm (accessed October 13, 2016).

15.

Section on Breastfeeding. Breastfeeding and the use of human milk. Pediatrics, 2012; 129:e827–e841.

16.

Uddin

, Simon

, Myrick

. Routine Prenatal Care Visits by Provider Speciality in the United States, 2009–2010. US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics, 2014.

17.

Center for Disease Control and Prevention. Infant feeding practices study II and its year six follow-up. Updated. 2014. Available at www.cdc.gov/breastfeeding/data/ifps/index.htm (accessed February 16, 2016).

18.

Fein

, Labiner-Wolfe

, Shealy

, et al. Infant feeding practices study II: Study methods. Pediatrics, 2008; 122 Suppl 2:S28–S35.

19.

Rothman

, Greenland

, Lash

. Modern Epidemiology. Lippincott: Williams & Wilkins, 2008.

20.

Taveras

, Capra

, Braveman

, et al. Clinician support and psychosocial risk factors associated with breastfeeding discontinuation. Pediatrics, 2003; 112(1 Pt 1):108–115.

21.

Kozhimannil

, Attanasio

, Yang

, et al. Midwifery care and patient–provider communication in maternity decisions in the United States. Matern Child Health J, 2015; 19:1608–1615.

22.

, Fein

, Chen

, et al. Why mothers stop breastfeeding: Mothers' self-reported reasons for stopping during the first year. Pediatrics, 2008; 122 Suppl 2:S69–S76.

23.

Lau

. Effects of stress on lactation. Pediatr Clin North Am, 2001; 48:221–234.

24.

Dewey

. Maternal and fetal stress are associated with impaired lactogenesis in humans. J Nutr, 2001; 131:3012S–3015S.

25.

Andresen

, Telleen

. The relationship between social support and maternal behaviors and attitudes: A meta-analytic review. Am J Community Psychol, 1992; 20:753–774.

26.

Costanian

, Macpherson

, Tamim

. Inadequate prenatal care use and breastfeeding practices in Canada: A national survey of women. BMC Pregnancy Childbirth, 2016; 16:100.

27.

Ridic

, Gleason

, Ridic

. Comparisons of health care systems in the United States, Germany and Canada. Mater Sociomed, 2012; 24:112–120.

28.

Starfield

, Shi

, Macinko

. Contribution of primary care to health systems and health. Milbank Q, 2005; 83:457–502.

29.

Shi

. The impact of primary care: A focused review. Scientifica (Cairo), 2012; 2012:432892.

30.

Ridic

, Gleason

, Ridic

. Comparisons of health care systems in the United States, Germany and Canada. Mater Sociomed, 2012; 24:112–120.

31.

Fisher

, Kane

. Consumer panelist versus random digit dial respondent performance revisited: How similar and how different. Res Res, 2004; 64.

32.

Hyman

, Lamb

, Bulmer

. The Use of Pre-Existing Survey Questions: Implications for Data Quality. Proceedings of the European Conference on Quality in Survey Statistics. 2006. Available at: https://pdfs.semanticscholar.org/4055/8314256349151a3cbbc7914b50aed99e1ff9.pdf (accessed December 7, 2017).

33.

, Scanlon

, Serdula

. The validity and reliability of maternal recall of breastfeeding practice. Nutr Rev, 2005; 63:103–110.