Breastfeeding Intentions of Female Physicians

Abstract

Background:

Methods:

We report data gathered from 50 physician volunteers, mainly affiliated with Johns Hopkins University (Baltimore, MD), using a questionnaire.

Results:

Consistent with previous physician studies, we found high breastfeeding initiation rates among our participants. However, the breastfeeding continuation rates of mothers in our study at 6 and 12 months were higher than those reported in previous physician studies. Our data showed that while physician mothers intended to breastfeed 64% of the infants for at least 12 months and while 97% of infants were breastfed at birth, only 41% continued to receive breastmilk at 12 months. This discrepancy suggests that work-related factors may influence physician mothers' breastfeeding behavior and might have a larger impact than these mothers' education and intentions on breastfeeding duration.

Conclusion:

This finding supports implementing workplace strategies and programs to promote breastfeeding duration among physician mothers returning to work.

Background

Based on extensive evidence of health benefits for both infant and mother, breastfeeding has become the preferred modality of infant feeding.^1–3 Although breastfeeding initiation rates have been increasing in the United States, rates of continuation and exclusive breastfeeding are still below desired levels.⁴ Interventions targeted at exclusive breastfeeding and breastfeeding duration are needed in order to achieve the Healthy People goals for breastfeeding.^5,6

Physician breastfeeding counseling is one of the interventions that successfully increase breastfeeding initiation and duration.^7–11 A strong predictor of physicians' breastfeeding advocacy is their successful personal or spousal breastfeeding experience.^12–15 To date, our understanding of breastfeeding among physicians in the United States has been based on cross-sectional questionnaire surveys administered to residents and/or practicing physicians.^16–19 Results of these studies suggest that while female physicians have excellent breastfeeding initiation rates, their continuation rates are lower than the general population.^16–20 Previous studies identified return to work, work schedule, diminishing milk supply, and lack of adequate time for milk expression as reasons for breastfeeding cessation among this population.^16–19 In fact, Miller et al.¹⁶ reported that 50% of resident mothers who had initiated breastfeeding weaned around the time of returning to work from maternity leave. Physician mothers who continued to breastfeed after return to work identified insufficient time and lack of appropriate place at work for milk expression as obstacles to breastfeeding continuation.^16–18 However, because of lack of data about breastfeeding intentions of surveyed mothers, definitive conclusions could not be made regarding whether the drop in physician mothers' breastfeeding rates after return to work resulted from their decision to wean earlier than recommended or whether it reflected influence of workplace-related factors that discouraged breastfeeding maintenance despite maternal intention to continue.

The purpose of this study was to determine initiation and duration rates of breastfeeding and exclusive breastfeeding in physician mothers, compare intention and duration rates, and identify modifiers for the decision to wean. Based on results of previous studies, we hypothesized that physician mothers would have excellent breastfeeding initiation rates and that they, particularly those in training at the time of childbirth, would be at risk of not maintaining desired breastfeeding practices because of personal and institutional factors.

Methods

Survey instrument development

Previous studies published on breastfeeding among physicians in the United States were reviewed, and questionnaires utilized for these investigations were requested. After review of published data and questionnaires that were made available,^12,14–19 areas pertaining to breastfeeding among physicians were identified. Survey items and response scales were developed to assess those areas, including breastfeeding intentions. The questionnaire was piloted among 20 Internal Medicine and Pediatric research fellows and faculty at Johns Hopkins University School of Medicine (Baltimore, MD) to ensure clarity. The criteria for participation in the pilot were (1) being present at a meeting of “Research in Progress” attended by faculty members and research fellows in General Internal Medicine and Pediatrics during which the survey was piloted and (2) volunteering to participate. The survey was administered to all volunteers, regardless of their gender, stage of career, having had biologic children, and infant feeding method. We sought feedback to “improve the survey instrument” and considered all suggestions, including those to improve clarity and readability. The questionnaire was revised based on suggestions of participants. Pilot participants were not included in the subsequent study. The final instrument contained 49 items and took approximately 15–30 minutes to complete. Questions included demographic information and previous breastfeeding education. Participants were asked a series of questions for each of their children, including current age, whether or not the infant was ever breastfed, and age at which the infant first received anything other than breastmilk.

Recruitment

Once approved by our institutional review board, recruitment was initiated. The principal investigator e-mailed the head of the Institution's Women's Task Force as well as residency program directors affiliated with Johns Hopkins University School of Medicine. The recruitment e-mail was sent one time and contained information about the nature of the study and e-mail and phone number of the Principal Investigator. Recipients were asked to disseminate the e-mail to their respective programs, so that physicians interested in participating in the study could contact the Principal Investigator.

We intended to interview as many mothers as possible with the goal of completing interviews in August 2009, the latest date the Principal Investigator would be able to conduct the interviews in person. Becausee physician mothers continued to enroll in the study throughout the study period, follow-up or reminder e-mails were not sent to the program directors. Interviews were set up as potential participants contacted us.

Participants

Criteria for participation included being a female physician and having had at least one biological child. Physicians were included whether they were in training (e.g., resident or fellow) or had completed training (e.g., faculty at academic site or community practice). Participants were included, regardless of their infant feeding methods. Initially, recruitment efforts only focused on physicians affiliated with Johns Hopkins University School of Medicine. However, when physicians not affiliated with the institution at the time of the study volunteered to participate, we included them as long as they were otherwise eligible.

Study

Fifty-two mothers volunteered to participate in the study from February to August 2009. All participants had had at least one biological child. Twelve physician mothers were in training, and 40 had completed training. Every attempt was made to conduct the interviews in person, but this was not always possible because of time constraints. Twenty-nine interviews were conducted in person, 11 physicians were interviewed on the phone, and 12 participants filled out the questionnaire on paper or online. Calls were made to all of these 12 physicians to clarify answers and to obtain further feedback about the participants' breastfeeding experience and about the questionnaire itself. One investigator (M.S.) conducted all interviews (by phone and in person).

Analysis

Descriptive statistics were calculated with SPSS software version 16 (SPSS, Chicago, IL). Two-tailed z-test comparisons of percentages were conducted via an online z-test calculator (www.dimensionresearch.com/resources/calculators/ztest.html). All comparisons were performed at a 95% confidence level.

Results

Fifty-two surveys were completed, and 50 of the survey results were used: One survey was not included as the participant was a Ph.D. and not an M.D., and the second was not complete, and we were unable to contact the participant to clarify and complete the answers.

The participants' ages ranged from 28 to 54 years at the time of study, with an average age of 36.8 years. Twelve (24%) were still in training, and 38 (76%) had completed training. Forty-four (88%) worked at Johns Hopkins Hospital or affiliated institutions, four participants (8%) now practiced or were doing research elsewhere, and two (4%) practiced in institutions not affiliated with Johns Hopkins University but learned about the study through their spouses who were affiliated with institution. Participants' specialties are described in Table 1. Additional demographics are characterized in Table 2. Only 22.1% (n = 19) of physicians in the survey reported receiving breastfeeding education in medical school and fewer still (2.3%; n = 2) during residency.

Table 1.

Specialty Distribution of Physician Mothers in the Study (n = 50)

Specialty	Number	%
Anesthesiology	3	6
General Surgery	2	4
IM or any IM subspecialty	34	78
IM/Pediatrics	1	2
Obstetrics/Gynecology	2	4
Neurology	4	8
Pediatrics	1	2
Rehabilitation	1	2
Ophthalmology	1	2
Urology	1	2

IM, Internal Medicine.

Table 2.

Demographic Data for Physician Mothers in the Study (n = 50)

	Number	%
Age (years)
28–30	5	10
31–34	13	26
35–39	20	40
40 +	12	24
Marital status
Married	45	90
Not married	5	10
Training level
Post-training	38	76
In training	12	24
Number of children
1	20	40
2	25	50
3	4	8
4	1	2

The 50 physicians included in the study had a total of 86 children, ranging in age from 10 weeks to 23 years old. Thirty-seven of the children (43%) were born after maternal completion of training, and 36 (42%) were born during the mother's training. More specifically, two children were born prior to the mother entering medical school, nine during the mother's medical school years, 19 while the mother was in residency, 17 during the mother's fellowship, and 37 after completion of maternal training. Two children were born after the mother had finished medical school and before she started residency (one mother was a foreign medical graduate preparing for the examinations, and the second one took a leave between medical school and residency).

Among this population, breastfeeding initiation rate was 97%, and continuation rates were 72% and 41% at 6 months and 12 months, respectively. We used the infant as the unit of analysis for calculation of rates because breastfeeding practices of some multiparous mothers were not similar for different offspring. In other words, 83 of the 86 children (96.5%) received breastmilk at birth, 62 (72.1%) continued to receive breastmilk at 6 months, and 35 (40.7%) continued to do so at 12 months (Fig. 1). Overall, 72 (83.7%) were exclusively breastfed at birth, 47 (54.7%) at 3 months, and 25 (29.1%) at 6 months of age (Fig. 1). Eleven children were still breastfeeding at the time of study.

FIG. 1.

Breastfeeding rates of physician mothers in our study: (left) exclusive breastfeeding rates at hospital discharge (initiation), 3 months, and 6 months and (right) rates of any breastfeeding at hospital discharge (birth), 6 months, and 12 months.

The median breastfeeding duration of infants breastfed during training was 10.17 months, compared to 10.3 months for infants breastfed after completion of training (z = −0.272). The rates of breastfeeding initiation and continuation as well as those of exclusive breastfeeding were generally lower in our study for infants breastfed during training than those breastfed after completion of training (Table 3). However, the differences between the two groups were not statistically significant.

Table 3.

Breastfeeding Rates During Training and Post-Training

	Infants breastfed in training (n = 36)	Infants breastfed after training (n = 37)	z value
Any breastfeeding
Birth	80.6%	83.8%	0.054
6 months	69.4%	70.3%	0.178
12 months	33.3%	43.2%	0.63
Exclusive breastfeeding
Birth	94.4%	97.3%	0.026
3 months	61.11%	54.1%	0.373
6 months	22.2%	32.4%	0.715

Interestingly, all mothers (100%) reported planning to breastfeed before childbirth. Thirty-two (64%) wanted to breastfeed their infant(s) for 12 months, seven (14%) for 6 months, and five (10%) for 3 months. The remaining six mothers did not state specific duration goals. The mothers' goal for duration of breastfeeding had a statistically significant association with their actual breastfeeding duration (r = 0.279, p < 0.05).

Reasons for intent to breastfeed mainly involved infant and/or maternal health, but also included bonding, ease of feeding, less expense, and obstetrician's recommendations. Reasons for supplementation with formula depended on the age of the infant at the time of supplementation and were mainly related to infant prematurity or low birth weight, inadequate milk supply, return to work/work schedule, and inability to breastfeed or express milk. Also, complementary foods such as cereal were introduced between 4 and 6 months for some infants because of pediatrician recommendations. Reasons for complete breastfeeding cessation also depended on the age of infant and are summarized in Table 4. The two most frequent reasons for breastfeeding cessation between 1 and 12 months were inadequate milk supply and lack of sufficient time at work to express milk. One mother eloquently stated that “lack of time led to lack of adequate milk supply.”

Table 4.

Reasons for Breastfeeding Cessation (n = 86)

	≥12 months	6–11.9 months	1–5.9 months	<1 month
Other	40%
Infant interest	34%	33%
Inadequate milk supply	17%	37%	83.3%
Inadequate time	8.6%	44.4%	25%
Too stressful	5.7%	7%
Maternal/infant health	5.8%		8%	100%
Maternal discomfort	2.9%
Maternal interest	2.9%	7%

Mothers could provide multiple reasons.

The one-on-one interaction with the survey participants allowed for collection of descriptive and qualitative data, even in response to forced response items in the questionnaire. Mothers in surgical specialties and procedure-based subspecialties (e.g., GI) universally reported one of their main challenges to be unavailability of lactation rooms close to operating rooms. Other examples mentioned influence of mother's residency rotations on milk supply. For example, one participant reported having no trouble maintaining breastfeeding during one rotation that “did not involve overnight call,” while experiencing a significant drop in her milk supply during a different rotation that involved overnight call and was more stressful.

Discussion

Our data demonstrated that 97% of infants were breastfed at birth, intent to breastfeed for at least 12 months was 64%, and yet only 41% of infants continued to receive breastmilk at 12 months of age. This discrepancy and the reasons for cessation (lack of time and adequate milk supply) between 1 and 12 months suggest that work-related factors not only influence physician mothers' breastfeeding behavior, but might have a larger impact than these mothers' education and intentions on their breastfeeding duration. In other words, physician mothers in this pilot study reported intention to breastfeed as well as awareness of benefits of breastfeeding and current recommendations. Their intentions and knowledge correlated with their breastfeeding initiation practices. However, their breastfeeding maintenance was then determined by interaction of personal factors, such as intent and knowledge, with work-related ones. Improving breastfeeding duration of physician mothers requires identification of other work-related modifiers and recognition of importance of institutional factors, such as accessibility to breastfeeding opportunities.

We were interested in whether maternal stage of career at time of childbirth might affect breastfeeding duration and included all physician mothers, regardless of their stage of career at the time of child birth. With the exception of Arthur et al.,^17,18 previous published physician studies have only included mothers during residency training. In our study, 43% of the children were born after their mothers had completed training. There were not sufficient numbers of mothers in each stage of career subset for a statistically meaningful comparison, but breastfeeding rates were generally lower for infants breastfed during training than those breastfed after completion of training (Table 3). Therefore, it is conceivable that our results might reflect the different breastfeeding practices of physician mothers based on their stages of career at the time of childbirth and breastfeeding. The working environments of a medical student, a resident, a fellow, and an attending physician could be very different and contribute to the ability and ease of the mother to breastfeed. Return to fulltime employment outside of home within a year after delivery has been identified as a barrier to breastfeeding in the general population.^21–23 Maternal full-time employment outside the home does not affect breastfeeding initiation in other professions, but it was associated with shorter breastfeeding duration.^23,24 However, women employed as professionals breastfeed longer than other working mothers.^24,25 It is possible that attending physicians have more control over their environment and schedules and are able to combine breastfeeding and working more successfully than physicians in training and medical students.

Maternal specialty might be another factor that influences breastfeeding duration of a physician mother. Thirty-nine (78%) of our participants were either training or had completed training in Internal Medicine or one of its subspecialties, a higher percentage than those reported by previously published physician studies.^16–19 The number of mothers in each specialty in our study was too small for detection of a statistically significant difference, but it is possible that physician mothers in some specialties/subspecialties have longer breastfeeding duration than others. Regardless of their specialty, all physician mothers face challenges universal to working mothers in return to work. However, each medical specialty also presents its own unique set of difficulties, inherent to its environment and “culture,” potentially making some specialties more “baby-friendly” than others. This is supported by anecdotal evidence from our study that suggest milk supply might be affected by overnight call, the level of stress experienced during different rotations, and availability and proximity of lactation rooms. Some differences in specialties might be institution-dependent, as attested to by the mothers in our study who returned to clinical work mainly in the operating room after maternity leave. Given the amount of time they had between cases and the distance between operating rooms and lactation rooms, many found it impractical to walk to and from the main lactation room in the hospital and chose instead to express milk in less optimal locations, such as unused operating rooms.

The high percentage of participants from Internal Medicine or Internal Medicine subspecialties suggests a potential recruitment bias that might be associated with the Principal Investigator's affiliation with the Internal Medicine department. Uneven distribution of the recruitment e-mail by program directors in different specialties might have also contributed to the recruitment bias. Although we attempted to recruit physician mothers who had not breastfed as well as those who had, it is possible that mothers who had chosen to breastfeed and had achieved their goal were more likely to volunteer to participate in the study. As such, another potential limitation of our study is selection bias resulting in over-representation of actual breastfeeding rates among physician mothers.

Another limitation is recall bias because the data gathered represented previous breastfeeding behavior. Retrospective infant feeding data based on maternal recall is a valid and reliable estimate of breastfeeding initiation and duration, especially when the duration of breastfeeding is recalled within 3 years.^26,27 However, validity and reliability of maternal recall for age of introduction of food and liquids other than breastmilk (e.g., duration of exclusive breastfeeding) are less satisfactory, and validity and reliability of maternal recall for breastfeeding intention are not clear.²⁷ A further potential limitation of this study is institutional bias as our study was mainly a single center study at an academic medical center. Although this allowed for detailed, in-depth analyses with extensive one-on-one interviews, the single institution aspect limits the generalizability of the findings.

Conclusions

It is known that physician mothers' breastfeeding behavior impacts their anticipatory guidance to their patients, which in turn influences patients' breastfeeding initiation and continuation.^7–15 As such, studying physician mothers' breastfeeding behavior is important, since it impacts not only the well-being of themselves and their families, but eventually the well-being of their patients and patients' families. Previous studies have indicated that despite excellent breastfeeding initiation rates, female physicians generally do not maintain breastfeeding as long as recommended.^16–20 We had previously hypothesized that physician mothers as a group are at risk of premature breastfeeding cessation.²⁰ The high breastfeeding initiation rates observed among this group support the important role of education and occupation in their decision to start breastfeeding, but their previously reported poor duration rates suggest that these characteristics alone are not sufficient for breastfeeding maintenance.

Our findings suggest that work-related factors might have a larger impact on breastfeeding maintenance in this group of mothers than their breastfeeding intentions or education and support implementing workplace strategies and programs to promote breastfeeding duration among physician mothers returning to work. Formal maternity leave policies, non-clinical duties when physician mothers first return to work, protected time and space for milk expression, sanitary storage, onsite daycare, and support and reinforcement at the work site are modifiable factors that might influence physician mothers' breastfeeding duration after return to work.

To our knowledge, this is the first published report of breastfeeding intentions in this population. Our results indicate that breastfeeding among physicians and the impact of workplace on their breastfeeding behavior would be an area for further investigation. As the breastfeeding continuation rates found in our study are different than those reported in previous physician studies, more investigation is warranted from a more diverse sample to determine whether a shifting paradigm is occurring across institutions and regions. Future research should also attempt to assess whether breastfeeding patterns differ among physician mothers in training and those who have completed training and among physician mothers in different specialties.

Footnotes

Disclosure Statement

No specific financial support was obtained for preparation of this article. The authors have no potential conflicts of interest to declare with respect to this manuscript.

References

World Health Organization, United Nations Children's Fund. Global Strategy for Infant and Young Child Feeding. World Health Organization: Geneva, 2001.

Gartner

, Morton

, Lawrence

et al. Breastfeeding and the use of human milk. Pediatrics, 2005; 115:496–506.

, Chung

, Raman

et al. Breastfeeding and maternal and infant health outcomes in developed countries. Evid Rep Technol Assess, 2007; 153:1–186.

Centers for Disease Control Prevention. www.cdc.gov/BREASTFEEDING/DATA/NIS_data/index.htm. 2009 November 6.

U.S. Department of Health and Human Services. Healthy People 2010: Conference Edition, Volumes I and II. Public Health Service, Office of the Assistant Secretary for Health: Washington, DC, 2000; 47–48.

U.S. Department of Health and Human Services. Healthy People 2010 Midcourse Review. U.S. Department of Health and Human Services: Washington, DC, 2005.

Lawrence

. Practices and attitudes toward breast-feeding among medical professionals. Pediatrics, 1982; 70:912–920.

Taveras

, Li

, Grummer-Strawn

et al. Opinions and practices of clinicians associated with continuation of exclusive breastfeeding. Pediatrics, 2004; 113:e283–e290.

Sikorski

, Renfrew

, Pindoria

et al. Support for breastfeeding mothers. Cochrane Database Syst Rev, 2002; 1:CD001141.

10.

Kramer

, Chalmers

, Hodnett

et al. Promotion of Breastfeeding Intervention Trial (PROBIT): A randomized trial in the Republic of Belarus. JAMA, 2001; 285:413–420.

11.

Labarere

, Gelbert-Baudino

, Ayral

et al. Efficacy of breastfeeding support provided by trained clinicians during an early, routine, preventive visit: a prospective, randomized, open trial of 226 mother-infant pairs. Pediatrics, 2005; 115:e139–e146.

12.

Freed

, Clark

, Sorenson

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

13.

Schanler

, O'Connor

, Lawrence

. Pediatricians' practices and attitudes regarding breastfeeding promotion. Pediatrics, 1999; 103:E35.

14.

Freed

, Clark

, Lohr

, Sorenson

. Pediatrician involvement in breast-feeding promotion: A national study of residents and practitioners. Pediatrics, 1995; 96:490–494.

15.

Freed

, Clark

, Cefalo

et al. Breastfeeding education of obstetrics gynecology residents and practitioners. Am J Obstet Gynecol, 1995; 173:1607–1613.

16.

Miller

, Miller

, Chism

. Breastfeeding practices among resident physicians. Pediatrics, 1996; 98:434–437.

17.

Arthur

, Saenz

, Replogle

. The employment-related breastfeeding decisions of physician mothers. J Miss State Med Assoc, 2003; 44:383–387.

18.

Arthur

, Saenz

, Replogle

. Personal breast-feeding behaviors of female physicians in Mississippi. South Med J, 2003; 96:130–135.

19.

Kacmar

, Taylor

, Nothnagle

et al. Breastfeeding practices of resident physicians in Rhode Island. Med Health R I, 2006; 89:230–231.

20.

Sattari

, Levine

, Serwint

. Physician mothers: An unlikely high risk group—call for action. Breastfeed Med, 2009; 5:35–39.

21.

Karinu

, Sorino

, Ezrine

et al.

Does maternal employment affect breastfeeding?

Am J Public Health, 1989; 79:1246–1250.

22.

Gielen

, Faden

, O'Campo

et al. Maternal employment during the early postpartum period: Effects on initiation and continuation of breastfeeding. Pediatrics, 1991; 87:298–305.

23.

Calnen

. Paid maternity leave and its impact on breastfeeding in the United States: An historic, economic, political, and social perspective. Breastfeed Med, 2007; 2:34–44.

24.

Ryan

, Zhou

, Arensberg

. The effect of employment status on breastfeeding in the United States. Womens Health Issues, 2006; 16:243–251.

25.

Visness

, Kennedy

. Maternal employment and breastfeeding: Findings from the 1988 National Maternal and Infant Health Survey. Am J Public Health, 1997; 87:945–950.

26.

Launer

, Forman

, Hundt

et al. Matenral recall of infant feeding events is accurate. J Epidemiol Community Health, 1992; 46:203–206.

27.

, Scanlon

, Serudla

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