Improving Drug Exposure Knowledge During Lactation: Quality Improvement Initiative in Low-Income Women

Introduction

Breastfeeding provides substantial health benefits for mother and child.^1–3 However, there remains a unique opportunity to support women in meeting exclusivity and duration goals. By 6 months of life, only one in four infants in the United States and 40% of infants globally are exclusively breastfed.^4,5 Women discontinue breastfeeding for a number of reasons, including concerns regarding decreased supply, returning to work, or lack of privacy to pump milk. ⁶ Medication use is also a major reason women stop breastfeeding, specifically due to concerns for the health of their infants.^7–9 Given that evidence shows the majority of medications are compatible with breastfeeding, ¹⁰ interventions aimed at improving health literacy on this topic could conceivably improve breastfeeding outcomes.

When there are maternal concerns regarding breastfeeding and medication use, women may discontinue or reduce the length of time they are taking medications. ¹¹ Others may choose not to breastfeed or terminate breastfeeding earlier than expected. ¹¹ Medication use is also a major reason women with chronic illnesses are discouraged from breastfeeding. ⁹ Health care providers often incorrectly counsel women regarding medication use while breastfeeding and the need for cessation, which in turn negatively affects breastfeeding rates.^7,8 Similarly, information regarding medication use and lactation available to the general public online may be unreliable. ¹² Furthermore, substance use (such as tobacco, alcohol, and opioids) is common in women of reproductive age. Some of these women and their infants benefit from breastfeeding. ¹²

While there are numerous efforts to improve provider knowledge regarding medication or substance use during pregnancy and lactation, there are few initiatives aimed at improving maternal knowledge. ⁷ The need for research specific to lactation is so great that a Congressional Task Force was established ¹³ and the taskforce identified the urgent need for research to improve knowledge surrounding medication usage and lactation. ¹⁴ We conducted a quality improvement (QI) initiative to assess and improve knowledge on medication/substance use and breastfeeding by providing prenatal lactation education for expectant low-income mothers.

Materials and Methods

This QI initiative was reviewed and determined to be exempt by the Institutional Review Board. The eligible population included pregnant women at any gestational age receiving prenatal care at four state-funded prenatal care centers serving low-income women in Hillsborough county, Florida. Participants spoke and understood English or Spanish. Postpartum patients and women with a contraindication to breastfeeding, such as HIV, were excluded. All participants provided informed consent.

Breastfeeding education sessions occurred from August to December 2016. Each woman participated in one group session at her prenatal care site entitled Milk and Meds. The educational sessions were presented in interactive PowerPoint format, lasted an hour, and were separate from routine prenatal care appointments. Topics included biology of breastfeeding, benefits of breastfeeding, infant feeding recommendations, and human milk expression and storage. Also discussed were common medical problems while breastfeeding, such as hypertension, diabetes, and depression, and compatibility of common medications used in the postpartum period (e.g., pain medications, antihypertensives, insulin, antidepressants). Education was also provided on substance use including marijuana, alcohol, and tobacco during breastfeeding. The informational materials were adapted from the American College of Obstetricians and Gynecologists breastfeeding support toolkit and patient friendly medication and lactation resources from the Organization of Teratology Information Specialists. ¹⁵ Innovative breastfeeding videos were also utilized.^16–18 Sessions were facilitated by credentialed doulas who are also certified lactation counselors with support from an international board-certified lactation consultant (IBCLC). Participants received a $20 incentive to participate in the session.

Data were collected at five time points between August 2016 and October 2017. Before the educational session, baseline demographic information, medical history, breastfeeding knowledge, and the infant feeding intention (IFI) scales were collected. The IFI is a validated scale designed to quantitatively measure maternal breastfeeding intentions. It is a 5-item self-report scale, which assesses if a mother intends to breastfeed or formula feed between 0 and 6 months postdelivery.^19,20 Each item has a 5-point Likert response option ranging from very much agree to very much disagree.^19,20 The scores range from 0 to 16 with a higher score indicating stronger intention to breastfeed exclusively for the first 6 months postpartum.^19,20 Postintervention, women completed breastfeeding knowledge and IFI scales, and provided a list of current medications they were taking and any potential concerns related to medication use while breastfeeding. The investigator developed breastfeeding knowledge questionnaire to assess participants' knowledge of the compatibility of medication and ramifications of substance use during breastfeeding. Questions include insulin, marijuana, alcohol, prescription opioid, antidepressant, and antihypertensive use while breastfeeding (Appendix Table A1). Delivery outcomes, infant and/or maternal complications, medication use, and infant feeding data were abstracted from the medical records. Detailed information on infant feeding and medication compliance were collected by doulas at home visit or by the research assistant through phone calls at 2–4 weeks postpartum. Data were collected on infant feeding and medication compliance by doulas or research assistant at visits or through phone calls at 6–8 weeks postpartum. Medical records were reviewed to assess directives given or information provided by health providers regarding breastfeeding cessation related to medication use.

Descriptive statistics were used to evaluate demographics and to compare characteristics of mothers who were taking medications at baseline to mothers who were not taking medications. Chi-square was utilized for categorical variables and McNemar's test was used for difference in knowledge score proportions pre- and postintervention. Logistic regression analysis was used to test for significant associations between medication use, baseline variables, and breastfeeding exclusivity. Data were considered to be missing completely at random and complete case analysis was used. An a priori level of 0.05 was used to determine significance. All analyses were conducted using IBM SPSS Statistics version 24.

Results

A total of 122 participants completed the education sessions; 1 participant was ineligible, resulting in a sample of 121 women. Baseline characteristics of the women are presented in Table 1. Participants had a mean age of 26.47 ± 4.98 and a mean gestational age of 25.31 ± 10.29. The majority were unmarried (68.9%), had at least a high school diploma/GED (75.2%), and reported Hispanic ethnicity (63.6%). Nearly 70% had health insurance, with Medicaid serving as the primary insurer for >80% of those who were insured. Most of the participants were unemployed (60.8%) and receiving Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) benefits (68.6%). Approximately three-quarters reported a household income of less than $30,000 per year.

Nearly 25% (n = 28) of women were on medications other than supplements such as prenatal vitamins and iron (Table 2) at the time of education. The most common medications used were antibiotics (5% [n = 6]), hypoglycemic (3.3% [n = 4]), and injectable anticoagulants (2.5% [n = 3]). Less than 10% (n = 10) of all women reported concerns prenatally regarding medication usage and infant safety during lactation. Knowledge regarding common postpartum medications and compatibility with breastfeeding increased significantly in five out of six areas (opioids, antidepressants, antihypertensives, insulin, marijuana, and alcohol) following the educational session (Table 3).

Despite increases in knowledge posteducation, women who were using medications at 2–4 and 6–8 weeks postpartum were less likely to exclusively breastfeed and 4 times as likely to mix/complementary feed or formula feed compared with postpartum mothers not taking medications, adjusted OR at 2–4 weeks = 3.150 (95% CI: 1.292–7.684) and adjusted OR at 6–8 weeks = 4.305 (95% CI: 1.076–17.225; Table 4).

Discussion

Our participants were primarily single, Hispanic, low-income, and enrolled in WIC. Previous research has indicated these factors are associated with lower exclusive breastfeeding rates among women. ⁴ Although a quarter of our study population were on medications other than prenatal vitamins and iron in the prenatal period, <10% reported concerns regarding medication usage. The percentage of women reporting concerns regarding medication use and lactation in our study is less than what is described in the literature. ⁸ A qualitative study by Spiesser-Robelet et al. explored the educational needs of 19 women in the immediate postpartum period through interviews and approximately one-third (31.6%) of these women reported feeling anxious about taking medication while breastfeeding. ²¹ The timing may contribute to the difference in reported concern. Perhaps the participants' concerns are higher when initiating breastfeeding as compared with the prenatal period or when breastfeeding is already established. Given that many of our patients were of lower socioeconomic status (SES), they may have been more focused on other responsibilities, such as providing food, shelter, transportation to appointments, and may have been less concerned about medication use while breastfeeding.

Following the educational session, knowledge regarding medication and substance use during lactation increased significantly in five out of six areas. Baseline knowledge scores indicated the majority of women were already aware of the risks of alcohol consumption during pregnancy therefore, although scores increased, the level did not reach statistical significance. Despite the increase in self-reported knowledge and limited reports of medication concerns, medication usage appeared to influence breastfeeding postintervention. Based on our results, women taking medication were more likely to use mixed feeding (breast milk and formula) or formula alone at 2–4 and 6–8 weeks postpartum. This is consistent with results of previous studies. In the 2017 literature review by Spiesser-Robelet et al., the authors found that depending on the study, a range of 2.4% to 26.0% of women refuse to breastfeed while taking medication. ¹¹ Among women already taking medication for chronic conditions, 11.5% to 88% of the women in these studies reported they did not initiate breastfeeding due to medication use. ¹¹ While these are wide ranges, they do support the idea that medication usage is a deterrent to breastfeeding.

Also highlighted by this initiative is the importance of educating patients when it comes to breastfeeding. While there are many efforts geared at educating providers, there are few initiatives aimed at educating and empowering patients to seek additional resources and advice from obstetricians, midwives, or lactation consultants. Studies evaluating the knowledge and practice of nonobstetric providers regarding medication and breastfeeding support a need for additional provider education. As evidenced by Hussainy and Dermele's review, many providers advise women to cease medications while breastfeeding, stop breastfeeding temporarily or permanently, and even advise against initiating breastfeeding when a woman is taking medication. ⁸ While the focus of our study was to educate patients, the effects of this deficit in provider knowledge are illustrated by our patient cases.

Even after completing the course, women followed the advice of the nonobstetric providers and subsequently stopped breastfeeding. While in the first case, the diagnosis of myocardial infarction and addition of multiple new medications at once may have been overwhelming to the patient, the second case is especially unfortunate, because women should be encouraged to continue breastfeeding, rather than to stop, when diagnosed with mastitis. For the remaining cases, the importance of close follow-up and support for breastfeeding women is illustrated. Because these patient were enrolled in this QI project, they were contacted at 2–4 and 6–8 weeks postpartum. Had they not been enrolled, they may not have been contacted and educated that these drugs (acetaminophen, labetalol, erythromycin) are safe while breastfeeding and subsequently not resumed breastfeeding. These cases highlight the importance of educating mothers, especially those who are underserved and may feel uncomfortable advocating for themselves, and the importance of increased postpartum support of breastfeeding mothers. While the primary goal of our QI initiative was to educate women, an important aspect of this education was to provide them with reliable and easily accessible resources. Our hope was that this would provide them with further learning opportunities once the class was over. Health care providers may misinform or misdirect patients regarding medication use and breastfeeding,^7,8 and so our hope was that these women would feel empowered to share this evidence-based information with providers who may not have access to these resources, but often interact closely with women in the postpartum period. ⁸

Two major research implications emerged from our QI initiative. Although the initiative improved patient knowledge on medication/substance use and lactation, postpartum medication use as well as nonobstetric providers may have influenced women's breastfeeding decisions. Future research in this area should focus on ways to not only educate women, but also health care providers that frequently interact with women during the postpartum period. A future goal is to provide women with a small, easy-to-carry resource, such as a card or magnet, with resources on breastfeeding and medication use that they can use themselves or share with health care providers. This may encourage more in-depth discussion and help educate providers so that patients are not incorrectly advised to stop breastfeeding.

We assessed breastfeeding rates and medication use at three time points from delivery until 6–8 weeks postpartum in a diverse population. We implemented a sustainable, innovative model of prenatal lactation education with ongoing classes and an opportunity for one-on-one counseling. We used evidence-based tools to educate participants and employed multimodal approaches to keep participants engaged and appeal to different learning styles. This model could be used for future education sessions given the evidence-based and generalizable approach.

The most notable limitation of this initiative was that there was no comparison group. While a randomized controlled trial (RCT) was considered, given the population with limited education and low SES, we did not want to deny any woman interested in breastfeeding education. Future work could include a RCT with more than one experimental arms to compare intervention effectiveness. Results from this study, such as the proportion of women with concerns about breastfeeding and medication use, could be used to inform sample size calculations for future intervention studies. Data from this initiative were reported for a smaller number of women and adjusted analyses resulted in wider confidence intervals. Future work with larger sample sizes may complement these results.

Conclusion

Empowering patients with evidence-based, accurate information about medication and substance use during lactation may help women to make informed infant feeding decisions. From our QI initiative, we learned that interventions in patient education, with a focus on prescribed or recreational substance use, are beneficial in increasing medication-specific knowledge among diverse low-income women.