OBJECTIVE: We tested the hypothesis that removing pacifiers from routine distribution in our mother-baby unit (MBU) would be associated with greater breastfeeding initiation or exclusivity during the birth hospitalization.
METHODS: We retrospectively compared exclusive breastfeeding, breastfeeding plus supplemental formula feeding, and exclusive formula feeding rates for 2249 infants admitted to the MBU at our university teaching hospital during the 5 months before and 8 months after restriction of routine pacifier distribution. Formula supplementation, if not medically indicated, was discouraged per standard practice, but access to formula was not restricted.
RESULTS: Of the 2249 infants, 79% were exclusively breastfed from July through November 2010, when pacifiers were routinely distributed. During the 8-month period after pacifier restriction, this proportion decreased significantly to 68% (P < .001). A corresponding increase from 18% to 28% was observed in the number of breastfed infants receiving supplemental formula feeds in the same period (P < .001). During the study period, the proportion of exclusively formula-fed infants increased from 1.8% to 3.4% (P < .05).
CONCLUSIONS: Restricting pacifier distribution during the newborn hospitalization without also restricting access to formula was associated with decreased exclusive breastfeeding, increased supplemental formula feeding, and increased exclusive formula feeding. Because high-quality, prospective medical literature addressing pacifier use and breastfeeding does not conclusively show an adverse relationship in women who are motivated to breastfeed, more studies are needed to help determine what effect, if any, pacifiers have on breastfeeding initiation and exclusivity in the immediate newborn period.
- AAP —
- American Academy of Pediatrics
- BFHI —
- Baby-Friendly Hospital Initiative
- EBF —
- exclusive breastfeeding
- EMR —
- electronic medical record
- MBU —
- mother-baby unit
What’s Known on This Subject:
Pacifiers may interfere with breastfeeding and thus are discouraged until 3 to 4 weeks of life, when they are recommended for sudden infant death syndrome risk reduction. Hospitals are restricting pacifier distribution as part of the Baby-Friendly Hospital Initiative.
What This Study Adds:
We describe a temporal association between reduced exclusive breastfeeding and pacifier restriction. This observation encourages research on breastfeeding promotion and the effects of pacifiers and pacifier restriction on breastfeeding.
Parents are faced with a dilemma when deciding whether to offer a pacifier to their newborn infants in the first days to weeks of life. Hospitals have traditionally provided pacifiers at birth; however, the American Academy of Pediatrics (AAP) Section on Breastfeeding recommends avoiding pacifier use until breastfeeding is fully established, at ∼3 to 4 weeks of age.1 Complicating the parents’ decision, to help prevent sudden infant death syndrome, the AAP Task Force on Sudden Infant Death Syndrome 2010–2011 issued a level A recommendation to consider offering a pacifier at naptime and bedtime.2 The AAP recommends postponing pacifier use until breastfeeding is well established, citing the Section on Breastfeeding’s policy statement.
The AAP and the American Academy of Family Medicine recommend exclusive breastfeeding (EBF) for the first 6 months of life, followed by continued breastfeeding with the addition of complementary foods for 1 year or longer as mutually desired by mother and infant.1,3 Likewise, the American College of Obstetricians and Gynecologists recommends EBF for ∼6 months and states that ongoing breastfeeding is also beneficial.4 World Health Organization recommends EBF for the first 6 months of life and continued breastfeeding with the addition of complementary foods into at least the second year of life.5 The World Health Organization, in partnership with the United Nations Children’s Fund, also publishes a list of Ten Steps to Successful Breastfeeding, which forms the basis for the Baby-Friendly Hospital Initiative (BFHI), a global program aimed at recognizing hospitals providing the recommended standard of care for postpartum mothers and infants.6,7 These steps are listed in Table 1, and they include not providing pacifiers to breastfeeding infants. Less than 1% of US hospitals had implemented all 10 steps in 2007 or 2009, according to the results of the national Maternity Practices in Infant Nutrition and Care survey, with only a slight improvement in the percentage of hospitals implementing 9 of the 10 steps during that 2-year period: from 2.4% in 2007 to 3.5% in 2009.8,9 The CDC breastfeeding report card from 2012 states that 6% of births in the US now occur in Baby Friendly facilities, which is up from less than 2% in 2008.10 The US Surgeon General’s Call to Action to Support Breastfeeding recommends that hospitals accelerate implementation of the BFHI Ten Steps.11 Similarly, increasing the proportion of births taking place in facilities providing the recommended level of care for lactating mothers and their newborn infants is one of the Healthy People 2020 breastfeeding objectives.12
The Ten Steps to Successful Breastfeeding have been studied as a group. When instituted together as a complete bundle, they have resulted in improvements in multiple breastfeeding outcomes, including rate of breastfeeding initiation, duration of EBF, and duration of overall breastfeeding.6,13–15 Certain subsets of the Ten Steps are also thought to improve breastfeeding outcomes. In 2007, the Colorado Department of Public Health and Environment released a report detailing the results of a population-based study investigating hospital maternity care practices that led to increased breastfeeding duration.16 These practices, called the “Colorado Can Do 5!” are listed in Table 2. Both the Colorado Can Do 5! and the Ten Steps specify that no pacifiers be given to breastfeeding infants.
Conventional wisdom holds that pacifiers interfere with breastfeeding. Observational studies suggest a significant decrease in breastfeeding duration associated with pacifier use, but this association has not been confirmed.17 A Cochrane review analyzed data from 2 randomized or quasi-randomized controlled trials (RCTs), involving a total of 1302 infants, that compared pacifier use versus no pacifier use in healthy full-term newborns who had started breastfeeding.18 The review found no difference in breastfeeding duration at 3 or 4 months between the 2 groups. Jenik et al19 randomized 1021 mothers who were highly motivated to breastfeed and whose newborns had regained their birth weight by 15 days into 2 groups. One group did and the other did not offer a pacifier to their infants. No differences in breastfeeding prevalence or duration at 3 months were found. Kramer et al20 performed a double-blinded RCT of 281 healthy, breastfeeding mothers in which participants were randomly assigned to different types of lactation counseling. This counseling either did or did not recommend avoidance of pacifiers along with suggested alternative soothing techniques. The authors found an observational association between pacifier use and early weaning but did not find an association when data were analyzed according to group randomization. They concluded pacifier use may be a marker, rather than a cause, of breastfeeding difficulties. A systematic review17 of 4 RCTs revealed no difference in breastfeeding outcomes in newborns with the following pacifier interventions: pacifier use during tube feeds, pacifier use at any time after delivery, an educational program for mothers emphasizing avoidance of pacifiers, and a Baby-Friendly hospital environment. One of the 4 studies in this review was an RCT of 700 newborns randomized to early (2–5 days) versus late (5 weeks) pacifier introduction.21 It found a detrimental effect of early pacifier introduction on overall breastfeeding duration, as well as decreased EBF at 4 weeks among the early pacifier users compared with those who had not yet introduced them.
As part of our quest to improve our EBF rates and in pursuit of Baby-Friendly hospital designation, we discontinued routine distribution of pacifiers in our MBU while tracking the different types of feeding offered. The purpose of our quality improvement initiative was to test the hypothesis that eliminating routine pacifier distribution would improve EBF rates during the birth hospitalization.
A multidisciplinary work group at Oregon Health and Science University is currently implementing Baby-Friendly practices at our institution. As part of this project, and in accordance with The Joint Commission perinatal core measures, in July 2010 the multidisciplinary group began reviewing types of feeds provided to infants on our MBU. Nurses enter this information into our electronic medical record (EMR) as part of their regular patient care duties, and members of our informatics division pull this data for review on a monthly basis. Members of the work group monitor the EBF, breastfeeding plus supplemental formula feeding, and exclusive formula feeding rates; the data are then shared with staff. In December 2010, as one step in the process toward becoming Baby-Friendly, physicians and nurses in the MBU were instructed to cease routinely offering pacifiers to breastfeeding newborns. Pacifiers began to be stored in a standard, locked, supply management system. Before our new policy, pacifiers had been routinely provided. Our new policy was initiated in a multidisciplinary collaborative fashion, and anticipatory guidance was provided to parents per previous standard of care. No specific script was instituted to verbally instruct parents on infant soothing techniques or use of pacifiers.
Under the new policy, if a decision was made to offer a pacifier, the employee taking one from stock “signed it out,” providing a written explanation for why it was given. Parents could also bring pacifiers into the hospital, but this behavior was not monitored. Pacifier use was considered indicated for infants undergoing painful procedures such as circumcision and for newborns with neonatal abstinence syndrome. Information discouraging pacifier use in the hospital, as well as the remainder of the Colorado Can Do 5! have been displayed prominently at bedside in each room in English and Spanish since shortly after the pacifiers were restricted in December 2010. The specific text discouraging pacifier use reads: “Don’t give your baby a pacifier in the hospital. If you give your healthy newborn a pacifier, it can make it harder for your baby to learn to breastfeed and decrease your milk supply.” No changes were made during the study period in the method by which infants receive formula on the MBU; formula is pulled from stock and given to infants on parental request or, in rare cases, when it is medically indicated.
Data and Statistical Analyses
All presented feeding data and demographic information were entered into the EMR per our standard of care and retrospectively extracted for analysis. Categorical demographics before and after pacifier restriction were compared using the 2-tailed Fisher exact test when there were binary variables and χ2 test when there were more than 2 categories. Continuous demographic characteristics were compared using the 2-tailed t test. We compared the percentage of infants in each of the feeding groups before and after the restriction of pacifiers using the z test for 2 proportions. Feeding data selected for analysis were from July 2010, when the work group began reviewing feeding data, through August 2011, the most recent data available at the time of analysis. Infants born in December 2010, the month when pacifiers were removed, were excluded from analysis to allow a “wash-in” period for the policy to become the standard practice.
Institutional Review Board Determination
The project was submitted to the Oregon Health & Science University institutional review board and was determined exempt, as it did not meet the definition of human subject research per 45 CFR 46.102(f).
We reviewed information on 2249 infants: 812 in the preintervention group and 1278 in the postintervention group. We excluded the 159 infants hospitalized during December 2010, as the pacifier restriction was introduced during this month, thereby allowing for the policy to be implemented and become standard of care. Of the remaining 2090 infants, 2075 had at least some demographic data available for comparison: 800 of the 812 infants in the preintervention group and 1275 of the 1278 in the postintervention group. As shown in Table 3, infants in both groups had similar birth weights, gestational ages, genders, modes of delivery, and lengths of stay. Maternal age, race, ethnicity, and marital status also did not differ between the 2 groups. Data regarding primary language spoken by the mothers were more completely available in the postintervention group than in the preintervention group, making it difficult for us to determine whether the groups differed significantly with respect to primary language. We found a significant difference (P < .05) in the type of insurance coverage that mothers had before and after December 2010: 39% of mothers had private insurance in the first group versus 45% in the second group, and 57% were covered by a government plan in the first group compared with 52% in the second group. We had no information about maternal education level or intention to breastfeed.
Feeding data from July through November 2010 and from January through August 2011 were compared using the z test for 2 proportions. As shown in Fig 1, 79% of infants were exclusively breastfed in the first time period, when pacifiers were routinely distributed. This proportion decreased significantly to 68% in the postintervention period (P < .001). A corresponding increase from 18% to 28% was observed in the number of breastfed infants receiving supplemental formula feeds in the same postintervention period (P < .001). The percentage of exclusively formula-fed infants increased significantly from 1.8% to 3.4% (P < .05) after pacifiers were restricted.
Our study demonstrates that, in our specific MBU, restricting pacifier distribution without also restricting formula access to medical necessity or providing scripted information on infant soothing was temporally associated with decreased EBF, increased supplemental formula use, and increased exclusive formula feeding during the birth hospitalization. We looked for differences in several key maternal and infant variables that could explain the increase in formula feeding after pacifier restriction and found no likely causative factors. We found no significant differences in maternal age, race, ethnicity, infant gestational age, birth weight, mode of delivery, or length of stay. There seemed to be a difference in maternal primary language before and after our pacifier intervention, although there were many more women with unknown primary language in the preintervention group, making this apparent difference difficult to interpret. Insurance type, which may be a surrogate marker for socioeconomic status, differed significantly before and after the intervention. Interestingly, the percentage of mothers with private insurance increased and the percentage of mothers with a government funded plan decreased after the intervention. Because breastfeeding initiation rates are known to be higher among less impoverished households, we would expect the increase in the percentage of mothers with private insurance coverage to skew our findings in the opposite direction, toward an increase in breastfeeding rates in the postintervention period.22 Unfortunately, we have no data on maternal prepartum intention to breastfeed or maternal education level.
No major systems changes occurred during this time period, such as adoption of an EMR or change in nursing or lactation consultant staffing ratios. Two specific changes did occur during this time period. We increased our skin-to-skin rates immediately after birth, and many of our floor nurses worked toward becoming certified lactation consultants. Our findings stubbornly persisted despite these efforts to embrace BFHI principles.
Our study has many limitations. Most important, pacifier restriction was not absolute, as pacifiers could be brought in from home or obtained on request. In addition, staff did not provide families with standard and scripted information about soothing techniques or recognition of hunger cues, and individual staff may have had varying opinions on pacifier use. In our MBU, pacifier restriction was not accompanied by formula use specifically restricted to medical necessity as recommended in step 6 of the BFHI. Because formula could be accessed on demand, breastfeeding rates with pacifier restriction should be interpreted in this context. Research was performed in the setting of a hospital quality improvement initiative, so we present group-level demographic and feeding data before and after the intervention, but we do not have individual data about which specific infants did or did not use pacifiers and their feeding type. Some of our data, including mother’s stated primary language, are incomplete. We have no information on mothers’ breastfeeding intention, one of the most important factors in breastfeeding initiation and exclusivity.
A variety of other unmeasured variables may have confounded this study as well, including, but not limited to, maternal education, breastfeeding difficulties, and maternal medical problems. Institutional protocol and care pathway changes could also affect EBF rates. The most potentially influential protocol, screening for hypoglycemia, was in place and unaltered during the presented period of this study. Types of feeding presented are recorded per nursing standards of care. Better methods of quantifying EBF rates in hospitals, such as formula use by specific order in the EMR, may help researchers quantify formula use more precisely.
To assess whether our initial EBF rate of 78% was higher than usual for our institution, we have subsequently performed an unpublished retrospective review of EBF rates in our MBU, starting from 2009, when nurses began initially recording infant feeds into our EMR. During the time preceding our quality improvement initiative, our institution’s EBF rate never dropped below 78.5%. However, the data from the earlier time are less complete; fewer feeds are recorded on average for each infant, especially at the inception of our EMR reporting. We have continued to follow our rates over time, since preliminarily reporting our findings. The EBF rate has hovered tightly around 70% through our most recent reporting period of August 2012.
We emphasize that we are reporting a temporal association. We do not claim a cause-and-effect relationship, nor do we seek to change current recommendations regarding pacifier use in the earliest breastfeeding newborns. We simply strive to encourage additional research on the impact of pacifiers on breastfeeding, a common topic important to many families culturally and socially. Also, we believe hospitals seeking Baby-Friendly designation should consider that although bundling the Ten Steps together has been shown to have a positive effect on breastfeeding initiation and exclusivity, there may be a complex relationship among the effects of individual steps. Implementing some of the steps in isolation may not achieve the desired outcome. Provider training and information given to parents about methods of infant soothing may also need to be more standardized and more intensive when restriction of pacifier use is implemented. We acknowledge education regarding pacifier use for sleep-making versus early soothing (with its potential to miss hunger cues and deny an infant the opportunity to breastfeed) should be distinguished. In addition, we understand hospital distribution of products, including pacifiers, engenders implicit recommendations for their use.
Our observation occurs in the setting of Oregon’s very high breastfeeding initiation rates: 91.2% of children are ever breastfed, compared with the US average of 74.6%.23 Our hospital already practices many of the Baby-Friendly tenets, although we have not yet fully completed staff training requirements and our institution does not purchase formula. Distribution of formula to parents via gift bags at hospital discharge was banned in 2007. It remains to be seen whether other centers would replicate our findings with different social and cultural supports for breastfeeding or at a different step along the BFHI journey. Although we report findings only during the birth hospitalization, this period is a critical time for establishing long-term, EBF. A recent report investigating hospital practices and maternal intent to breastfeed suggests that not distributing pacifiers may be associated with EBF after discharge.24 But in that study, giving only breast milk in the hospital was by far the most important factor associated with mothers meeting their intent to exclusively breastfeed, and it was the only significant factor when adjusting for other practices, including pacifier distribution. A more thorough understanding and relative weighting of the impact and long-term implications of the Ten Steps are needed.
We report a temporal association between reduced EBF and pacifier restriction during the birth hospitalization. This observation occurred in the absence of accompanying restriction of formula to medical necessity and without provision of scripted information on infant soothing. Published data about the relationship between pacifier use and breastfeeding initiation, exclusivity, and duration are limited, and the best available evidence does not indicate that pacifiers interfere with breastfeeding in women who are motivated to breastfeed. Given the high-quality evidence supporting the recommendation to offer a pacifier before putting an infant down for sleep to protect against sudden infant death syndrome and other sleep-related death, it is important that pediatric health care providers have nuanced discussions with parents about the risks and benefits of pacifier use. We believe additional scientific inquiry is needed to support the current, universal recommendation of not offering pacifiers to breastfeeding infants in the first 3 to 4 weeks of life.
We thank Beven Kair, MS, for his assistance with statistical methods and Stanley Jones-Umberger, BSCS, for his help with data collection.
- Accepted December 12, 2012.
- Address correspondence to Laura R. Kair, MD, Mail Code CDRCP, 707 SW Gaines St, Portland, OR 97239. E-mail:
Dr Kair assisted with study design, performed data interpretation and analysis, and drafted the initial manuscript; Mr Kenron assisted with study design and data collection, assisted with data analysis, and revised the manuscript; Ms Etheredge assisted with study design and data collection, assisted with data interpretation, and revised the manuscript; Dr Jaffe assisted with study design, aided in data interpretation, and critically reviewed and revised the manuscript; Dr Phillipi conceptualized and designed the study, coordinated data collection, assisted with data interpretation and analysis, and revised and critically reviewed the manuscript; and all authors approved the final manuscript as submitted.
This research was presented in part at the annual meeting of the Pediatric Academic Societies; April 28–May 1 2012; Boston, MA.
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
FUNDING: No external funding.
- Moon RY,
- Task Force on Sudden Infant Death Syndrome
- ↵American Academy of Family Physicians. Breastfeeding (policy statement). Available at: www.aafp.org/online/en/home/policy/policies/b/breastfeedingpolicy.html. Accessed July 6, 2012
- ↵American College of Obstetricians and Gynecologists Committee Opinion. Breastfeeding: maternal and infant aspects. Available at: www.acog.org/∼/media/Committee%20Opinions/Committee%20on%20Health%20Care%20for%20Underserved%20Women/co361.pdf?dmc=1&ts=20120716T0001274881. Accessed July15, 2012
- ↵World Health Organization Health Topics. Breastfeeding. Available at: www.who.int/topics/breastfeeding. Accessed July 6, 2012
- Philipp BL,
- Merewood A,
- Miller LW,
- et al
- ↵Baby-Friendly USA. The ten steps to successful breastfeeding. Available at: www.babyfriendlyusa.org/eng/10steps.html. Accessed July 7, 2012
- ↵Center for Disease Control and Prevention (CDC). Breastfeeding report card–United States, 2012. Available at: www.cdc.gov/breastfeeding/data/reportcard.htm. Accessed February 12, 2013
- ↵US Surgeon General’s call to action to support breastfeeding 2011. Available at: www.surgeongeneral.gov/library/calls/breastfeeding/calltoactiontosupportreastfeeding.pdf. Accessed July 7, 2012
- ↵Healthy People 2020 breastfeeding objectives. Available at: www.usbreastfeeding.org/LegislationPolicy/FederalPoliciesInitiatives/HealthyPeople2020BreastfeedingObjectives/tabid/120/Default.aspx. Accessed July 8, 2012
- ↵Colorado Department of Public Health and Environment. Getting It Right After Delivery: Five Hospital Practices That Support Breastfeeding. Available at: www.cdphe.state.co.us/ps/mch/gettingitright.pdf. Accessed July 8, 2012
- ↵Jenik AG, Vain NE, Gorestein AN, et al. Does the recommendation to use a pacifier influence the prevalence of breastfeeding? J Pediatr. 2009;155(3):350-354.e1
- Howard CR,
- Howard FM,
- Lanphear B,
- et al
- Jones JR,
- Kogan MD,
- Singh GK,
- Dee DL,
- Grummer-Strawn LM
- ↵Kaiser Family Foundation. Available at: www.statehealthfacts.org/comparemaptable.jsp?ind=501&yr=1&typ=2&rgnhl=39. Accessed July 8, 2012.
- Perrine CG,
- Scanlon KS,
- Li R,
- Odom E,
- Grummer-Strawn LM
- Copyright © 2013 by the American Academy of Pediatrics