BACKGROUND AND OBJECTIVE: Breastfeeding has many well-established health benefits for infants and mothers. There is greater risk reduction in health outcomes with exclusive breastfeeding (EBF). Our urban academic facility has had long-standing low EBF rates, serving a population with breastfeeding disparities. We sought to improve EBF rates through a Learning Collaborative model by participating in the Best Fed Beginnings project.
METHODS: Formal improvement science methods were used, including the development of a key driver diagram and plan–do–study–act cycles. Improvement activities followed the Ten Steps to Successful Breastfeeding.
RESULTS: We demonstrated significant improvement in the median adherence to 2 process measures, rooming in and skin-to-skin after delivery. Subsequently, the proportion of infants exclusively breastfed at hospital discharge in our facility increased from 37% to 59%. We demonstrated an increase in sustained breastfeeding in a subset of patients at a postpartum follow-up visit. These improvements led to Baby-Friendly designation at our facility.
CONCLUSIONS: This quality improvement initiative resulted in a higher number of infants exclusively breastfed in our patient population at “high risk not to breastfeed.” Other hospitals can use these described methods and techniques to improve their EBF rates.
- AAP —
- American Academy of Pediatrics
- BFB —
- Best Fed Beginnings
- BFUSA —
- Baby-Friendly USA
- CWH —
- Center for Women’s Health
- EBF —
- exclusive breastfeeding
- EMR —
- electronic medical record
- STS —
- UCMC —
- University of Cincinnati Medical Center
Breast milk is the optimal source of nutrition for newborns, conferring many health benefits to mothers and infants. Breastfed infants have a lower risk of infant mortality and childhood illnesses, including respiratory and gastrointestinal infections, otitis media, and childhood leukemias.1–4 Breastfed infants are also less likely to develop diabetes and obesity.1,5 Maternal benefits include lower risk of postpartum hemorrhage, breast and ovarian cancers, and type 2 diabetes.1,6
Despite these benefits, national breastfeeding rates are below the goals set by Healthy People 2020.7 Ohio’s exclusive breastfeeding (EBF) rates are in the lowest quartile nationally.8 In addition, racial and socioeconomic disparities exist.9,10 Antenatal education and hospital practices that support breastfeeding significantly affect breastfeeding success, exclusivity, and duration, regardless of socioeconomic status.11 In addition, the number of practices a mother experiences is associated with improved breastfeeding duration and exclusivity.12
University of Cincinnati Medical Center (UCMC) has long-standing low rates of EBF, and in June 2012 we began participation in Best Fed Beginnings (BFB), a quality improvement collaborative targeting facilities serving populations at highest risk not to breastfeed. BFB was led by the National Institute for Children’s Health Quality in partnership with the Centers for Disease Control and Prevention and Baby-Friendly USA (BFUSA). Through BFB, we received coaching, technical assistance, and guidance from breastfeeding and quality improvement experts and access to shared knowledge.
Our objective was to incorporate evidence-based hospital practices to improve EBF rates to 90% by August 2014 and to achieve Baby-Friendly designation by September 2014. We used the Learning Collaborative model where representatives from similar health care organizations share their experiences while individually implementing best practices.13 Group learning sessions alternate with local action cycles. Learning sessions are led by national content experts. We chose this model because of its success in other national health care collaboratives, such as improving end-of-life care, decreasing appointment wait times, and reducing adverse events. Here, we share our improvement methods, experiences, and outcomes to assist hospitals intending to make similar improvements.
UCMC is a large, urban academic medical center that trains obstetric and pediatric residents and serves southwestern Ohio, northern Kentucky, and southeastern Indiana. Fifty-five percent of women delivering at UCMC receive prenatal care at the Center for Women’s Health (CWH), a hospital-based resident and midlevel provider practice; the remainder receive care at community health centers, local health departments, and the academic physicians’ office, all staffed by UCMC faculty. In fiscal year 2011 (the year of application to BFB), there were 2352 deliveries; 30% of patients were white, 51% black, and 7% Hispanic; 82% had Medicaid coverage, and 15% were privately insured. Because of these characteristics, our interventions needed to engage trainees, faculty, staff, and hospital leadership. Our hospital’s 7-year plan included an initiative to pursue national certification and recognition that coincided with Ohio’s campaign to lower the infant mortality rate.
In July 2012, we assembled a multidisciplinary team that included an administrative leader, obstetrics and pediatrics physicians, a mother–infant nurse manager, labor and delivery and postpartum staff nurses, and lactation consultants. The team evolved to include a Women, Infants, and Children administrator and peer.14 The team developed a key driver diagram (Fig 1) and identified improvement activities and plan–do–study–act cycles to meet the aims.15 The UCMC Office of Research Compliance determined that our project was a quality improvement initiative and not human subjects research.
The team attended 3 BFB Collaborative Learning sessions, participated in regular webinars, reported monthly data, implemented and tested changes, and shared resources with other hospital teams. The desired BFB outcome was Baby-Friendly designation of all participating hospitals by September 2014. BFB surveyors conducted a mock site visit to evaluate readiness for a BFUSA assessment. They reviewed infant feeding policies, compliance with BFUSA guidelines, and interviewed providers and patients. Visit findings provided opportunities to correct deficiencies before the actual BFUSA assessment.
Our improvement activities followed the Ten Steps to Successful Breastfeeding,16 endorsed by the American Academy of Pediatrics (AAP), the World Health Organization, and other health care organizations. A dose-dependent relationship exists between breastfeeding duration, exclusivity, and the number of these steps a mother experiences during the delivery hospitalization.12,17
Step 1: Have a written breastfeeding policy that is routinely communicated to all health care staff.
Policy written and implemented (online Supplemental file).
Step 2: Train all health care staff in the skills necessary to implement this policy.
Nursing education (for which staff was compensated): online breastfeeding modules (15 hours), skill laboratory (4 hours), breastfeeding policy in-service (1 hour).
Pediatric providers: 2 hour lectures for residents and fellows, AAP slide sets distributed with posttest18 (3 hours).
Obstetric providers: American College of Obstetrics and Gynecology publications19,20 and breastfeeding videos (taped neonatal grand rounds and Stanford University breastfeeding/hand expression)21 with posttest (4 hours).
New employees complete training within 6 months of hire.
Step 3: Inform all pregnant women about the benefits and management of breastfeeding.
CWH staff education: Lactation consultant lecture (3 hours).
Community (non-CWH) nurse case managers: online breastfeeding modules (20 hours), training by lactation consultant (3 hours).
Policy written and implemented outlining prenatal breastfeeding education topics.
Prenatal education checklist generated in electronic medical record (EMR) to facilitate efficient documentation.
Pocket scripts developed for breastfeeding education reflecting policy (Supplemental Information A).
Step 4: Help mothers initiate breastfeeding within 1 hour of birth.
Skin-to-Skin (STS) Task Force created.
Infant nurse role developed to facilitate STS for all medically appropriate deliveries regardless of feeding plan.
Step 5: Show mothers how to breastfeed and maintain lactation, even if they are separated from their infants.
Initiate timely pumping within 6 hours of delivery if couplet separated. Extra pumps and hands-free pumping bras obtained.
Step 6: Give infants no food or drink other than breast milk, unless medically indicated.
Step 7: Practice rooming-in: allow mothers and infants to remain together 24 hours a day.
Nursery was staffed only for procedures and was subsequently named the Newborn Observation Unit to minimize separation, optimize family recognition of feeding cues, and prevent delayed feedings.
Computer workstations on wheels obtained for family-centered rounds.
Infant location documented in EMR.
Step 8: Encourage breastfeeding on demand.
Infant-led feedings encouraged for breastfed and formula-fed infants.
Nursing personnel, including labor and delivery staff, trained in latching techniques and positions.
Step 9: Give no pacifiers or artificial nipples to breastfeeding infants.
Pacifiers eliminated from the postpartum unit, except for medical indications. Scripts developed to respond to maternal requests for pacifiers; smart phrases developed in the EMR for documentation of counseling (Supplemental Information C).
Alternative feeding methods introduced (syringe, cup, and supplemental nursing system).
Case managers counseled patients in prenatal visits regarding risk of early pacifier use.
Step 10: Foster the establishment of breastfeeding support groups and refer mothers to them on discharge from the hospital or birth center.
A Women, Infants, and Children–funded peer provided part-time counseling.
Outpatient lactation visits promoted.
Telephone support system (warm-line) provided to patients upon discharge.
Hospital-based breastfeeding support group developed.
There were 4181 total deliveries during the study period. The study population included all infants cared for on the postpartum unit at UCMC from July 2012 through December 2014. Infants were excluded for NICU transfer or maternal contraindication to breastfeeding (Supplemental Information D).
The primary outcome measure was the percentage of infants exclusively fed breastmilk. Infants were excluded from this measure if the mother exclusively formula fed from birth, after being informed of the benefits of breastfeeding. If supplements were medically indicated, breastfeeding was considered exclusive. A secondary outcome measure was the percentage of mothers reporting breastfeeding at a CWH postpartum visit.
We identified 2 process measures. Rooming-in was the percentage of infants separated from their mothers for ≤1 hour per day; STS was the percentage of infants placed STS with their mothers within 5 minutes of birth, until the first breastfeed or 1 hour of life.
From July 2011 to June 2012, we reviewed 30 charts per month to establish a baseline EBF rate for the year preceding BFB. From July 2012 to October 2012 we selected every fourth medical record from monthly delivery reports (n = 50). Starting in November 2012, 30 infant charts were reviewed, per BFB project guidelines.22 We extracted the following data from the infant hospital charts: breastfeeding assistance, exclusivity, supplementation, rooming-in, feeding on cue education, and discharge lactation support referral. Maternal charts were reviewed to assess breastfeeding intent, prenatal breastfeeding education, and STS compliance at delivery. We performed an EMR query for sustained lactation success. This included charts of CWH patients seen postpartum with a documented feeding plan from November 2012 to December 2014. The data collection periods for inpatient and outpatient were different because of the 6-week postpartum period and coincidental implementation of an EMR system at our hospital.
Statistical process control charts and run charts evaluated the impact of interventions on the outcomes over time. These tools can evaluate the effectiveness of change over time and distinguish common cause variation (causes that are random in the process over time) from special cause variation (causes that are not part of the process but arise due to the process).23 Observed changes in the mean and median line are considered statistically significant when the line shifts on a chart by 8 points in either direction, 6 consecutive points either increase or decrease, or 14 consecutive points alternate above or below the line.
Our initial efforts focused on 2 key hospital practices changes: STS and rooming-in. Before BFB, STS after delivery was rarely practiced. The STS Task Force, made up of key stakeholders, established weekly meetings to discuss small tests of change. The STS Task Force acknowledged challenges voiced by delivery personnel and adapted practices to satisfy STS goals and provider concerns. Although improvement in STS was achieved quickly with vaginal deliveries, we observed slower progress in cesarean deliveries. Mock drills demonstrated the complex documentation, monitoring, and supportive role of the circulator nurse at cesarean delivery; a dedicated infant nurse allowed for STS in the operating room without compromising patient safety. Since STS was instituted for all infants, staff and families shared their observations of innate newborn behaviors, and consequently several patients changed their feeding plans from formula to breastfeeding. This unexpected consequence, along with data from plan–do–study–act cycles, increased acceptance of STS benefits. With these measures implemented by 8 months, the median STS rate was 24.5%; by 15 months, STS increased to 59% for all modes of delivery (Fig 2).
Before project initiation, infants rarely roomed in with their mothers, and causes and duration of couplet separation were not documented in the EMR. During early discussions, provider-cited reasons for separation included newborn teaching rounds, newborn assessments, circumcision, and the perception that mothers preferred uninterrupted rest. Additionally, providers referenced the nursery as a place to care for multiple infants whose mothers lacked family support. A pediatrician tested newborn teaching rounds in patient rooms and reported back to the physician group to address concerns. We obtained portable equipment to facilitate performance of newborn assessments. Our obstetrician champion engaged the postpartum team to adopt a patient-centered rather than provider-centered approach to circumcisions. Specifically, this approach included individualized timing rather than provider-convenient, designated times for circumcisions. Nurses performed bedside postcircumcision checks rather than using the Newborn Observation Unit. Acknowledging that hospital staff frequently interrupt mothers’ sleep, we instituted daily quiet times to promote maternal rest. We explained that rooming-in empowers mothers to care for their infants and encourages maternal recognition of feeding cues, without affecting maternal sleep.24–26 Through these initiatives, we improved rooming-in to a median rate of 70% at 7 months and 98% at 15 months (Fig 3).
In the year preceding BFB, the baseline rate of EBF was 37%, and within 6 months of the project, EBF increased to 59% (Fig 4). After observing an unexplained deviation in March 2014, we subsequently audited all newborn charts to ensure validity and eliminate sampling error. We found a small but significant improvement in our secondary outcome of sustained breastfeeding in women with prenatal care at CWH. The median number of women reporting breastfeeding at an outpatient visit rose from 42% to 50% in 18 months (Fig 5).
Before our initiative, the cultural norm did not include EBF, rooming-in, or STS after delivery. Through participation in the BFB Learning Collaborative model, we successfully implemented evidence-based hospital practices and increased breastfeeding rates measured at hospital discharge and a postpartum visit. These changes provided the foundation for Baby-Friendly designation at our urban academic health center in December 2014.
By June 2015, only 31 of 89 Learning Collaborative hospitals achieved Baby-Friendly designation.22 We believe our organizational leadership and concurrent, mutually beneficial initiatives were key factors in our success. We strategically associated this department quality improvement project with the hospital mission to provide life-changing, patient-centered care. Our participation in BFB aligned our practices with the Joint Commission Perinatal Core Measures on EBF,27 a publicly reported accountability measure. The timing of this project coincided with public health efforts to lower the infant mortality rate in Hamilton County and Ohio, which our leadership champion used to solidify the hospital’s financial commitment and justify the increased budget for maternity and neonatal services. Practically, this required the Purchasing Department to negotiate a fair market price for formula, purchase online modules, and cover payroll costs for nursing education time. The obstetrics and pediatrics departments absorbed physician training expenses.
Before BFB, many individuals supported breastfeeding mothers but functioned in silos. We encountered numerous challenges in our efforts to implement practice changes. The didactic and clinical skill sessions across multiple units involved considerable time and labor. Maintaining documentation compliance required unremitting vigilance and accountability. No additional personnel were hired for these activities, so workloads shifted to maintain clinical coverage while accommodating the extended duties of data collection and education required for the project. The intrinsic motivation of many individuals was a critical, but difficult to measure, key to our success.
For salaried members of the STS Task Force and staff, including pediatric and obstetrics residents, no additional time was allotted for meetings and training. We believe critical buy-in was gained from these groups, given the public health benefits of breastfeeding and the evolving demographics of resident physicians. Six obstetric residents became parents during our project. Nationally, more residents become parents during residency than ever before,28 and similar personal motivation is probably shared by other residency training hospitals. A residency training hospital supportive of breastfeeding for its patients indirectly supports the breastfeeding of resident trainees and their future patients.
It is worth noting the increase in our EBF rates was significant but not as large as improvements we experienced with STS and rooming in, and we did not meet our goal of 90%. Our experience mirrored that of other collaborative hospitals, because the average increase in EBF rates across all BFB facilities was only 27%.22 We found that families continue to request non–medically indicated formula despite our counseling, and many have the prenatal intention of both breastfeeding and formula feeding. We suspect there are broader cultural and generational biases to be addressed well before pregnancy, delivery, and even conception.
We must acknowledge certain limitations to our results. The data sampling measures may not accurately reflect the EBF rate. Our outpatient data suggest that sustained breastfeeding increased, as other authors in the literature have described,29 although this outcome was measured only in a subgroup of patients receiving prenatal care in our clinic. In addition, this measure is based on maternal report and does not describe exclusivity. Although exact timing of the postpartum visit was not available, most visits occurred within 6 weeks. Because of practical constraints, the baseline for this outcome was generated with EMR implementation rather than project initiation. It is possible the baseline could be lower, resulting in greater improvement; given the preexisting culture, we do not suspect the baseline would be higher.
Some barriers we faced implementing the Ten Steps16 were unique to urban or academic health facilities. Our demographics and hospital dynamics may not generalize to all facilities, especially those already serving populations with high breastfeeding rates.30 In fact, our own evolving demographics may provide false reassurance of sustainability. By 2014, our deliveries increased by 7.8%, which included a greater percentage of privately insured patients. We cannot rule out the possibility that these changes factored into the improvements we experienced. Despite these demographic changes, added complexity remains in caring for underserved populations in the context of an academic medical center with many faculty and resident physicians transitioning monthly.
Another noteworthy challenge has been reconciling AAP recommendations to consider pacifier use to prevent unexplained infant death31 while promoting step 9. Although the AAP recommends delaying pacifier use until 1 month for breastfed infants,31 we have found that this recommendation might represent a mixed message with regional infant mortality reduction efforts for parents, families, nurses, physicians, and public health officials. Nevertheless, we are dedicated to collaborating with all stakeholders, and we strive for a unified approach as we move forward.
Baby-Friendly designation marks not the end of a journey but rather an ongoing pathway for which sustainability has distinct barriers. Overall, we are gratified by the changed attitudes and diminished biases among our staff, physicians, and families. This project fostered standardized communications with patients and families and collaborative interaction between providers. As an added benefit, we found the educational processes required for Baby-Friendly designation improved staff retention, helped recruit nurses, and promoted resident physician well-care.
This initiative represents a multitiered approach to improve the health of our community and to address existing disparities. Although challenging, Baby-Friendly designation is achievable, and it sets the standard for breastfeeding support and mother–infant care. It is particularly important for academic centers training future health care providers to adopt best maternity practices to improve health outcomes for mothers and infants.
- Accepted July 25, 2016.
- Address correspondence to Laura P. Ward, MD, IBCLC, Cincinnati Children’s Hospital Medical Center, Division of Neonatology, 3333 Burnet Ave, MLC 7009, Cincinnati, OH 45229. E-mail:
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
FUNDING: Best Fed Beginnings was supported by the Centers for Disease Control through the National Institute for Children’s Health Quality.
POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.
- US Department of Health and Human Services
- Center for Disease Control and Prevention
- Institute for Healthcare Improvement
- About WIC Breastfeeding Peer Counseling
- Langley GI,
- Moen RD,
- Nolan KM,
- Nolan TW,
- Norman CL,
- Provost LP
- Baby-Friendly USA
- Perrine CG,
- Scanlon KS,
- Li R,
- Odom E,
- Grummer-Strawn LM
- ↵American Academy of Pediatrics Breastfeeding Residency Curriculum Tools. Accessible at: https://www2.aap.org/breastfeeding/curriculum/tools.html. Accessed June 4, 2016
- American Congress of Obstetricians and Gynecologists
- Committees on Healthcare for Underserved Women and Obstetric Practice
- Stanford Medicine Newborn Nursery at LPCH
- National Institute for Children’s Health Quality
- Benneyan JC,
- Lloyd RC,
- Plsek PE
- Ball HL,
- Ward-Platt MP,
- Heslop E,
- Leech SJ,
- Brown KA
- The Joint Commission
- Hughes V
- Brodribb W,
- Kruske S,
- Miller YD
- Moon RY; Task Force on Sudden Infant Death Syndrome
- Copyright © 2017 by the American Academy of Pediatrics