Improving Guidance and Maternal Knowledge Retention After Well-Newborn Unit Discharge Free

This project was conducted at a 280-bed community hospital in northeastern Connecticut with 1300 deliveries annually. Our team of 6 academic pediatric hospitalists round on >80% of all newborns in the well-newborn unit, with 2 community pediatrician groups caring for the rest. More than 95% of mothers admitted to the hospital prefer English as their primary language; 67% identify as White, 16% as Hispanic, 12% as Black, and 3% as American Indian. Fifty percent are privately insured, 35% are insured through Medicaid, and 15% insured through TRICARE. Newborn discharge guidance by the hospitalist delivered verbally in person has historically occurred either on the day of discharge or the day before. Nursing education, on the other hand, was completed along the continuum of the newborn’s stay, but the majority of topics were reviewed the night before discharge. On the basis of a preintervention survey of staff, nurses spend anywhere from 10 to 40 minutes on newborn education compared with 5 to 10 minutes for hospitalists. We discovered that certain topics were communicated differently by nurses and hospitalists, such as the definition of fever. Other educational content discrepancies were uncovered for normal voiding and stooling patterns and language use about car seats, safe sleep, and circumcision care. Because there was no formal guidance curriculum, nurses followed the nursing discharge educational tool in our electronic medical record, which included an unstructured, overly long list of suggested topics. Hospitalists, on the other hand, tended to focus on a core set of topics, including safe sleep, car seat safety, fever, feeding, and voiding and stooling patterns. Prioritization of topics for both nurses and hospitalists happened on an individual level and depended on how much time was available between patients. Independent from verbal education, all mothers of newborns were given a booklet on newborn and postpartum care.

To better understand barriers to successful guidance and maternal knowledge retention, we conducted a modified failure mode effects analysis.⁹  Failure modes were developed around the lack of standardization of educational content; teaching responsibilities among team members; prioritization of topics; patient-centered teaching approaches; effective teaching strategies focused on adult learners and learning style preferences; and underuse of available teaching materials, such as the aforementioned booklet. We also identified a lack of understanding of maternal educational needs.

Additionally, we used Kern’s approach to curricular development in medical education, beginning with problem identification and general and targeted needs assessment (steps 1–2).¹⁰  We performed an educational needs assessment during the project baseline period (October 2018 through December 2018) and surveyed over a 3-month period 120 mothers admitted to the well-newborn unit on their learning preferences. The survey was part of our general needs assessment in the early stages of the project to help us to better understand the population we were targeting as well as personal satisfactions and preferences with educational materials provided and content. Surveys were completed at the time of discharge, which the mother either left in her room or handed to the nurse at the time of discharge (see Supplemental Information). The results informed the development of our key driver diagram.

We next developed a phone follow-up quiz (Table 1) for mothers recently discharged from the well-newborn unit to assess maternal knowledge retention of core AAP anticipatory guidance. The quiz was tested by a small group of hospitalists and nurses for immediate feedback and edits. We subsequently trialed the quiz over the phone with 20 mothers, which resulted in additional edits. Two hospitalists conducted all calls and adhered to a script. Withing 2 to 3 days after discharge, mothers were contacted randomly from a list of recent 4-day hospital discharges in our electronic medical record. The hospitalist contacted mothers 1-by-1 on the discharge list, and mothers had to agree to take the time for the quiz to be included. Our aim was to contact 5 to 10 mothers weekly (25%–30% of admissions). Mothers of infants who were discharged from the NICU or did not have custody of their newborn were excluded. Mothers whose preferred language was not English were excluded. Parents did not know that they were being contacted because we selected participants at random. Mothers answered their phones were asked whether they would be willing to take a 3-minute quiz to inform a project aimed at improving education for mothers in the well-newborn unit. Mothers who answered quiz questions incorrectly received immediate feedback on the correct answer. Some mothers also took the phone call with “their nursery pediatrician” as an opportunity to seek advice or ask questions that had come up before their infant’s first newborn visit. During these calls, the conversation was mutually beneficial. No incentive was offered, and 1 to 2 mothers per month chose not to participate.

Pareto analysis⁹  identified high-yield guidance topics that mothers missed during our baseline data collection through these phone follow-up quizzes (Fig 1). These results guided the prioritization of our interventions.

From December 2018 to November 2019, we sought to optimize the newborn discharge educational process and assess how to strengthen maternal retention of newborn anticipatory guidance. We formed a multidisciplinary QI team that comprised hospitalists, nurses, and a lactation consultant and met monthly starting in December 2018 to develop goals and objectives and discuss educational strategies and implementation (Kern’s steps 3–5).¹⁰  Formal interventions started in January 2019. We centered our interventions around 4 key drivers (Fig 2) and implemented a series of interventions aimed at restructuring the educational process.

With a bedside topic checklist, we performed Plan-Do-Study-Act (PDSA) cycles around an educational topic checklist reflecting the recommended core content of the AAP discharge guidance. After feedback from nurses and hospitalists, the wording and layout were modified. We then tested the checklist’s utility on a subset of new mothers. During the next PDSA cycle, more topics were added and organized into 3 guidance sections—newborn care, breastfeeding, and postpartum care—which were developed together with our obstetrics colleagues. The final list of newborn educational topics were bulb syringe use, cord care, car seat safety, safe sleep, temperature taking and fever, voiding and stooling patterns, jaundice, circumcision care, take-5/abusive head trauma prevention, and when to call a physician (Fig 3). During the third PDSA cycle, the list was tested on all admitted mothers over the course of 1 month. At present, every mother receives a checklist, which is kept next to the bedside computer. Nurses check off completed educational topics along the continuum of the mother-infant dyad’s stay. The checklist is a communication tool between nurses and hospitalists to coordinate teaching and break down educational silos. The checklist represents an educational summary at the time of discharge when the mother and nurse sign the form to confirm that all topics were covered. Signatures are intended to add accountability to the educational process. Upon discharge, the list is scanned into the electronic medical record.

For reviewing and reorganizing the teaching content, we created and implemented teaching scripts and disseminated them among nurses and hospitalists. Different PDSA cycles evolved around obtaining feedback. Content was vetted by a lactation specialist and other content experts, such as a urologist who clarified the circumcision care teaching script. Content was incorporated during safety huddles and staff meetings. Laminated versions for reference are now kept at the bedside next to the topic checklist. Scripts have been incorporated into new staff orientation for both nurses and hospitalists.

Car seat safety courses were provided for nursing staff during 2 nursing skills days, with a focus on proper seat positioning, harness slot adjustment, and harness clip positioning. A car seat infant mannequin was placed in a hospital-owned car seat to facilitate demonstration and hands-on teaching with the parent, which is especially useful when parents have not yet brought in their own car seat. We also created a mandatory car seat safety teaching video for nurses and hospitalists to model the ideal educational interaction with parents. Script development went through PDSA cycles after feedback from nurses, hospitalists, and car seat technicians. The video also models a teach-back moment by the parent at the time of discharge. A checkbox for the car seat teach-back was later added to the educational checklist.

For teaching temperature taking, we addressed maternal knowledge gaps with regard to fever recognition and temperature taking through incorporation of a 1-time rectal temperature demonstration before a baby’s bath and through a gift of a digital thermometer. The ideal timing of the rectal temperature demonstration was tested in PDSA cycles. Shared teaching scripts were used to complement the visual teaching.

Congruent educational content within the team was enhanced by script sharing through e-mail, during huddles, through laminated script sheets kept at the bedside, and through orientation materials. Educational refreshers were conducted later in the project phase after the initial rollout, during staff and faculty meetings, and during daily morning safety huddles. The hospitalist signout tool had checkboxes about fever recognition, safe sleep, and car seat safety guidance.

Maternal self-study was emphasized through the introduction of a handbook given to mothers after admission to the postpartum unit. This handbook included the explanation of a phone application that can be downloaded for a more-detailed content review through videos.

The primary measure was the weekly percentage of correct answers to the phone follow-up quiz questions (Kern’s step 6).¹⁰ 

Phone follow-up quiz data were collected real time through a quiz form created in Qualtrics (Qualtrics International Inc, Provo, UT). The data were exported monthly into a spreadsheet (Microsoft Excel) and subsequently plotted on p-charts using statistical process control (SPC) methodology (Cincinnati Children’s Hospital SPC chart template). The mean was shifted when 8 consecutive data points were above the centerline, signaling special cause variation. Control limits were created as 3 SDs above and below the centerline (mean).

The project was deemed under the category of QI and exempt from a formal institutional review board review.

Over the course of the 1-year project, from October 2018 to November 2019, 333 mothers were contacted by phone within 2 to 3 days after discharge from the well-newborn unit. During the baseline period (October 2018 to December 2018), 93 mothers were reached, and retention of discharge guidance material was poor, with an average of 69% of correct answers. Common incorrect answers based on Pareto analysis were related to urination habits, car seat harness clip positioning, and fever recognition (Fig 1). After the formation of our multidisciplinary team in December 2018, the first intervention and PDSA cycle were conducted in January 2019. Special cause was achieved in March 2019 after the first 2 interventions (Fig 4 A). The centerline shifted from the 69% correct phone follow-up answers to 83% correct answers on the SPC chart (p-chart). The final educational script version was adopted in early April 2019. Interventions 3 through 7 resulted in a second shift of the centerline at the beginning of August 2019 to an average of 86% correct answers (Fig 4 A). This average was sustained throughout the rest of the project. The centerline for proper harness clip positioning shifted after interventions 1 and 2 from a baseline of 41% to 67% followed by a second shift to 85% by the end of July 2019 (Fig 4 B). The centerline for rectal temperature checks shifted from a baseline of 59% to 78% after interventions 1 and 2 to 85% by early August 2019 (Fig 4 C). Data point 3/11 was excluded in the centerline calculation as it was below our cutoff of 5 calls per week. Variation seemed to decrease in the last 4 months of the project phase, likely as a result of overall stable performance of the mothers on the knowledge quiz. Incorrect answers, although showing overall improved average percentages, centered around fever recognition, harness clip positioning, and expected number of wet diapers on day of life 3.

By using QI methodology and informed by AAP guidelines and adult learning theory combined with an established framework for curriculum development, we standardized a multidisciplinary approach to newborn discharge education in 1 community hospital well-newborn unit with the primary outcome of improved maternal knowledge retention of newborn discharge guidance. We achieved improvement of maternal knowledge from an average of 69% correct phone follow-up quiz answers to 86% correct answers over 1 year.

To our knowledge, this educational QI initiative is the first aimed at improving AAP policy adherence of recommended well-newborn anticipatory guidance in a community hospital setting. The novel component of this QI initiative was that we applied frameworks of adult learning theory, curricular development in medical education, and a multidisciplinary team approach to maternal education and subsequently achieved a measurable change in knowledge retention in a community hospital within a large health system. Given the importance of successfully helping mothers to transition out of the hospital with their newborn and getting them safely to their first pediatrician appointment,¹  the results of this QI initiative reveal an important model on which to base how mothers receive newborn anticipatory guidance in well-newborn units.

In a recent study, researchers evaluated the educational needs of parents in the well-newborn unit setting through a 4-question survey to assess parents’ preexisting knowledge base. The results of the study revealed that mothers answered basic knowledge questions correctly 88% of the time.⁷  Our quiz not only included the same questions but also explored topics such as car seat safety, safe sleep recommendations, and the most accurate temperature-taking method. Mothers struggled most in these knowledge areas when probed with slightly more challenging questions, which is why our knowledge average was poor at 69% correct answers. In our project, we also noticed that the number of previous children did not affect the average of correct answers, stressing the existence of unmet educational needs and the importance of effective discharge guidance for all mothers and caregivers.

In other studies, investigators suggested that parents may be susceptible to educational information overload, which in turn reduces knowledge retention.^7,8  Similarly, knowledge retention of anticipatory guidance in the outpatient setting has been negatively affected by a larger number of topics discussed.¹¹  Hence, one of our first QI interventions focused on aligning and narrowing the educational content to core topics recommended by the AAP policy “Hospital Stay for Healthy Term Newborn Infants.”¹  We conducted PDSA cycles around an educational discharge checklist. The adoption of its final version was crucial to standardizing and shifting our team’s educational focus. The checklist also created accountability and served as a visible communication tool of each mother’s “educational status.” Compliance overall was very good, although we did not track this process measure in real time.

Effective adult learning is reliant on multimodal teaching strategies.^10,12,13  By incorporating visual and interactive demonstrations in this QI project, particularly rectal temperature taking and hands-on car seat use with a newborn mannequin, we created a more practical and immersive learning environment. Previous studies on parents’ car seat safety misuse, for example, have revealed car seat safety to be a prevalent issue.¹⁴  Numbers are improved if car seat safety technicians are involved,¹⁴  underscoring the positive impact of effective and possibly hands-on teaching.

Our project has several limitations. Given the global approach to restructuring newborn discharge guidance, with several educational topics and interventions addressed concurrently, it is difficult to ascertain the impact of each single intervention on the outcome of specific knowledge areas. Although we could have narrowed our focus, the project’s success reflected the motivation of our multidisciplinary team, which was determined to overhaul the entire educational process, not just individual topics. We did not track health care outcomes such as health care utilization and readmissions after discharge because these outcomes were beyond the scope of this project. Such outcomes will be important to investigate in our continued work and are now more feasible with longer-term data. The exclusion of non-English speaking mothers from knowledge quizzes may have created a selection bias and a risk for potentially exacerbating health disparities. For context, in our community hospital setting, <5% of families are non-English speaking; therefore, we chose to trial different interventions in English first. We plan to address this disparity in our ongoing QI work, using bilingual materials and including Spanish-speaking mothers. In our project, we focused on educating the mother as the presumed main caregiver and for consistency in data collection. The exclusion of fathers or other caregivers is a limitation and an opportunity to address in our future work. We did not explore technology-based media for delivering guidance in the current project. Technology-based educational approaches have been successfully tested in comparison with the more traditional approach to patient education of handouts and verbal instruction.^8,15–17  Bilingual videos could help to address the need for individualized learning and potentially close the knowledge retention gap that remains. We are currently analyzing the impact of several videos that we have created.

With regard to sustainability, the project was paused during the early months of the COVID-19 pandemic. Data collection resumed in November 2020 with the goal of implementing video- and Spanish-based guidance. Average correct answers had dropped to a baseline of 80% as awareness around the project itself had faded and newer staff had onboarded. Guidance at that point had been incorporated into new staff orientations, and the discharge topic checklist was still actively used for all mothers along with laminated guidance scripts kept at the bedside. As a result, our improvement efforts were sustained months beyond the project intervention period and in the midst of the pandemic.

In conclusion, the implementation of a series of educational interventions to standardize newborn discharge guidance in a community hospital resulted in marked and sustained improvement in maternal knowledge retention of newborn care. On the basis of this success, the next evolution of this QI initiative is the creation of a formal newborn discharge guidance curriculum that will be inclusive of non-English speaking mothers and enhanced by bilingual video teaching materials. Videos may also benefit mothers with lower literacy levels. Systemwide spread to other hospital sites in our health system, including the large tertiary academic medical center, will be evidence of the value that community-based pediatric hospitalist sites bring to all levels of care.

The authors thank the Unit Based Practice Council, Educational Subcommittee, at Lawrence and Memorial Hospital in New London, Connecticut, for their support.