Discharge Best Practices of High-Risk Infants From Regional Children’s Hospital NICUs

With a longitudinal goal of optimizing NICU discharge and transition processes by improving efficiency, safety, readiness, and education for medically complex patients, we used techniques from QI methodology including cause and effect diagram (fishbone) and key driver diagrams (KDDs). Using discussion and voice of customer with individually identified local expert stakeholders to evaluate current state, 7 CHNC sites participated in constructing their own fishbone diagram assessing unit-specific discharge vulnerabilities. Considerable overlap of contributory factors to discharge complexities among units became evident. Common themes and vulnerabilities were extracted from the 7 fishbone diagrams and merged into 1 multisite fishbone diagram (Fig 1). The multisite fishbone allowed for shared mental model for system vulnerabilities among participating CHNC sites. These diagrams identify 7 core themes: Education, Process, Documentation, Providers, Nursing, Family and Caregivers, and Patients (Infants).

The development of a multisite fishbone established a robust perspective of clinician and provider’s point of view. Recognizing the importance of family engagement, the clinical team sought to include a parent family advisor group to complement the clinical perspective. CHNC centers each recruited a parent of medically complex NICU graduates. This collaborative was fortunate to have 6 volunteer parents to provide insights into family experiences and needs and help inform future intervention plans.

Next, exploring many vulnerabilities in the complex NICU discharge process, 15 participating centers built a comprehensive KDD. Literature review, clinician experience, and lived experience from our parent advisors identified 21 primary drivers related to successful discharge. The group condensed the 21-component KDD to 8 primary drivers focusing on discharge education, inpatient discharge process, and caregiver engagement (Fig 2). Physician, nurse, and patient clinical categories were considered out of scope for this stage; documentation was bundled with process. On the basis of the barriers identified in the fishbone diagram, drivers, and literature review, we compiled a list of 52 potential best practices without exclusion for successful discharge preparation from regional children’s hospital NICUs (Supplemental Information).

To achieve consensus of the 52 best practice statements among NICU stakeholders, we used a modified Delphi approach. The Delphi survey is designed to capture opinion and perspectives via multidisciplinary vote casting.^15–17  This modified Delphi approach reflects collective center management rather than individual preference or choice of the respondent.

For survey 1, participating CHNC site leaders recruited expert stakeholders for survey participation, (n = 131), spanning the neonatal and outpatient continuum, including parents, neonatal, pediatric, and ancillary providers. We distributed the 52-point survey to participants via REDCap, a secure data system, to vote as content experts and provide feedback for the potential best practices. For each practice, participants indicated their level of agreement with including the practice as a potential best practice recommendation statement on a 5-point Likert scale, with a comment box prompt for answers indicating disagreement. Consensus was defined as ≥85% agreement (tallied as Agree and Strongly Agree on the Likert scale) among respondents. For survey 2, we clarified language for some of the 52 practices on the basis of participant feedback from survey 1. Electronic reminders were sent via RedCap to survey nonrespondents to maximize response rate.

To identify implementation feasibility and understand unit-level priorities, we developed a prioritization and feasibility assessment survey to rank the top 5 best practices. Recognizing that 52 potential best practices may be challenging to implement, we condensed the best practices to 38 when practices were similar or overlapping. For this survey, we invited 1 member from each CHNC site to answer on behalf of their site (n = 15). Participants indicated whether the potential best practice was either already implemented in their NICU, or whether the practice was feasible and/or a priority for their unit. On the basis of our feasibility assessment findings, we initiated the implementation phase for an improvement intervention cycle focused on screening for SDoH, including referral for resources, as indicated. We began with a gap analysis at participating sites and collected data from each site regarding current screening practices.

The focused KDD is shown in Fig 2. For Process, variability and accountability in the multidisciplinary discharge process were identified as primary drivers. Brainstorming potential actions for a future intervention bundle, we focused on discharge task organization, standardization yet customization of processes, and determining early discharge readiness. For Education, variability in education plans, delivery methods, and parental presence and engagement were identified as the biggest challenges. For Parental Engagement, variation in health literacy regarding complex NICU discharge processes and education, variable challenges affording parental engagement, and variation in language barriers, cultural barriers, and racial biases, as primary drivers. We compiled a list of tools, strategies, and technology that could enhance education as possible future interventions.

The primary goal of the Delphi surveys was developing a list of potential best practices for optimal discharge from regional children’s hospital NICUs focusing on process and education. The overall response rate for the 2 modified Delphi surveys was 46%, with responses from NICU parents; neonatologists; bedside nurses; advanced practice providers; case managers; a discharge coordinator; social workers; a psychologist/behavioral health specialist; QI advisors; physical, occupational, and speech language therapists; outpatient providers; and follow-up clinic providers (Table 1). Responses demonstrated a strong level of agreement for both process (33 best practices) and education drivers (19 best practices) between survey 1 (n = 47, 90%) and survey 2 (n = 50, 96%) respondents (Fig 3). After these 2 iterative modified Delphi method surveys, 50 of the 52 statements met the predefined criteria for consensus.

We condensed the potential best practice statements from 52 to 38, because some practices were overlapping or closely aligned. Condensing the potential best practice statements allowed for improved focus on prioritization and feasibility assessments for each individual unit. The prioritization survey of these 38 potential best practice statements received 12 unique responses, including 11 regional children’s hospital NICUs and 1 parent (Fig 4). Among the potential best practices, those already implemented at centers were ranked and are presented in Fig 4. Two were already universally implemented (n = 11 centers), “identifying and including home care providers in addition to parents (eg, grandparents) that need discharge education” and “providing consultation with dieticians and pharmacists for infants with complex formula and medication needs.” The highest-ranked practice as a top 5 priority was “consider assessment of families’ SDoH with a standardized tool” (n = 6).

Nine sites confirmed participation in the first improvement intervention cycle. Four sites currently conduct SDoH screening, albeit not universally within their NICU. Five sites had no current standardized screening with a validated tool. Sites identified team leaders for this cycle. The standardized tool embedded in each site’s local electronic medical record was chosen for screening; sites without this option may use a paper version of a standardized screening tool. Each site selected operational practices that met their respective units’ needs to optimize flexibility and encourage participation. These practices included the role of performing the screening (nursing, social work, case management) and timing of screening (within the first week of admission or the week before discharge).

There are potential unit-specific and multisite-wide barriers to successfully implementing SDoH assessments for discharge optimization. Unit-specific barriers include limited dedicated resources leading project work, the time reserved to promote multidisciplinary member and parent engagement, the time reserved for educating staff regarding the planned tests of change and corresponding documentation, and technology resources for data collection and reporting strategies. Additionally, resources are needed to follow implementation, addressing specific family vulnerabilities identified during the assessment. Potential challenges to successful implementation strategies also occur at a system level. Referring to our multisite national collaborative, examples include variation in local culture in the different geographical NICU settings, different policies and procedures regarding respective units’ discharge processes, and variation in dedicated discharge resources involved in complex NICU discharge planning. Another potential barrier is the risk of bias in data collection relating to situational discharge circumstances associated with local culture, policies, and processes.

Strengths of this work include the engagement of a national and multidisciplinary group, extensive vetting of various perspectives from regional children’s hospital NICUs, and use of a rigorous methodology to gain consensus. Translating high-level guidelines and recommendations to “boots on the ground” actionable steps with designated roles for each team member has been done by forming multiinstitutional QI collaborations.¹⁸  Respondent centers reported being amenable to creating bundle interventions focused on discharge task organization and education, standardization yet customization of these processes, and the identification of early discharge readiness. The CHNC provides an infrastructure to implement these changes in a systematic and methodological way. Limitations to this work include lack of universal involvement of all CHNC sites and potential bias from respondents. Lack of available baseline data regarding discharge processes for multiple CHNC sites, resulting in the use of a modified Delphi approach to reach consensus on potential best practice statements, is another noted limitation. These results reflect highly resourced regional children’s hospital NICUs and may not be generalizable to all NICU populations.

Given that consensus guidelines and recommendations for discharge preparation and transition planning of NICU patients have only recently been published,⁸  regional children’s hospital NICUs have been left to design these programs on the basis of individual center needs and resources leading to significant practice variation. The multidisciplinary CHNC Transitions of Care and Follow-Up Focus Group developed many potential best practices, defined priorities for improvement, and concluded that process and education drivers were the highest priority for complex NICU discharges. We achieved consensus with the modified Delphi survey regarding practices within these drivers and determined that, though many were already implemented within respondent NICUs, variability remained.

Processes that address discharge preparation and transition planning for the NICU population cannot be one size fits all because of the complexity of infants’ medical and family needs. Many NICU graduates are discharged with families facing social, environmental, economic, and mental health challenges that further complicate discharge. To accomplish the enormous task of transitioning families home, multidisciplinary teams that include NICU staff, parents, social workers, mental health providers, and discharge planners and a systematic approach are needed. Many regional children’s hospital NICUs have these providers, but how they function and engage the family before discharge and their accountability for given discharge steps is not always clear and consistent, both within and between centers. Focus group members also identified that variability in discharge processes contributed to delays and complications. Our parent advisory group similarly advocated for a discharge process that was simplified, timely, and provided necessary resources that addressed mental health, financial needs, and parental support at home.

Our group’s work also highlighted the need for better educational tools to prepare families for the transition home.^19–21  The consensus statement by Smith et al provides a battery of recommended educational content to address individual family needs for caring for their infant at home. Development of this educational content and resource guides must be further customized to individual family and patient needs, especially those of medical and surgical complexity. Recognizing variation in adult learning styles and health literacy, the content and delivery of these topics must be carefully curated and presented. Anticipatory guidance and education can provide a realistic idea of life beyond the NICU and facilitate their transition home.

Our findings call for infants to be screened for adverse SDoH during NICU admission. Identifying concerns related to SDoH has important implications in optimizing discharge, especially for infants with medical complexity. Unmet basic needs and stressors may impair a family’s ability to be present in the NICU to receive optimal education regarding their infant’s condition and care required after discharge.²²  Recognizing and addressing families’ challenges, such as housing insecurity and transportation needs, is essential for the care of infants requiring durable medical education and multiple subspeciality follow-up appointments. This echoes the mandate from the American Academy of Pediatrics recommendation of universal screening for adverse SdoH and delivery of referral for community resources for unmet basic needs within pediatric clinical care.²³  Our survey revealed many regional children’s hospital NICUs do not screen routinely, which may be because of lack of NICU-specific screening tools, personnel, unit resources, or systematic guidelines. A recent mixed methods study identified several themes regarding the need and applicability of screening of adverse SDoH in the NICU.²⁴  NICU providers felt addressing families’ unmet needs was central to the care of infants and highlighted that standardizing screening is feasible and beneficial. This is consistent with our survey, with the highest priority score given to the best practice consider assessment of families’ SDoH with a standardized tool.

This study represents a first step toward developing resources and implementing workflows that can be used in NICUs to improve complex discharge preparation and transition planning. Our work highlights the lack of universal screening SDoH in the NICU, a critical step to help understand unmet needs of families that may influence the health and care of their infant once home. Specific focus on equity, inclusion, and antiracial biases screening practices will be key to reducing risk of discrimination and improving equity of care. Having a systematic, multicenter approach to implement quality discharge initiatives is the natural next step to demonstrate necessary change. Better support of families navigating the complex NICU discharge process has the potential to improve infant health outcomes.

We thank the parental advisory group, including Emily Bogarin, Brendan Etue, Monia Jones, Krystal Kendall, Courtney and Dwamelle McGill, Trisha O’Shea, and Brittany Whittman. Beverly Brozanski, Jacquelyn Evans, Theresa Grover, Karna Murthy, Michael Padula, Eugenia Pallotto, Anthony Piazza, Kristina Reber, and Billie Short (ex-officio David Durand, Francine Dykes, and Jeanette Asselin) are executive members of the Children’s Hospitals Neonatal Consortium, Inc, who developed and manage the CHND (thechnc.org). For more information, please contact [email protected]. We also thank the following CHNC participating institutions that serve the infants and their families; these institutions (and their site sponsors) also have invested in and continue to participate in the CHND: Alfred I. duPont Hospital for Children, Wilmington, DE (Kevin Sullivan); All Children’s Hospital Johns Hopkins Medicine, St Petersburg, FL (Victor McKay); American Family Hospital, Madison, WI (Jamie Limjoco, Lori Haack); Advent Health for Children, Orlando, FL (Narenda Dereddy, Raj Wadhawan); Ann & Robert H. Lurie Children’s Hospital of Chicago, Chicago, IL (Karna Murthy, Gustave Falciglia); Arkansas Children’s Hospital, Little Rock, AR (Becky Rogers); Boston Children’s Hospital, Boston, MA (Anne Hansen); Brenner Children’s Hospital, Winston-Salem, NC (Cherrie Welch); Cincinnati Children’s Hospital, Cincinnati, OH (Beth Haberman); Children’s Healthcare of Atlanta at Egleston, Atlanta, GA (Anthony Piazza); Children’s Healthcare of Atlanta at Scottish Rite, Atlanta, GA (Gregory Sysyn); Children’s Hospital and Medical Center, Omaha, NE (Nicole Birge); Children’s Hospital Colorado, Denver, CO (Theresa Grover, Danielle Smith, Stephanie Bourque); Children’s Hospital of Orange County, Orange, CA (Michel Mikhael, Irfan Ahmad); Children’s Hospital of Philadelphia, Philadelphia, PA (Michael Padula, David Munson); Children’s Hospital of Pittsburgh of UPMC, Pittsburgh, PA (Toby Yanowitz); Children’s Hospital of Wisconsin, Milwaukee, WI (Michael Uhing, Ankur Datta); Children’s Medical Center, Dallas, TX (Rashmin Savani, Luc Brion); Children’s Mercy Hospitals and Clinics, Kansas City, MO (Julie Weiner); Children’s National Medical Center, Washington, DC (Billie Short, Lamia Soghier); Children’s of Alabama, Birmingham, AL (Carl Coghill, Allison Black); Children’s Hospital of Los Angeles, Los Angeles, CA (Steven Chin, Rachel Chapman); Children’s Hospital of Michigan, Detroit, MI (Girija Natarajan); Connecticut Children’s Medical Center, Hartford, CT (AnneMarie Golioto); Cook Children’s Medical Center, Fort Worth, TX (Jonathan Nedrelow, Annie Chi, Yvette Johnson); Hospital for Sick Children, Toronto, Ontario, Canada (Kyong-Soon Lee); Le Bonheur Children’s Hospital, Memphis, TN (Mark Weems); Nationwide Children’s Hospital, Columbus, OH (Kris Reber); Nemours Children’s Hospital, Orlando, FL (Aaron Weiss); Oklahoma Children’s Hospital, Oklahoma City, OK (Trent Tibble); Primary Children’s Hospital, Salt Lake City, UT (Con Yee Ling, Shrena Patel); Rady Children’s Hospital, San Diego, CA (Mark Speziale, Brian Lane, Laurel Moyer); Riley Children’s, Indianapolis, IN (William Engle, Lora Simpson, Gregory Sokol); Seattle Children’s Hospital, Seattle, WA (Elizabeth Jacobsen-Misbe, Robert DiGeronimo); Stead Family Children’s Hospital, Iowa City, IA (Julie Lindower); St Christopher’s Hospital for Children, Philadelphia, PA (Vilmaris Quinones); St Louis Children’s Hospital, St Louis, MO (Beverly Brozanski, Rakesh Rao); Texas Children’s Hospital, Houston, Houston, TX (Gautham Suresh, Lakshmi Khatakam); and UCSF Benioff Children’s Hospital Oakland, Oakland, CA (Art D’Harlingue, Priscilla Joe).