Examining Barriers and Facilitators of Pediatric Enhanced Recovery Protocol Implementation

Background: Enhanced Recovery Protocols (ERPs) are an evidence-based intervention to optimize post-surgical recovery. Several studies have demonstrated that use of an ERP for gastrointestinal surgery results in decreased length of stay, shortened time to regular diet, and fewer administered opioids, while also trending toward lower complication and 30-day readmission rates. Yet, implementation of ERPs in pediatric surgery is lagging compared to adult surgery. The purpose of this study was to conduct a theory-guided evaluation of barriers and facilitators to ERP implementation at US hospitals with a pediatric surgery service. Methods: We conducted semi-structured interviews at 18 hospitals with 48 participants, including pediatric surgeons, anesthesiologists, gastroenterologists, nurses, and physician assistants. Interviews were conducted online, audio-recorded, and transcribed verbatim. To identify barriers and facilitators to ERP implementation, we conducted a thematic analysis using combined inductive as well as deductive logics based on the National Implementation Research Network’s Five Active Implementation Frameworks (AIFs). Results: Framed by the 5AIFs, factors serving as barriers and facilitators to ERP implementation included: delity to and operationalization of ERP elements and measurement of outcomes (usable innovations); establishing and standardizing ERP protocols (stages); organizational support, institutional and electronic health record capacity to track patients getting ERP, training on protocol implementation, education on benets of ERP use and how to adhere to ERPs (implementation drivers); and team, patient, and family buy-in and engagement early in the ERP implementation process (teams). Applying the implementation framework facilitated the identication of potential strategies to support effective practices (evidence-based ERPs and implementation tools), effective implementation (local team infrastructure support including team member roles and responsibilities; a learning collaborative for pediatric surgery groups with monthly support, coaching and training by topic experts, tools to engage key stakeholders), and enabling contexts (a toolkit to support implementation, sustainability assessments, quarterly data-driven feedback sessions, and access to a QI expert on the implementation teams), to achieve the desired outcomes. Conclusions: Comprehensive and systematic application of the 5AIFs allowed organization of the identied barriers and facilitators. Future

Guided by the Active Implementation Frameworks, our research indicates there are modi able barriers to ERP implementation in pediatric settings including: need for pediatric and family engagement, challenges based on surgical volume, and limited site-speci c evidence of ERP effectiveness to drive stakeholders buy-in.
These results provide insight into how to optimize ERP implementation for pediatric surgery.

Background
Adult surgical disciplines have widely adopted Enhanced Recovery Protocols (ERPs), evidence-based interventions that have been shown to reduce patient hospital length of stay, surgical complications, and costs, and optimize patient post-surgical recovery. [1][2][3][4][5] Although there is some evidence of use of ERPs in pediatric surgery, most reports are at a single institution or for a speci c procedure. 6-9 Implementation of ERPs, more broadly, in pediatric surgery is lagging, even for ERP elements that have proven bene ts. For example, despite the American Society of Anesthesiologists (ASA) recommendation for a clear liquid diet, followed by a 2-hour preoperative fasting window, many hospitals still routinely counsel patients to have nothing to eat or drink after midnight of the day preceding a surgical procedure. 10,11 A single center study of an ERP for pediatric gastrointestinal (GI) surgery patients showed a steady increase in the number of ERP elements being used, over time, with a simultaneous decrease in length of stay, in median time to resumption of a regular diet, in median doses of intraoperative and postoperative opioids, and a trend toward lower complication and 30-day readmission rates. 8 Yet, in a recent study to evaluate the adoption of 21 ERP elements in GI pediatric surgery at 18 pediatric hospitals across the United States (US), on average only 6 of 21 elements were widely adopted. 12 Implementation of a comprehensive ERP can reasonably be expected to be challenging, as ERPs consist of a large number of elements, spanning the pre-, intra-, and post-operative process, requiring practice changes at the hospital-, department-and practice-level, and by a multitude of clinicians. Previous studies about ERP implementation have evaluated implementation barriers, processes, and outcomes in adult surgical settings. [13][14][15] To date, however, no comprehensive study about ERP implementation has been conducted in pediatric surgery. Such evaluations are important because of the distinct characteristics of the pediatric surgical environment compared to the adult environment. The long-term goal is to achieve similar outcomes and bene ts from implementation of ERPs in a pediatric population. To achieve this goal, it is necessary to understand the nuances of implementation in a new setting, and in this case, a pediatric setting. 16 The complexity of ERPs and the lag in uptake of its elements highlight the importance of using a systematic implementation framework to change practice comprehensively and proactively.
We conducted a formative qualitative assessment of barriers and facilitators to ERP implementation in 18 pediatric hospitals with a pediatric surgery service for pediatric patients undergoing GI surgery. We chose to use the National Implementation Research Five Network's Active Implementation Frameworks (5 AIFs), 17 to identify which strategies can be designed to optimize ERP implementation in pediatric surgery.
The rationale for selecting the 5AIFs is based on the Active Implementation Formula, 18 which states that to achieve equitable outcomes, three key factors, effective practices, effective implementation, and enabling contexts are needed. 17 The 5 AIFs have been used to facilitate implementation of programs to improve child well-being in various settings, 19,20 but have not yet been rigorously applied in pediatric surgery. By identifying barriers and facilitators within each of the 5 Active implementation frameworks (5 AIF), we aim to inform and optimize the planned implementation of ERPs in a multi-center trial to implement an ERP for pediatric GI surgery.

Sample
This study includes a purposive sample of pediatric surgery teams with a surgeon championing ERPs. We conducted online, web-based semi-structured interviews with clinicians and staff involved, or likely to be involved, in the use of an ERP at 18 diverse hospitals with a pediatric surgery service across the US. The hospitals were selected based on their planned participation in a future multi-center implementation trial to improve ERP implemenation. 21 The interviews were conducted as part of a mixed-methods baseline assessment of ERP use at the hospitals. Hospitals were in urban and rural settings and included pediatric surgery services within freestanding children's hospitals and nested within adult hospitals. The study protocol was evaluated by the Institutional Review Boards at Ann & Robert H. Lurie Children's Hospital of Chicago, Children's Healthcare of Atlanta, Emory University, and Kaiser Permanente Georgia and considered exempt from full review.

Data Collection
A semi-structured interview guide ( Figure 1) was developed, based on the results of a previously administered survey at the 18 hospitals about their use of pre-, intra-, and post-operative ERP elements. 22 The interview guide included questions about hospital and clinician characteristics, processes for GI surgeries, perceived barriers and facilitators to ERP implementation, and successful strategies and recommendations for future ERP implementation. ERP adherence and measurement practices from the prior survey 22 were compiled into a hospital-speci c summary and used to initiate the interview and discuss use of the individual ERP elements in the pre, intra, and post-operative stages. The guide also addressed barriers and facilitators to implementing each element and sought recommendations for further enhancement of ERPs. The interview guide was pilot tested and re ned with one pediatric surgeon with experience in ERP implementation.
We conducted semi-structured interviews between October and December 2019 with clinicians and staff from each of the 18 hospitals participating in the collaborative study. Each interview lasted approximately 60 minutes. Interviews were conducted with groups with at least two interviewees participating per hospital. All participants were consented verbally and the interviews were audio-recorded and transcribed, verbatim.

Theoretical Framework
The National Implementation Research Network identi ed 5AIFs focusing on best practices for implementation and includes suggested mechanisms and strategies for the three key factors to achieve the desired outcomes. 17,23−25 In the Active Implementation Formula 18 (Figure 2), successful implementation is predicated upon having effective practices, effective implementation, and enabling context to achieve desired outcomes. To achieve effective practices, the Usable Innovations framework outlines criteria to ensure that the innovation is usable. To achieve effective implementation, the Stages framework emphasizes having well-operationalized innovations that are teachable, learnable, doable, and readily assessable in practice. Further, the implementation Drivers framework includes common features for successful implementation. Finally, the enabling context is determined by the role of implementation Teams and Improvement Cycles. 17 We used these 5 frameworks to guide our analyses. Table 1 describes the 5AIFs and provides de nitions, how each framework was operationalized and applied to ERP implementation. hospitals. Table 1 summarizes the codebook de nitions of the components within each framework, as applied to ERPs in pediatric surgery. The remaining interviews were concurrently coded in dyads to identify the AIF codes and the barriers and facilitators. The data were then organized to examine barriers and facilitators for each AIF. Two research team members (TD and WS) discussed each framework and the associated barriers and facilitators to con rm inductive coding consistency and reconciled any differences. 28 MaxQDA 2020 (VERBI Software, 2019) 29 was used to support data storage, coding, and analysis.

Results
We conducted 18 semi-structured interviews with 48 participants including surgeons (n=24), anesthesiologists (n=10), a gastroenterologist (n=1), nurse practitioners (n=5), research specialists (6), a physician assistant (n=1), and a clinical pharmacist (n=1). Table 2 provides quotations of each identi ed barrier and facilitator, categorized by the 5AIFs. Some additional exemplary quotes are provided in-text. The research team identi ed barriers and facilitators for each of the 5 frameworks, except for the "Improvement Cycle" framework, which was expected, as the hospitals were all in the pre-implementation phase that precedes improvement cycles. We characterized the barriers and facilitators to implementation of pediatric ERPs into three key factors: effective practices, effective implementation (implementation stages and implementation drivers), and enabling contexts.

Effective Practices: Usable Innovations Framework
The usable innovations framework emphasizes making practices effective by making them teachable, learnable, doable, and readily assessed in practice. 17 Participants reported a lack of training on ERP elements among team members as a barrier to operationalization of ERPs. Reported facilitators to ERP implementation included having clear de nitions of ERP elements and early inclusion of patients and families in the implementation process.
Measuring delity of ERP implementation was identi ed as a consistent barrier for hospitals. Indeed, no hospital reported the existence of a system to track ERP implementation or associated clinical and patient outcome measures. Participants discussed only having an informal process for monitoring outcomes, "... I haven't been really good at making sure preoperatively [that] they've been getting or following their preoperative orders. But what I have been doing, if I do know if they're scheduled surgeries, I try to do kind of like a social round after their surgery, and kind of see how things are going, making sure they're getting the instructions from our wound team, and then our dieticians have this ostomy eating plan that they can take home, and kind of refer to as needed." [MD_11_1] The absence of any monitoring system was attributed to a lack of resources and institutional support. Participants at hospitals with low surgical volume discussed their "insu cient power" to conduct outcome analyses with limited data. One participant noted that ensuring that the team understood their role in ERP implementation was an operational barrier, stating, "It's just operationalizing it in a way everybody understands their role and how to do this." [MD_10_2] None of the reported barriers could be characterized as part of the function framework component.
Rather, functional resources were identi ed as facilitators, including the ability to leverage existing processes from other departments, such as adult surgery, or from hospitals with higher volumes that use ERPs.
"I think it's identifying those things so that you're not recreating the wheel, and it's resources that the nurses are used to using." [MD_11_1] Participants reported that a key philosophy to facilitating ERP implementation is the belief that ERPs lead to improved patient outcomes. Participants emphasized that using available evidence to inform end-user perception of ERPs' clinical bene t and establishing buy-in was essential to facilitating ERP implementation. Participants further discussed the importance of establishing goals for ERP implementation, including alignment with an institution's desire to improve quality of care, such as decreased length of stay, minimizing opioid use, avoiding postoperative ileus, decrease patient nausea, or more promptly resume a regular diet and maintenance of patient homeostasis.

Effective Implementation: Implementation Stages Framework
The implementation stages framework takes teams through the different stages of effective implementation, including exploration, installation, initial implementation, and full implementation. 17 Study participants identi ed speci c barriers to exploration, including a lack of prioritization of ERPs at Page 24/35 the hospital, resistance among clinical partners (e.g., surgeons, anesthesiologists) due to a lack of familiarity with ERPs, and physicians needing time to adjust to ERPs in practice. Report facilitators included having experts available within the hospital in non-surgical specialties (e.g., gastroenterology, pain specialists) and national collaborators available to provide training to individual hospital-based ERP implementation teams.
Reported installation barriers included challenges with identifying and involving all stakeholders in the ERP onboarding process. Starting the ERP implementation process was reported as comparable to "moving the battleship" given the number of moving parts involved in getting a site ready to implement an ERP. Reported facilitators included having an ERP champion with protected time to prepare the site for ERP implementation.
Reported barriers of the initial implementation stage included a lack of work ow or implementation tracking, variation in ERP implementation, di culties with identifying barriers to implementation, low Participants reported that when surgeons have complex patients with challenging operative cases, they are more likely to abandon ERPs and revert to their traditional practices. Reported facilitators of full implementation included having a protocol available for distribution in multiple formats and discussing the protocol in a multidisciplinary team setting.

Effective Implementation: Implementation Drivers Framework
The implementation drivers framework includes development of competencies, obtaining organizational supports, and engaging leadership to lead to effective implementation. 17 Participants reported that organizational drivers (e.g., electronic health record (EHR) systems to ag or remind providers of ERP elements) were not available at some hospitals and represented a major barrier to creating order sets in the EHR or to identify ERP cases, electronically. Participants also noted barriers to ERP implementation during transitions between phases of care, from pre-to intra-to post-operative settings. For example, order sets for ERPs may not be transferrable between inpatient and outpatient settings ( Table 2). Participants indicated the availability of decision support systems could facilitate ERP implementation through the use of an alert in the EHR to identify patients eligible for ERPs, making ERP checklists available, electronically, for eligible patient encounters, generating a standardized ERP clinical decision support tool in the EHR, and using the EHR for process and outcome measure tracking.
The major barrier reported for competency drivers was staff turnover. For example, trainees (e.g., residents, fellows) and nurses rotating onto the service and not speci cally part of the GI surgical team may require ongoing training and orientation to become familiar with the ERP. One surgeon noted, "I think one of the barriers is that we don't have a speci c team of nurses. We have lots of nurses and the hospital's philosophy overall is that nurses should be trained in everything. So yes, in general we have people who preferentially want to be in one room or another, but that doesn't always work. And so that would be infrequent ... Nurse infrequent with the process, has to have discussion about that. Although, they're trying to put the right people in the room." [MD_8_2].
Participants also noted the need to train and coach surgeons and anesthesiologists to conduct speci c ERP procedures (e.g., regional blocks, perioperative analgesic adjuncts). While reported facilitators of ERP implementation included educating all providers on ERPs, having reciprocal education between different groups of clinicians (e.g., from nurses to surgeons and from surgeons to nurses), and using evidence as part of clinician education being key.
Participants reported barriers to leadership drivers, including a lack of leadership buy-in to support ERP implementation. They reported the need to navigate "o ce politics" to ensure leadership support and resource investment including personnel necessary to measure outcomes. Reported facilitators to implementation included administrative support, protected time for staff and clinicians to implement ERPs, and institutional buy-in to the ERP culture, all of which are more likely to occur when there are littleto-no costs related to implementation. Similarly, several participants brought up hospital and department size as a potential barrier to ERP implementation. Speci cally, they reported that for hospitals and departments with low surgical patient volume, it is di cult to implement education/training due to the infrequency of surgical cases. Participants indicated that pediatric surgery services that are a "wing" in an adult hospital have di culty implementing innovations speci c to pediatric surgery." One barrier which is that we are a children's hospital within an adult hospital… So, we have to think in terms of our patients the early standpoint, because we are sharing preop, postop with the adult population, so we have to go an extra mile to make sure things which we want for pediatric patients are implemented we follow up on those things." [MD_9_2] Enabling Context: Teams Framework The teams framework emphasizes having teams de ne infrastructures and support strategies and improve outcomes; this creates a foundation or enabling factors for ERP implementation. 17 Receptiveness and buy-in were identi ed as key barriers and facilitators to ERP implementation. Key reported barriers included a lack of institutional support, paucity of hospital-speci c outcome data, di culty in identifying the right stakeholders at their hospitals, variation in levels of baseline knowledge and consensus about of ERPs, and lack of clinician receptiveness for speci c ERP elements (e.g., early enteral nutrition). Reported facilitators related to receptiveness and buy-in included having outcomes data demonstrating how ERPs could reduce hospital costs and resource utilization, standardized procedures and protocols for clinicians to follow, and ERPs in other surgical procedures as a reference, and leveraging leadership support to garner buy-in from more clinicians.
Collaboration among team members was noted by participants to be hindered if there was a lack of team consensus on ERPs and poor communication between clinicians. Reported communication barriers could preclude collaboration both within a profession (e.g., among surgeons) and across professions (e.g., surgeons, nurses, anesthesiologists). However, having strong relationships within and across teams and clinician groups was reported to foster collaboration and facilitate implementation. Participants highlighted examples of having regular meetings to discuss ERP implementation broadly, and ERP cases, speci cally. Clearly de ning and communicating the work ow within and across teams was another aspect that was also reported as facilitating collaboration. Some participants from smaller pediatric surgery teams reported being able to access each other quickly by telephone with questions and other teams mentioned having recurring meetings to facilitate communication. As one surgeon highlighted, team engagement is a step in the right direction but not a panacea, "So it really highlighted to us that even when you have engaged family and you know the institutions are moving towards this pathway, that there are hurdles." [MD_1_1] Team engagement barriers included lack of communication among teams, inadequate team composition, and team size. Team engagement was reportedly hindered by the lack of an established ERP protocol. Some participants reported inadequate team composition and lack of engagement of critical team members, such as an ERP representative, patient liaison, child life specialists, and outcomes data collectors. Other participants reported that large team size would be more challenging to engage in the ERP implementation process.
Participants endorsed that engaged teams would be a facilitator to ERP implementation. Early engagement of patients and families, presence of services that were supportive of personnel on the team and having a coordinator to monitor ERP cases throughout the care pathway were important facilitators. Furthermore, the existence of ERP champions (e.g., executive sponsor, surgical lead, anesthesia lead) and a multidisciplinary team (e.g., surgery, nursing, pain management) would facilitate ERP implementation.

Discussion
In-depth interviews with 48 pediatric clinicians and staff caring for pediatric surgical patients at 18 hospital across the US provided insight into the potential barriers and facilitators to the implementation of comprehensive ERPs in pediatric surgery. Guided by the 5AIFs, 17,19,20,30 our analysis identi ed several barriers and facilitators that are unique to pediatric surgery practice implementation, including the need for pediatric patient and family-speci c engagement, challenges of limited pediatric surgical volume, barriers related to pediatric services embedded in an adult hospital with different priorities, and limited available evidence of ERP effectiveness to drive buy-in (e.g., medications).
In the Active Implementation Formula 18 , successful implementation is predicated upon having effective practices, effective implementation, and enabling context to achieve desired outcomes. Applying this framework in our analysis suggests that pediatric surgical services will need to implement evidencebased protocols that are data-driven (effective practices), and clearly identify roles for team members.
Further, success will be contingent upon providing consistent training and support to engage and prepare teams (effective implementation) and providing toolkits to enable implementation and data-driven feedback during the learning process (improvement cycles). Patient and family engagement, data collection and tracking, training and education, surgical volume and practice size, buy-in from stakeholders, and an ERP champion were dominant themes of implementation drivers necessary for successful implementation of ERP implementation in pediatric surgery. This study extends our understanding of barriers and facilitators to ERP implementation for a diverse and complex population not previously studied, namely pediatric surgical patients. 31 Prior studies in adults identi ed barriers of team and administrative buy-in, team receptiveness, 32 team engagement, resources, inter-department collaboration, and standardization of order sets.
Patient and family representation and buy-in were identi ed by study participants as important components throughout the process of implementing ERPs in pediatric GI surgery. Opportunities to involve patients and families include involvement in developing ERPs and introducing patients and families to ERPs prior to surgery. Participants noted the need for administrative support to actively engage families and patients as implementation team members. Including patient and family members on the implementation team can be used as an implementation strategy to co-produce strategies to solve issues around adherence (e.g., drinking a carbohydrate liquid prior to surgery). 33 Patients and families were identi ed as an important part of the care pathway for co-production of educational materials and tools supporting implementation. Further, patients and families need to be introduced to ERPs in an early phase of care, (e.g., preparation prior to surgery for different steps and phases of engagement throughout the perioperative pathway). Such engagement has been modeled at pediatric hospitals successfully implementing ERPs and includes preoperative counseling to introduce patients and families to the "why" of ERPs, the daily goals involved in ERPs, and expectations for a successful discharge and return to baseline activity after surgery 34 .
Although data collection, speci cally surrounding ERP adoption, was noted as important to successful ERP implementation, all participants reported that their hospital had substantial barriers to data tracking.
No hospital was reported to have a formal tracking system in place that could be used for monitoring ERP implementation, noting lack of human resources and clinical tools to properly track which ERP elements were implemented and frequency of implementation. Several key ERP elements cannot be easily abstracted through retrospective review of an EHR, highlighting the need for prospective data collection. Formal audits of ERP implementation are the best way to monitor delity 35,36 and can also be used to increase protocol adherence. 37 Hospitals would bene t from having systematic processes developed to measure adherence to ERP elements based on clearly de ned de nitions of each element.
Training and education were also identi ed as essential components to ERP implementation and sustained compliance. 38, 39 In addition to achieving standardization, training is important to engage all team members throughout the perioperative continuum. Training surrounding ERPs was reported as necessary to garner buy-in from all hospital stakeholders including hospital administration and clinicians.
Training and using data-driven outcomes were endorsed as methods to reduce resistance to ERPs among colleagues and hospital leadership. Multi-level leadership engagement preceding resource allocation is necessary for ERP implementation. Using data to enhance support has also been identi ed as an important facilitator in other studies of ERP implementation among adult surgical populations. 13,14 Hospitals expressed support for use of their own outcomes data to enhance receptivity. Further, extending training to all team members is a barrier when all teams do not have a speci c team of ERP clinicians, rather a rotation of team members. This is a well-known challenge encountered within modular care delivery systems, such as nursing. 40,41 Potential strategies to overcome this barrier include the assignment of a 'Nurse Coordinator' and efforts to centralize patients on one surgical oor. The nurse coordinator can help disseminate critical information to peers, guide the patient throughout the process, reinforce the importance of various ERP elements, and clarify misunderstandings. 37 Similarly dissemination to peers can be facilitated by appointing designated surgeon and anesthesia champions.
Finally, low case volumes of pediatric GI surgeries were a major reported concern. In some hospitals, low case volume limits their ability to use internal data for education and to generate institutional buy-in.
Hospitals with low surgical volume would bene t from having access to clinical outcomes data speci c to pediatric surgery from other hospitals (individual or compiled from multiple hospitals). Because the percentage of patients undergoing non-urgent gastrointestinal surgery procedures is generally low in each hospital, clinicians (e.g., surgeons and anesthesiologists) may not identify a patient as a candidate for many of the aspects of the ERP. Further, the low case volume makes it di cult to ensure standardization of the process and conduct of rapid improvement cycles. Lastly, sustaining the changes of a complex ERP if only applied to a small number of patients can be a challenge. One potential solution is to look for opportunities to horizontally integrate ERPs to other patient populations who can bene t from ERPs. This allows surgical team to increase volume of surgeries performed using ERPs. And hospitals to achieve economies of scale.
As highlighted by a surgeon participant, ERP implementation is a team effort that requires numerous factors to be set into place before the ERP elements can be implemented, measured, and assessed.
Successful implementation needs to go beyond checking off the boxes of the protocol. The qualitative interview data presented in this manuscript, in combination with survey data from the 18 hospitals, 22 are being used to inform an evidence-based toolkit and the development of protocols for ERP implementation. Individual hospitals do not have the resources, time, or administrative support to both develop protocols and standardize procedures.
A strength of this research is the rich, descriptive data that provides a better understanding of ERP implementation in pediatric surgery. Having interviews with surgical teams, compared to one-on-one interviews, provided a more expansive perspective on barriers and facilitators to ERP implementation.
Each site had an ERP champion to help recruit other team members; in this purposive sample, the champions helped identify other team members with working knowledge to discuss their site's implementation. This sampling plan also means this work may not be applicable to sites in precontemplation or without plans to implement ERPs. Those sites may have different barriers to implementation that are not represented in this research.
The next steps in this research are to conduct a multi-site, randomized, controlled trial, ENhancing Recovery In CHildren Undergoing Surgery (ENRICH-US), that supports protocol development and implementation of an ERP for pediatric GI surgery patients and evaluates both the implementation of the ERP and its effectiveness on patient outcomes. 21,37 Strategies will include: (1) Development of implementation strategies and a toolkit targeting these barriers and enhancing facilitators of implementation; (2) Use of a learning collaborative approach to implement an ERP across 18 pediatric surgery services, and (3) Implementation a uniform ERP measurement system and measuring delity of implementation.

Conclusion
This study comprehensively and systematically applied the 5AIFs to organize the identi ed barriers and facilitators of ERP implementation. Ultimately, using this theoretical foundation to code and analyze the data his facilitated the identi cation of potential strategies to support effective practices (evidence-based ERPs and implementation tools), effective implementation (local team infrastructure support including team member roles and responsibilities; a learning collaborative for pediatric surgery groups with monthly support, coaching and training by topic experts, tools to engage key stakeholders), and enabling contexts (a toolkit to support implementation, sustainability assessments, quarterly data-driven feedback sessions, and access to a QI expert on the implementation teams), to achieve the desired outcome. This study establishes a contextual understanding of barriers and facilitators to ERP implementation in hospitals serving pediatric patients undergoing GI surgery. The current study was able to inform and optimize next steps, which are to apply and evaluate these strategies in a future stepped-wedge, cluster randomized trial to increase implementation of ERPs in 18 pediatric surgery centers, which will allow us to determine if the strategies identi ed in this study will be able to increase ERP implementation.

Declarations
Ethics approval and consent to participate The study protocol was evaluated by the Institutional Review Boards at Ann & Robert H. Lurie Children's Hospital of Chicago, Children's Healthcare of Atlanta, Emory University, and Kaiser Permanente Georgia and considered exempt from full review.

Consent for publication
This manuscript does not require consent for publication.

Availability of data and materials
The data that support the ndings of this study are stored with Northwestern University but restrictions apply to the availability of these data, which were used under license for the current study, and so are not publicly available.

Competing interests
The study team does not have any competing interests. WS made signi cant contributions to the study design, data analysis, interpretation, and drafting and revising the work.
SCB made signi cant contributions to the study design, data acquisition, data analysis, interpretation, and drafting and revising the work. SC made signi cant contributions to the study design, data acquisition, data analysis, interpretation, and drafting and revising the work. MVR made signi cant contributions to the study design, data acquisition, data analysis, interpretation, and drafting and revising the work. Figure 1 "See image above for gure legend"