Throughout the project, we gathered feedback from clinical personnel and family advisors regarding various processes and materials in order to improve things as we went along. In this section, we summarize key learning from that feedback as well as reflections from the core project team.

Composition of core team

  • In addition to a physician lead, we felt that it was important to have a clinical team member (in this case a congenital heart surgery nurse practitioner) who had knowledge of the patient population. It was especially helpful that she was the same person who managed the data collection and entry for the STS database.
  • One of the challenges we faced was the multiple roles for the TeamSTEPPS facilitator. We started out with the same person having responsibility for training, coaching and observing teamwork behaviors. Not only was this a lot for one person to do, but it introduced potential bias to have the same person training a team and then scoring their teamwork behavior. We addressed this issue by training student interns to conduct teamwork observations post-training.

Inclusion of family advisors

  • Participation from family advisors was extremely valuable, even more so than we had anticipated. Their feedback helped us prioritize and gave us direction for communication areas in which to focus. For example, the idea to summarize daily goals for families in such a way that empowers them to be proactive in helping their child get better came directly from discussions with families.
  • The coordination and time required to include family advisors was more involved than what we had planned and required some reprioritizing on our part. Because these families are busy and volunteering their free time, we often ended up engaging them one or two at a time rather than finding time when they could all meet together. We would suggest planning specifically for this area, given its importance, and utilizing patient- and family-centered resources in your institution, if available.
  • In addition, our physician leader completed training through the Institute for Patient and Family Centered Care, and we suggest including someone with experience and training in this area to oversee appropriate incorporation of family and patient advisors in a project.

Incentives for participation

  • Although the process to have our project approved for credit for the quality improvement portion of Maintenance of Certification (MOC) through the American Board of Pediatrics was cumbersome, it turned out to be a huge incentive for physician participation. Participating physicians were required to complete TeamSTEPPS training and several project activities, such as attending peer review conferences, piloting tools, and submitting feedback surveys. We believe the MOC credit was a large part of our having 100% of the physician faculty participating fully in the requirements without difficulty.
  • Continuing education for completing TeamSTEPPS training was also helpful. Another suggestion would be to include activities as part of hospital credentialing and resident education requirements so that participation in this project fulfills already existing requirements.

Cardiac rounds

  • Respondents to a feedback survey reported that having standardized cardiac rounds in the intensive care unit improves communication among cardiothoracic surgery, cardiology, and ICU teams. It allows everyone to be present at the same time and cuts down on time required to follow up later and potential miscommunication.
  • However, efficiency is a challenge since cardiac patients are not grouped together in our intensive care unit, and the cardiologists are competing with timing of rounds for the intermediate care unit.
  • Suggestions for improvement include grouping cardiac patients together, limiting the time for each patient, having more consistency in reporting, and bringing in additional subspecialties, when needed.

Peer review quality improvement conference

  • We initially found timing of this conference difficult, so to optimize attendance, we placed it immediately after the standing weekly cardiac catheterization/surgical conference, which already included several of the teams. In addition to cardiology, cardiothoracic surgery, anesthesiology, and critical care, we expanded the attendance to include medical staff from neonatology and perfusion.
  • We found that the attendees were very eager to participate in this conference when real time institution-specific data was made available in addition to the typical morbidity and mortality discussions of peer review.
  • Through this forum, we formed subcommittees or expert panels that worked at standardizing different practices and created guidelines for certain areas of care such as operative (regional anesthesia, fast track extubation, surgical access issues), NICU to PICU/CICU transition (handoffs, timing of transfers, coordination of care efforts), PICU/CICU to cardiac intermediate care (handoffs, clinical pathway continuation, preparation of families to leave ICU). These subcommittees were not originally planned but were natural progressions of this structure.
  • We found that soliciting feedback from this peer review group to include format, frequency of meetings, and topics to be discussed was important for participation. Though this conference was initially planned quarterly, the conference attendees voted to hold the meeting bi-monthly rather than monthly or quarterly.

Teamwork training

  • In the evaluations of Ready Training, participants indicated that the most valuable parts of training were learning effective communication techniques and tools, and examining the importance of teamwork and team organization. Several participants also noted that the training helped them feel more empowered and confident in their role and ability to speak up.
  • Those who attended Booster Training said that a review of the communication tools was helpful.
  • Participants in both trainings liked having examples that were specific to their unit or role, and a few Ready Training participants suggested that the training could be improved by having more interaction or role play and more instruction on use of the tools.

Implementation of integrated clinical pathways

  • We found the development of clinical pathways more time consuming than we had anticipated originally. Although we had access to existing clinical pathway templates and designed our own template with key components of the ICP Checklist in mind for completeness, placing these pathways into the correct context of our institution was complicated. For example, we do not have a separate ICU for pediatric cardiac surgery patients, so we needed strategies for identifying the pathway patients (as described in “Integrated Clinical Pathways”).
  • Another complication of ICP implementation was that each clinical team had to address separately its role and process in the pathway. For example, the person completing the ICP might have been a fellow or attending physician in the ICU and a resident physician or nurse in the intermediate care unit.
  • Once we found our way in designing and implementing our first pathway, the approach to other pathways was easier to envision and complete. Because we had developed a process and educated staff on using ICPs, significant ground work was already done for the second pathway.
  • Reviewing the pathways with the expert panels also led to collaborative efforts to address transfer criteria between units in order to be more consistent and have more agreement across groups.
  • For the most part, physician groups welcomed the standardization of the ICPs presented. There was some concern that the daily template could get too busy, especially with more complex patients, but this did not turn out to be a problem.
  • After the first year of the project, we realized that, in addition to clinical pathways for specific surgical procedures or patient populations, we needed clinical pathways and guidelines for the management of morbidities or complications common to this patient population. This was a natural response to our multidisciplinary discussions regarding our dashboard and outcomes, and was further justified by analysis of the STS database1.
  • As the patients reached transition to the cardiac intermediate care unit, a more generic discharge goals sheet was developed for all cardiac patients in that unit (including those not on a specific pathway) to improve utilization. The frontline staff and medical team in the intermediate care unit felt that if discharge goals sheets were used for all of the cardiac patients, then there would be improved adherence to pathway goals for the ICP patients.
  • As the clinical pathways were implemented and accepted by frontline staff, we found that some of the team members wanted to spread the ICPs to similar patient populations not necessarily fitting one of the clinical pathway diagnoses. For example, a patient that had a cardiac surgical procedure that allowed for a short ventilation period and ICU stay (atrial septal defect) may have a similar post-operative plan and goals as the VSD patient. In these cases, the team requested to use the VSD clinical pathway to standardize the care of these patients. In response, we are creating generic cardiac clinical pathways that focus on expected length of stay for patients that do not fit a current specific pathway. This process could be designed from the beginning focusing on similar surgical populations and not specific surgical repair types. This may allow for a decreased total number of clinical pathways or may allow for standardized care while more specific clinical pathways are developed.


  • We found it extremely helpful that the cardiothoracic nurse practitioner, who treats all of our patients, also had been the STS data manager for our congenital heart surgery patients for the past several years. She had the historical perspective of STS and data definitions, and could remember details about the patients.
  • Some of the institutional data (not from STS) had not been reviewed in this way before, and it took several rounds of running the data to get what we needed. Given that this challenge is likely unavoidable in your institution as well, we recommend gathering a data team with clinical and data experts from both the project team and the data source or department as early in the project as possible in order to figure out the best way to obtain data.

1 Pasquali SK, He X, Jacobs JP, et al. Evaluation of Failure to Rescue as a Quality Metric in Pediatric Heart Surgery: An Analysis of The STS Congenital Heart Surgery Database. Ann Thorac Surg. 2012;94:573-80.