Abstracts

Rationale:
Expert consensus statements developed through consortia and task forces are critical in defining a collective standard of care. Recommendations from The ACNS Critical Care Continuous EEG Task Force (J Clin Neurophysiol. 2015 Apr;32[2]:87-95) are broadly used to guide continuous EEG in critically ill adults and children. The ultimate effectiveness of any expert recommendation depends upon implementation. Local implementation may vary, and is dependent upon available resources, demand, and stakeholder engagement. At our center, requests for critical care continuous EEG (CCEEG) have increased by more than 50% over the past five years. In this demanding environment, lack of standardization around CCEEG initiation, monitoring, notification, and discontinuation has contributed to perceived practice uncertainty and burnout. Such factors may also frustrate efforts to gain consensus around process improvement. Traditional consensus-building tools may be subject to several unintended biases, including expertise, authority, and reactive devaluation bias. The primary aim of this single-center implementation science project was to create institutional CCEEG care guidelines with high stakeholder consensus and engagement with minimized bias.

Methods:
Thirty-two stakeholders involved in daily CCEEG operations (four intensivists, three neonatologists, four EEG technologists, four residents, ten neurologists, one surgeon, five epileptologists, and one administrator) were identified. Stakeholder contributions derived from focus groups and semi-structured interviews were combined with expert consensus statements, process maps, and a primary literature review to create institutional draft guidelines (Figure 1). Draft guidelines were divided into three sections: 1) CCEEG indications, 2) staff roles and responsibilities, and 3) CCEEG communication flowchart. Employing Delphi techniques online using Microsoft Whiteboard, stakeholders provided anonymous draft guideline feedback visible to all reviewers, using electronic sticky notes placed around each page of the draft guideline document (Figure 2). The draft guideline was iterated at the end of each one-week feedback cycle. Feedback comments were collated and validated by a Delphi cycle facilitator (DJP). Comments from past cycles were aggregated and remained visible during subsequent rounds, to promote inter-cycle feedback stability. Consensus was determined by stakeholder assent and by cessation of novel feedback.

Results:
Overall, four weeklong feedback cycles were completed. Fifty-four feedback statements achieved consensus. Feedback was divided into three sections: 1) CCEEG initiation timeframe, 2) communication, and 3) non-epileptologist scope of EEG interpretation. More than 80% of responders agreed that final guidelines were acceptable, feasible, and represented critical consensus on post-cycle testing.

Conclusions:
Adoption of expert guidelines is dependent upon local implementation. Clinical burden, siloed communication, and bias are all potential barriers to effective implementation. Here we demonstrate successful use of the Delphi method to achieve high levels of consensus and engagement, resulting in feasible and representative institutional care guidelines.

Funding:
None