Our study provides the first multi-center analysis of EHR alert design elements in pediatric care settings, revealing substantial variation in content and design. These findings underline the need for future research to experimentally explore EHR alert design best practices to improve efficiency and effectiveness.
While there are many published pediatric AI models, the number of AI implementations is minimal with no standardized reporting of outcomes, care processes, or human performance measures. More comprehensive evaluations will help elucidate mechanisms of impact.
Current models of operations and research involving clinical decision support (CDS) are too site-specific, not scalable, and do not measure translational outcomes. We propose to create a collaborative infrastructure centered around reusable CDS metrics that more effectively measures clinical outcomes and compares these benefits against the tradeoffs of alert burden and fatigue. Through our collaborative approach across ten CTSA Hubs, this study will be the first to use a translational science approach to understand the impact of CDS systems.
This panel highlights some of the Pediatric CDS Collaborative’s most ignominious CDS roaches – hard to kill tools that far outlived their utility – to share hard earned lessons and successes organized as roach facts
In this panel, we describe some of our most inglorious misfires as case vignettes in clinical decision support (CDS) implementation to share hard-earned lessons organized into 8 sociotechnical dimensions.
Further refinement of alert burden metrics is needed as current metrics do not adequately represent the impact on end users when viewed through different dimensions. The best practices we describe here will allow institutions to establish monitoring and optimization programs to reduce alert burden.
Estimates of the clinical areas with highest alert burden varied substantially by institution and based on the metric used.
We discuss benefits and challenges in developing a pediatric learning collaborative for clinical decision support benchmarking and sharing of best practices, describe the strengths and weaknesses of different approaches to alert burden measurement, demonstrate how alert prioritization can inform quality improvement initiatives to reduce alert burden, examine unique measurement challenges for non-EHR alerts such as phone calls and text messages, and discuss strategies for linking alert burden analytics to governance strategies that facilitate improvement.