Designing a Clinical Tool for Visualizing Variation in Pediatric Perioperative Opioid Use
Abstract/Contents
- Abstract
- Anesthesiologists integrate diverse sets of variables to determine appropriate opioid doses for surgery that sufficiently manage patient nociception and pain during the operation, while minimizing adverse effects. Understanding and reducing unnecessary variation in opioid administration is key to achieving optimal patient safety and health outcomes. However, relevant data is currently disparate, and direct comparisons amongst anesthesiologists are complicated by case-specific patient demographics and surgical factors. We engaged healthcare specialists associated with surgical care at a pediatric hospital to develop end-user requirements for an opioid utilization data dashboard and identified patient characteristics and surgical factors that may impact opioid administration. After aggregating and cleaning 5.33 years of surgical case data, mathematical modeling determined variable importance and overall percent of variation explained (R2) by all variables currently available. Finally, we worked with stakeholders to integrate modeling results to iteratively build and deploy an opioid utilization dashboard. Modeling identified patient weight, patient age, and surgery type to be most responsible for opioid-use variability. Machine learning algorithms using a comprehensive set of the available patient and surgical variables explained 58.6% of variation in intraoperative opioid dose delivered to children. This suggests that a large portion of variation results from uncaptured data, such as patient vital signs and difficult to capture data, including random and potentially unnecessary clinical variation between similar cases. The iterative dashboard design process identified two mechanisms to enable comparative visualizations with meaningful case-mix adjustment: an interactive data filtration tool and machine learning-based normalization. When deployed in real practice, our dashboard identified a number of low opioid use outliers who effectively reduced utilization with opioid alternatives while maintaining patient outcomes. These results demonstrate one way in which the dashboard can help physicians learn from historical practice. We developed algorithms and an interactive interface that applies contemporary data informatics to provide anesthesiologists with timely data on their practice patterns while adjusting for case-mix differences. Normalized comparative visualizations can empower physicians to track changes and variation in opioid administration over time and to understand their practice relative to their peers. Access to transparent, interpretable data may help reduce unnecessary variation in surgical care opioid utilization and thereby improve patient health outcomes.
Description
| Type of resource | text |
|---|---|
| Date created | June 2021 |
Creators/Contributors
| Author | Safranek, Conrad W. |
|---|---|
| Primary advisor | Scheinker, David |
| Advisor | Bergmann, Dominique |
| Advisor | Glynn, Peter |
| Degree granting institution | Stanford University, Department of Biology, 2021 |
Subjects
| Subject | Anesthesiology |
|---|---|
| Subject | Stanford University Department of Biology |
| Subject | Stanford University |
| Subject | Decision support systems |
| Subject | Clinical informatics |
| Subject | Machine learning |
| Subject | Pediatrics |
| Genre | Thesis |
Bibliographic information
Access conditions
- Use and reproduction
- User agrees that, where applicable, content will not be used to identify or to otherwise infringe the privacy or confidentiality rights of individuals. Content distributed via the Stanford Digital Repository may be subject to additional license and use restrictions applied by the depositor.
- License
- This work is licensed under a Creative Commons Attribution Share Alike 3.0 Unported license (CC BY-SA).
Preferred citation
- Preferred Citation
- Safranek, Conrad W.; Scheinker, David; Bergmann, Dominique; and Glynn, Peter. (2021). Designing a Clinical Tool for Visualizing Variation in Pediatric Perioperative Opioid Use. Stanford Digital Repository. Available at: https://purl.stanford.edu/yf740hg5785
Collection
Undergraduate Theses, Department of Biology, 2020-2021
View other items in this collection in SearchWorksContact information
- Contact
- conradws@stanford.edu
Also listed in
Loading usage metrics...