A Case Study: Optimizing CDS for Pediatric Oncology Trials by Transitioning from Interruptive to Passive Alerts
Bibliographic record
Abstract
Many children with cancer are treated as part of interventional clinical trials. Ensuring that the correct chemotherapy treatment plan is used is paramount.The objectives of this report were to: (1) highlight the initial design of a clinical decision support (CDS) tool that was intended to help ensure the correct matching of research studies to research chemotherapy medications, (2) discuss the issues identified with the CDS tool, and (3) review the redesign of the tool that was done to overcome the issues identified.We previously utilized an interruptive alert developed by Epic Systems to identify mismatches between a patient's chemotherapy plan and research study. We identified an issue with the logic of the alert resulting in the alert firing inappropriately.We estimate that the chemotherapy-research plan alert fired when 93.4% of treatment plans were applied (17.3 alerts/provider/year). A high number of misfiring alerts were identified due to the inclusion of our institution name as both (1) a "tag" in the research protocol, and (2) an unallowed tag in the research study record. Since the tag was included in all protocols, but also unallowed in all research records the alert fired with the application of almost all treatment plans. We developed a new mechanism to provide CDS that did not involve an interruptive alert. Within the research study record, we manually associate compatible treatment plans to that study record, and then when an oncologist goes to order chemotherapy the system prioritizes the display of compatible treatment plans to the oncologist. The goal of the redesigned CDS approach is to eliminate interruptive alerts while ensuring the correct chemotherapy plan is selected.With end-user engagement and creative approaches to CDS design, interruptive alerts can be transitioned into passive and effective CDS tools.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.083 | 0.235 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.004 | 0.003 |
| Science and technology studies | 0.003 | 0.002 |
| Scholarly communication | 0.007 | 0.007 |
| Open science | 0.004 | 0.004 |
| Research integrity | 0.005 | 0.005 |
| Insufficient payload (model declined to judge) | 0.005 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".