Topic: Data Science
Podium Presentation in Room 3 on Thursday at 9:00 (Chair: Lindsay Bazydlo)
Authors: Benjamin K Beppler
Introduction: While periodic urine drug testing for patients prescribed opiates for chronic pain is either strongly recommended or required by many state and federal agencies (e.g., the Centers for Disease Control recommends at least annual testing, while the State of New Mexico requires at least biannual testing), interpretation of testing results often presents a challenge to the ordering physician. In response, clinical laboratories have begun to offer interpretive reports, which accompany urine drug test results and attempt to aid the ordering physician in the assessment of patient compliance with prescribed medications. However, the generation of these reports can be tedious and time consuming manual process. Our goal was to develop a review by exception workflow for these interpretive reports, whereby the majority of reports can be resulted automatically, while holding only difficult or questionable cases for manual review.
Methods: Ordering physicians requesting this service are required to supply opiate prescription information at order (e.g. patient is prescribed oxycodone). This information, along with instrument results from a LCMSMS based urine drug confirmation, are fed through a series of calculations in Data Innovations (DI), the software we currently use as middleware for all of our mass spectrometry testing.
Results: This presentation will detail how the DI calculations were built, why we chose to build what we did while excluding certain scenarios, and ways the calculations can be expanded to include additional information. Case studies will be presented to demonstrate several common and not so common testing results generated during validation.
Discussion: Based on historical opiate testing results, we expect this workflow to be able to result approximately 85% of all interpretive reports without direct human interaction, leaving a much more manageable review workload for our clinical staff.
|Planning to mention or discuss specific products or technology of the company(ies) listed above:||