OAII-004 - CARE TEAM ATTRIBUTES PREDICT LIKELIHOOD OF UTILIZING PHARMACOGENOMIC INFORMATION.

Z. Huang¹, M. Jack², R. Knoebel², K. O'Leary³, E. Nutsecu⁴, T. Chen², K. Yeo², S. Hartman², A. Choksi², G. Ruhnke², M. Perera³, M. Ratain², D. Meltzer², P. O'Donnell²; ¹University of Chicago Pritzker school of Medicine, ²The University of Chicago, ³Northwestern University, ⁴University of Illinois Chicago.

Background: Medication prescribing is imperfect, and unintended side effects complicate patient care. Pharmacogenomics (PGx) is an emerging solution to this challenge that associates individual genotypes with personalized drug-related outcomes, but it has not been widely adopted. We hypothesize that patient and provider attributes may predict and help promote PGx utilization.

Methods: We studied PGx using data from the ACCOuNT study, a prospective clinical trial that examined the feasibility of implementing PGx information for self-identified African American inpatients across multiple institutions [Clinicaltrials.gov NCT03225820]. Inpatients and their care teams, including physicians, pharmacists, and mid-level providers, were enrolled from 2016 to 2021. Patients were genotyped and their PGx information was made available within each institution’s informatics dashboard. PGx use was left to enrolled provider discretion. Our primary end point was to identify patient and care team attributes associated with PGx use. We identified statistically significant univariate predictors and utilized a logistic regression to compare relative predictiveness.

Results: This study included 186 patients (60.2% female, mean age 54.2 [sd = 16.2], 75.3% treated at the University of Chicago, 17.7% at Northwestern University, and 7.0% at University of Illinois Chicago) and 188 providers (63.8% MD, 22.3% PharmD, 6.3% PA, and 7.4% APN). In multivariate analysis, we found that use of PGx information in at least one patient significantly predicted use in subsequent encounters during the trial (OR 6.9 [2.7 – 20.7], p < 0.05). Similarly, pharmacist presence on care teams significantly predicted PGx use (OR 3.9 [1.98 – 7.87], p < 0.05). No individual patient factors predicted care team PGx use.

Conclusion: These findings suggest that controllable care team attributes, such as pharmacist presence or successful initial adoption measures, correlate with increased PGx use. Future studies should examine the impact of care team composition on PGx adoption and how to better engage patients in personalized medicine.