Lead/Presenter: Christine Ramsey, COIN - Pittsburgh/Philadelphia
All Authors: Ramsey CM (Corporal Michael J. Cresenz VA Medical Center, Philadelphia, PA; Yale University School of Medicine, New Haven, CT), Hull LE (Center for Healthcare Organization and Implementation Research, VA Boston Healthcare System), Lynch KG (Corporal Michael J. Cresenz VA Medical Center; Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA) Oslin DW (Corporal Michael J. Cresenz VA Medical Center; Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA)
Objectives:
The PRecision Medicine In MEntal health (PRIME) Care study is a prospective randomized control trial examining genotype-guided therapy for major depressive disorder compared to treatment as usual. We examined characteristics of mental health (MHP) and primary care (PCP) providers associated with patient referrals for pharmacogenetic (PGx) testing.
Methods:
Sample: 413 VHA providers (70.2% MHP, 29.8% PCP) from 19 stations who enrolled in PRIME Care from July 11, 2017 to August 8, 2018. Data sources: Data for patients of enrolled providers were obtained from VHA electronic health records. Provider demographic data and knowledge and comfort with genetic testing were collected in a baseline survey administered to providers. Measures: Provider type (MHP or PCP) was our primary predictor of interest. Our outcome was rate of referral for PGx testing. Patients diagnosed with depression, no serious mental illness or substance use, and ?2 encounters with a provider since enrolment were eligible for referral. Additional covariates included provider demographic and clinical characteristics, and self-rated knowledge and comfort with genetic testing. Analysis: Baseline characteristics of MHPs and PCPs were compared using chi-squared tests. Association between provider characteristics and referral for PGx testing was assessed with Generalized Estimating Equations.
Results:
MHPs saw more patients who met study eligibility criteria (p = 0.013), and were more likely to feel well-informed about the role of PGx testing in choosing a psychotropic medication compared to PCPs (58.6% vs. 23%, p < 0.001). Self-rated training and comfort in ordering genetic tests was low in both MHPs and PCPs ( < 50% agreement on all measures). After adjustment, rates of referral were 4.24 times higher among MHPs (95% CI: 2.13, 8.45) and 1.58 times higher among providers who felt well-informed about the role of PGx testing in choosing a psychotropic medication (95% CI: 1.11, 2.24).
Implications:
Providers' self-rated knowledge, training, and access to expertise in genetic testing differs by provider type and is associated with referral of patients for PGx testing.
Impacts:
Training opportunities and access to expertise are needed to improve providers' knowledge and comfort in the use of PGx testing for Veterans in the VHA. Identifying differences in PCPs and MHPs can inform targeted training and resource allocation.