Lead/Presenter: Mark Bounthavong,
All Authors: Bounthavong M (Pharmacy Benefits Management (PBM) National Academic Detailing Service) Harvey MA (PBM National Academic Detailing Service) Himstreet JE (PBM National Academic Detailing Service) Popish SJ (PBM National Academic Detailing Service) Wells DL (PBM National Academic Detailing Service) Lau MK (PBM National Academic Detailing Service) Kay CL (PBM National Academic Detailing Service) Torrice V (VA Central Office) Valentino M (VA Central Office) Christopher ML (PBM National Academic Detailing Service)
Objectives:
In 2015, the Interim Under Secretary of Health mandated that all VISNs implement an Academic Detailing Program in order to improve evidence-based delivery of mental health and pain management medication services. Therefore, we sought to evaluate the impact of the VA Pharmacy Benefits Management National Academic Detailing Service (ADS) on the Opioid Overdose Education and Naloxone Distribution (OEND), Opioid Safety Initiative (OSI), Psychotropic Drug Safety Initiative (PDSI) campaigns.
Methods:
We performed three separate retrospective, repeated measures cohort studies that examined the impact of ADS on naloxone prescribing for patients at-risk of opioid overdose, opioid utilization in Veterans on high-dose opioids, and benzodiazepine prescribing in patients with posttraumatic stress disorder. This was a provider-level analysis where providers exposed to academic detailing (AD-exposed) were compared to providers unexposed to academic detailing (AD-unexposed). Longitudinal analyses were performed using generalized estimating equations with autoregressive correlation structure to handle repeated observations within-subjects adjusting for provider-level characteristics.
Results:
In the OEND analysis, AD-exposed providers (N = 750) had a greater rate of naloxone prescriptions prescribed compared to AD-unexposed providers (N = 2563) over two years (7% greater rate; 95% CI: 2%-13%). In a separate study evaluating ADS's impact on high-dose opioids and average monthly MEDD, ADS reported that AD-exposed providers (N = 1813) had a greater rate of reduction in the proportion of patients with high-dose opioids (58% versus 34%, respectively; p = 0.013) and average MEDD (59% versus 31%, respectively; p < 0.001) compared to AD-unexposed providers (15,993) over 36 months. Additionally, in a separate analysis on the PDSI, AD-exposed providers (N = 274) had a greater reduction in the proportion of PTSD patients on benzodiazepines compared to AD-unexposed providers (N = 1424) over 12 months (14% versus 7%, respectively; p = 0.045).
Implications:
These were the first empirical evidences of ADS's impact using real-world data. This suggests that provider behaviors can be modified to be aligned with evidence-based practices. Further research is needed to determine the elements of academic detailing that have the greatest impact on changing providers' behavior.
Impacts:
Although ADS has demonstrated its effectiveness, serious examination of its elements will help decision makers optimize the program and expand into other disease areas.