Sponsored Health IT and Evidence-Based Prescribing Among Medical Residents

Failure to follow evidence-based prescribing (EBP)--the use of current evidence in prescribing medications--has been linked to an increase in drug abuse and misuse, morbidity, mortality, and increased healthcare costs. A substantial proportion of the three billion prescriptions written annually are not evidence-based. Direct marketing by pharmaceutical companies has been shown to impact how physicians prescribe medications. Concurrent with the increased use of mobile applications for decision making at the point-of-care, pharmaceutical-sponsored messaging within applications has been increasing, potentially impacting prescribing patterns. As such, there is a pressing need to understand how this new form of targeted marketing may lead to non-EBP. Additionally, physician training programs need to reaffirm the basic principles of EBP, particularly in the context of sponsored health IT messaging.

This project developed SMARxT, a series of six web-based videos examining strategies for overcoming EBP challenges including biases introduced by pharmaceutical marketing and the use of applications to guide clinical decision making. Researchers developed a knowledge-based assessment framework for each of the six videos. This assessment was piloted with a sample of medical students, and items were refined and leveraged to conduct a longitudinal study in a sample of resident physicians.

The specific aims of the project were as follows:

• To determine the feasibility and acceptability of SMARxT among resident physicians. 
• To assess the efficacy of SMARxT in increasing knowledge surrounding EBP practice. 

Knowledge-based items were developed by iteratively reviewing and refining the items, expert review, pilot testing, and evaluation of psychometric properties. The resulting 38-item assessment was augmented with additional items assessing acceptability of the modules and attitudes about covered topics. Residents completed a knowledge pre-test prior to watching videos. The immediate post-test and 6-month post-test assessed knowledge, acceptability, and clinical attitudes related to SMARxT.

Overall, the SMARxT program demonstrated acceptability as well as preliminary efficacy. Knowledge scores increased significantly in all six domains before and immediately after implementation. Six-month followup scores remained significantly higher than pre-test scores, indicating that knowledge was retained. Assessments of acceptability were favorable, although some participants indicated preference for shorter or faster videos. As a self-paced implementation tool, including additional features like higher speed playback or interactive web content may further enhance program acceptability. The SMARxT program may be considered as a possible candidate for classroom implementation.