Leveraging Electronic Health Record Data in a Risk Prediction Model to Increase HIV PrEP Intake in Primary Care

"PCPs notified by EHR messages if they had an upcoming in-person or video visit with an eligible patient who had a predicted risk of incident HIV diagnosis of 0.2% or higher. Study team members manually sent messages marked as high priority one business day prior to scheduled visits, alerting PCPs that a patient may be at increased risk for acquiring HIV and may benefit from a conversation about PrEP. Additional information was provided about the study, the prediction model, clinical indications for PrEP, strategies for initiating discussions about HIV prevention and PrEP, and how to refer patients to KPSF's PrEP program. Messages did not include the numeric predicted HIV risk from the model. PCPs received a maximum of two messages per day; with the two patients with the highest predicted risk were selected. Patient with multiple visits were flagged no more than once monthly."