Clinical Decision Support for PrEP (PrEDICT)

Optimizing EHR-Based Prediction Models to Improve HIV Preexposure Prophylaxis Use in Community Health Centers

OCHIN, Inc., Oregon Health and Science University, National Institute of Mental Health (NIMH), Beth Israel Deaconess Medical Center

Scale-up of HIV preexposure prophylaxis (PrEP) is a key strategy of the federal initiative to end the HIV epidemic. However, healthcare providers lack tools to identify patients who are at increased risk for HIV infection and thus likely to benefit from PrEP. This pilot study will test the hypothesis that an electronic health record (EHR)-based clinical decision support system that incorporates an HIV risk prediction model can help providers identify patients at increased risk for HIV infection and improve PrEP prescribing in safety-net community health centers. The clinical decision support system will be implemented in the EHR at 2-3intervention clinics, while 2 control clinics will receive standard of care. The primary outcome is PrEP prescriptions. Other key metrics of PrEP-related care to be assessed include medication persistence, adherence to monitoring guidelines for PrEP, and rates of HIV/STI testing and diagnoses. The expected outcome is the foundation for a large-scale cluster randomized trial to test whether EHR-based clinical decision support tools for PrEP can improve PrEP prescribing and prevent new HIV infections in a national network of community health centers.

HIV, preexposure prophylaxis (PrEP), risk prediction, clinical decision support, electronic health records

The intervention will be implemented in 2-3 clinics, with 2-3 control clinics receiving standard of care. Eligible community health centers will be those with at least 500 patients and a minimum of 10 HIV diagnoses. Clinics located in counties identified as HIV hotspots will be prioritized. Clinics specializing in HIV or sexual health care, where providers may have less need for PrEP decision support, will be excluded. Overall, 26 clinics are expected to meet these eligibility criteria, with a total of 831 incident HIV diagnoses in 2018. Of eligible clinics, 2-3 will be recruited to receive the intervention. Using a covariate-constrained matching procedure with EHR data on important covariates (e.g., urbanity and the sex, race, and age distributions of the patient population), 2-3matched control clinics will be selected.

EHR-based decision support tools to support PrEP discussions and prescribing for patients who have increased predicted HIV risk

Two control clinics will be selected based on a set of matching criteria, e.g., urbanity and the sex, race, and age distributions of the patient population. Control clinics will not participate in study activities.

Healthcare providers at 2 intervention clinics (total of approximately 60 providers anticipated) will be prompted by an EHR-based tool to discuss PrEP with patients whose demographics and clinical history indicate increased predicted HIV risk. Providers will be offered clinical decision support tools to guide sexual health discussions and support PrEP prescribing.

Rates at which providers view alerts about patients with increased predicted HIV risk and rates at which providers access the EHR-based decision support tools using Epic date-time stamps

Inclusion Criteria: Primary care providers at participating clinics who are licensed to prescribe PrEP Exclusion Criteria: N/A