Clinical Decision Support System for Guidelines-based Treatment of Gonococcal Infections, Screening for HIV, and Prescription of Pre-Exposure Prophylaxis (PrEP): Design and Implementation Study (Preprint)

Abstract

The syndemic nature of gonococcal infections and HIV provides an opportunity to develop a synergistic intervention tool that could address the need for adequate treatment for gonorrhea, screen for HIV infections, and offer pre-exposure prophylaxis (PrEP) for persons who meet the criteria. By leveraging information available on electronic health records, a clinical decision support (CDS) system tool could fulfil this need and improve adherence to CDC treatment and screening guidelines for gonorrhea, HIV, and PrEP.

The goal of this study was to translate portions of CDC treatment guidelines for gonorrhea, and relevant portions of HIV screening and prescribing PrEP that stem from a diagnosis of gonorrhea, as an EHR-based CDS intervention. We also assessed whether this CDS solution worked in real-world clinic.

We developed four tools for this CDS intervention: a form for capturing sexual history information (‘SmartForm’), rule-based alerts (‘Best Practice Advisory’; BPA), an enhanced STI order set (‘SmartSet’), and a documentation template (‘SmartText’). A mixed-methods, pre-post design was used to measure the feasibility, acceptance, and usefulness of the CDS solution. The study period was 12 weeks with a baseline patient sample of 12 weeks immediately prior to the intervention period for comparison. While the entire clinic had access to the CDS solution, we focused on a subset of clinicians who frequently engage in the screening and treatment of STIs within the clinical site under the name ‘X-Clinic’. We measured the utilization of the CDS solution within our target population of patients who had either a confirmed gonococcal infection, or an STI-related chief complaint. We conducted four mid-point surveys and three key informant interviews to quantify perception and impact of the CDS solution and solicit suggestions for potential future enhancements. Findings from qualitative data were determined using a combination of explorative and comparative analysis. Statistical analysis was conducted to compare the differences between patient populations in the baseline and intervention periods.

Within the X-Clinic, the CDS alerted clinicians (as a BPA) in one-tenth (348/3,451, 10.08%) of clinical encounters. These 348 encounters represented 300 patients; SmartForms were opened for half of these patients (157/300; 52.33%) and was completed for most for them (147/300, 89.81%). STI test orders (SmartSet) were initiated by clinical providers in half of those patients (162/300, 54%). HIV screening was performed during about half of those patient encounters (191/348, 54.89%) and the odds of ordering these tests were eleven times higher in these patient encounters (OR:11.28, CI:8.86-14.38) compared to those encounters where CDS did not alert clinicians (302/3,103, 9.73%).

We successfully built and implemented multiple CDC treatment and screening guidelines into a single, cohesive CDS solution. The CDS solution was integrated into the clinical workflow and had a high rate of utilization.

Not applicable