(CCCNP003) INTEGRATING ARTIFICIAL INTELLIGENCE FOR QUALITY IMPROVEMENT: SGLT2-INHIBITOR INITIATION FOR PATIENTS MEETING CLINICAL PRACTICE GUILDELINE CRITERIA. A NURSE PRACTITIONER PATIENT OPTIMIZATION INITIATIVE

Background: Patients with heart failure with reduced ejection fraction (HFrEF) or coronary artery disease (CAD) and type 2 diabetes (DM2) have strong indication for initiation of Sodium-glucose Cotransporter-2 inhibitors (SGLT2-i). This is supported in Canadian Cardiovascular Society (CCS) clinical guidelines. In daily clinical practice, pharmacotherapies may be overlooked, especially with multiple therapeutic options available. Efforts to optimize guideline directed medical therapy (GDMT) is a priority to improve patient outcomes.

METHODS AND RESULTS: Vita Diagnostics is an outpatient cardiology practice in Calgary, Alberta with a large volume of patients with HFrEF and/or CAD/DM2. A quality improvement (QI) initiative, led by the Nurse Practitioner (NP) and supported by a Registered Nurse, integrated artificial intelligence into the electronic medical records to identify those who may qualify for SGLT2-i initiation to optimize care. AI technology was provided through ENSHO Health Intelligent Systems Inc., a health technology company whose mission is to empower healthcare providers to do more with data.

The AI uses a CCS guideline algorithm to identify patients with HFrEF and/or CAD/DM2 who are not currently identified at taking an SGLT2-i. Data is mined at 90–180-day intervals. Thus far, 212 patients were identified. Patients were then reviewed for indication, safety, contraindications, prior use or intolerance by the NP/RN. Suitable patents were contacted to discuss initiation of therapy. Currently, 30 patients have optimized therapy with SGLT2-i initiation, 9 deferred therapy, 6 deceased and 27 are pending contact/follow-up. This QI initiative is ongoing.

Conclusion: Quality improvement initiatives can optimize patient care with the goal of improving outcomes. We have demonstrated that purposeful implementation of guideline directed medical therapies for patients with HFrEF and/or CAD/DM2 was enhanced through the utilization of AI technology for patient identification.