Implementation of a Nurse-Navigator Run AAA Program Using Natural Language Processing and Electronic Medical Records Successfully Identifies AAA Patients Not Being Actively Followed

Abdominal aortic aneurysms (AAAs) are often identified incidentally on imaging studies. Patients and/or providers are frequently unaware of these AAAs and the need for long-term follow-up. We evaluated the outcomes of a nurse-navigator run AAA program, which uses a natural language processing (NLP) algorithm applied to the electronic medical records (EMRs), to identify patients with imaging report-identified AAA not being actively followed up. A commercially available AAA-specific NLP system (Illuminate, Overland Park, KS) was run on EMR data (Epic, Verona, WI) at a large, academic, tertiary hospital with an 11-year historical look back (January 1, 2010 to December 31, 2020) to identify and characterize AAAs. Beginning January 1, 2021, a direct link between the NLP system and the EMRs enabled a real-time review of the imaging reports for new AAA cases. A nurse-navigator (1.0 FTE) used software filters to categorize the AAAs according to predefined metrics, including repair status, and adherence to the Society for Vascular Surgery imaging surveillance protocol. The nurse-navigator then interfaced with the patients and providers to reestablish care for patients not being actively followed up. The nurse-navigator characterized the patients as case closed (eg, deceased, appropriate follow-up elsewhere, refused follow-up), cases awaiting review, and cases reviewed and undergoing ongoing surveillance using AAA-specific software. The primary outcome measures were the yield of surveillance imaging performed or scheduled, new clinic visits, and AAA surgery for patients not being actively followed up. The contribution margin was the estimated revenue minus the direct costs. During the prospective study period (January 1, 2021 to December 30, 2021), 6,340,505 imaging reports were processed by the NLP (Fig). After filtering for studies likely to include the abdominal aorta, 243,889 imaging reports were evaluated, resulting in the identification of 5609 nondeceased patients with an AAA. When stratified by the maximum aortic diameter, 2324 AAAs were 2.5 to 3.4 cm, 478 were 3.5 to 3.9 cm, 521 were 4 to 5 cm, and 271 were >5 cm. In addition, 204 AAAs were saccular, 303 had previously received open repair, and 840 had previously received endovascular repair. The status for 688 AAAs was unknown. Of these, 1621 cases had been reviewed and closed, 1393 had been reviewed and placed in ongoing surveillance, and 2595 were awaiting review. This yielded 40 finalized imaging studies, 29 scheduled imaging studies, 29 new patient clinic visits, and 4 AAA surgeries among the patients not being actively followed up. The program generated $60,032.24 in contribution margin. Application of an AAA program leveraging NLP successfully identified patients with AAAs not receiving appropriate surveillance or counseling and repair. This program offers an opportunity to improve best practice-based care across a large healthcare system.