(#65) AI AIMS Screening Tool for Improved Early Detection of Tardive Dyskinesia

Abstract: Tardive dyskinesia (TD) is a medication-induced movement disorder associated with antipsychotic use. Although routine screening with the Abnormal Involuntary Movement Scale (AIMS) is recommended, it is often underused due to time constraints and workflow barriers. This quality improvement (QI) project evaluated whether increased provider use of an artificial intelligence (AI)-assisted AIMS screening tool would improve screening consistency and facilitate earlier detection of TD. A Plan-Do-Study-Act framework guided this 90-day project in an outpatient psychiatric clinic. Providers referred patients at risk for TD to a remote, video-based AI-assisted AIMS platform. Data collected included the number of screeners sent, the number of screeners completed, days to completion, and TD risk alert scores. Pre- and post-implementation outcomes were compared. Screening rates did not reach the target increase. Provider participation remained limited, and screening completion decreased (43 vs. 22). However, identification of moderate-to-high TD risk increased from 27.9% to 31.8%. Time to completion improved overall and among higher-risk patients, decreasing from 9.4 to 3.1 days. AI-assisted AIMS screening demonstrated potential to improve early detection and timeliness of TD identification; however, limited adoption and completion rates highlight the need for improved workflow integration and stakeholder engagement to optimize effectiveness.

Keywords: tardive dyskinesia, AIMS, artificial intelligence, quality improvement, early detection, telehealth

Short Description: DNP Quality Improvement Project on using an AI-assisted AIMS screening tool to improve early detection of tardive dyskinesia.

Name of Sponsoring Organization(s): N/A