Artificial Intelligence to Measure Adherence to Oral Medication

The challenge of oral medication adherence in first-episode psychosis (as in any potentially chronic illness) is enormous, and numerous studies have linked non-adherence to increased rates of relapse/ hospitalization. Determining whether a patient is adhering to medication is a challenge. Pill counts, pharmacy records, technology-assisted monitoring, biological assays, and a range of self-report and interviewer-rated scales have been employed. In other areas of medicine such as antiretroviral treatment Direct Observation of Treatment (DOT) has been employed with excellent success both in monitoring and demonstrated improvement of important clinical outcomes. DOT is clearly not feasible in community settings. An artificial Intelligence (AI) platform that can be downloaded as an application onto a smart phone app represents a novel approach to offering DOT to support participant oral medication adherence and thereby improve outcomes. Aims of the project are: 1) to document acceptability to patients of the AI smartphone app based upon participation in the study and 2) to compare rates of hospitalization and emergency room visits between participants who receive the AI smartphone app with participants who receive standard care.