Impact of Artificial Intelligence Assisted Drug Therapy Optimization on Drug Related Problems Amongst Older Adult Patients: A Systematic Review

Drug-Related Problems (DRPs) are common among hospitalized older adult patients and can lead longer hospital stays, worsening clinical conditions, increased healthcare costs, and even death. Artificial Intelligence (AI)-assisted interventions may aid in the optimization of medications and prevention of DRPs in this population. This review elucidates the impact of AI-assisted interventions on drug therapy optimization and identifies and prevents DRPs in hospitalized older adult patients. A comprehensive literature search was conducted in PubMed, Scopus, and Google Scholar for studies published between January 2012 to December 2022, pertaining to AI in drug therapy optimization in older adult care. The quality of the included studies was also assessed. A total of 567 relevant studies were identified, with 194 records excluded as duplicates. Following screening and eligibility determination, 13 studies were included in the review (five randomized controlled trials, two non-randomized trials, and six cohort studies). The quality ratings ranged from 79 to 100%. The studies examined clinical decision support systems and their impact on outcomes such as potentially inappropriate medications, medication appropriateness, adverse drug events, and AI recommendation acceptance rates ranging from 18 to 54%. Overall, AI tools aided in the improvement of prescribing practices; however, definitive effects on key outcomes were lacking. Alert fatigue and disobeying AI advice were significant barriers. AI interventions have the potential to improve medication safety and optimize treatment regimens for older adult patients in hospitals. However, more research is necessary to determine the true clinical significance of the CDSS and other technological interventions, especially in terms of critical clinical outcomes, such as adverse drug reaction identification and prevention, healthcare utilization, patient morbidity, and mortality.