JMIR Publications

ABSTRACT

Background:

Delirium is a complex neuropsychiatric syndrome with severe and lasting consequences in hospitalized patients, which are largely considered preventable. Prevention relies primarily on the early identification of at-risk patients prior to syndrome onset and the consequent initiation of preventive measures. However, the actual impact of such preventive measures is yet to be determined.

Objective:

The aim of this study was to assess the effectiveness and impact of targeted clinical pharmacist interventions (medication review) and multidisciplinary preventive non-pharmacological measures in at-risk patients, identified through artificial intelligence [AI]-based high delirium risk stratification.

Methods:

All patients admitted to a university hospital trauma ward between June 1, 2024, and May 31, 2025, were stratified by delirium risk (AI-based prediction tool). At-risk patients received a clinical pharmacist intervention, as well as multidisciplinary non-pharmacological preventive measures. As primary outcome measure, the length of hospital stay (LOS) was evaluated. Outcomes were compared to patients without delirium risk during the same year, as well as to population cohorts of the years before (June 1, 2021, until May 31, 2024). Secondarily, trends in delirium medication prescriptions were analyzed and compared to these benchmarks.

Results:

The mean LOS decreased significantly in the at-risk patient population from 8.4 days to 6.5 days. The LOS decrease significantly correlated with the number of clinical pharmacist interventions within the at-risk population. The duration of symptomatic pharmacological delirium treatment, as well as the relative number of delirium medication prescriptions also significantly decreased over time. No comparable trends were observed in patients without delirium risk (control group) in the same observation period.

Conclusions:

The delirium strategy project demonstrated consistent and clinically meaningful improvements among patients at risk of delirium (significant LOS reduction, decrease in duration and number of symptomatic delirium medication). These findings support the effectiveness of a targeted, interdisciplinary approach to delirium prevention and management, in combination with an AI-based delirium risk prediction.