Effects of Implementation of a Standardized, Interdisciplinary Algorithm on Prevalence, Treatment, and Outcomes of Delirium in Medical Inpatients.

Background
With the aim to optimize delirium management, an algorithm was introduced in general internal medicine wards of a Swiss University Hospital in 2018. We evaluated its impact on prevalence, therapeutic measures, and outcomes of delirium in a real-world setting.Methods
Retrospective before/after study of patients aged ≥ 65 years hospitalized in general internal medicine. During 2018, an interprofessional delirium algorithm was introduced. We compared the delirium prevalence and secondary outcomes including mortality, length of stay (LoS), discharge home, delirium duration, and use of pharmacological and non-pharmacological interventions between the pre-intervention (2017) and the post-intervention (2019) periods in unadjusted and adjusted analyses.Key Results
Of 4855 cases (pre 2253/post 2602), the overall delirium prevalence was 10.3% (n = 499); the prevalence in 2017 was 9.5% (n = 213) and 2019 11.0% (n = 286) (p = 0.08, adjusted OR 1.04, 95% CI 0.93 to 1.15). The delirium duration was similar in the pre- and post-intervention periods (adjusted mean difference -0.61 days, 95% CI -2.35 to 0.12). The use of antipsychotic medications decreased (adjusted OR 0.45, 95% CI 0.23 to 0.88) and non-pharmacological interventions were increasingly used (involvement of relatives adjusted OR 16.59, 95% CI 4.74 to 58.07; delirium screening instruments adjusted OR 25.22, 95% CI 7.24 to 87.93).Conclusion
The introduction of an interdisciplinary delirium algorithm did not change delirium prevalence, but was associated with a decreased use of antipsychotic medications and an increased use of non-pharmacological interventions. However, there is further need for quality improvement to unlock the full potential to prevent and treat delirium in the acute hospital wards.