Development and evaluation of a concise nurse-driven non-pharmacological delirium reduction workflow for hospitalized patients: An interrupted time series study.

The study was done on one hybrid geriatric and general medicine unit at one academic medical center, so other hospitals may get different results. Delirium was measured with a nurse screening score, not a diagnostic delirium assessment. The screening tool has high specificity but low sensitivity, 98% and 42%, so some delirium cases may have been missed or counted differently. The post-intervention period overlapped COVID-19, patient characteristics changed, and workflow fidelity fell, so the pandemic and the workflow cannot be cleanly separated. Audit coverage was low overall, 54% during the workflow period and 36% after it, so many patient days were not checked for workflow delivery. The intervention was a bundle of five morning and five evening components, so the study cannot show which component, if any, caused the change. The study was not randomized, so unmeasured changes over time could have influenced outcomes.

The supplied text does not include an author conflicts statement; the publication types list NIH extramural and non-U.S. government research support.

Do not conclude that this workflow reliably prevents delirium or shortens hospital stays. The initial drop in delirium screening days was not sustained after COVID-19, length of stay did not change, and the study used a screening score rather than a delirium diagnosis.