Detecting delirium in Parkinson's disease: an evaluation of diagnostic accuracy of bedside tools.

Background: Delirium is a serious, acute neuropsychiatric condition associated with fluctuating attention and altered arousal. Delirium in Parkinson's disease (PD) is common but often missed in hospital due to shared clinical features. This study aimed to evaluate the accuracy of current tools used to identify delirium in inpatients with PD.

Methods: People with PD admitted to all hospital wards were invited to take part. Participants completed a standardised delirium assessment based on the Diagnostic and Statistical Manual of Mental Disorders 5th Edition (DSM-5) criteria, in addition to standard bedside tools including the 4 As Test (4AT), arousal and cognition. This was a secondary analysis of a prospective observational study; bedside tools were not completed independently of, or blinded to, the DSM-5 criteria. Accuracy was assessed using Receiver Operating Characteristic area under the curve (AUROC).

Results: Participants included 115 people with PD (200 hospital admissions); 66.1% (n = 76/115) had delirium. Considering all admissions, the diagnostic accuracy of tools was good, ranging from 74% to 89% (AUROC = 0.764-0.923, P < .001 for all). The 4AT scores had the highest sensitivity (96.7%, AUROC = 0.922, P < .001). However, accuracy decreased in those with underlying cognitive impairment (AUROC = 0.499-0.886).

Conclusions: Current bedside tools can accurately identify delirium in PD inpatients. Although tools were comparable, the 4AT may have greater clinical utility as it had high sensitivity, is quicker to complete and already widely used clinical. However, caution is recommended as tools did not differentiate between symptoms typical in PD and acute symptoms associated with delirium; this should be a focus for future research.