Validation of the Stanford Proxy Test for Delirium (S-PTD) among critical and noncritical patients

Background: The Stanford Proxy Test for Delirium (S-PTD) is a tool developed to be completed by nurses at the end of their shift. It was designed to use the knowledge acquired during a full shift of nurse-patient interaction. The objective of our study was to validate the S-PTD among a mixed sample of patients in both the intensive care unit (ICU) and non-ICU settings. Methods: A cross-sectional study was conducted in an ICU and three general medical wards in a tertiary care hospital. Patients were independently and blindly assessed for delirium by (1) the patients' primary nurses using the S-PTD at the end of their shift, and (2) a Consultation liaison psychiatrist who conducted a neuropsychiatric evaluation based on the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5). Results: A total 288 patients were included in current analysis. Using the S-PTD, delirium was identified in 72 (25.0%), while an expert neuropsychiatric examination, based on DSM-5 identified delirium in 75 (26.0%) patients. This study demonstrated that the S-PTD has very strong discriminative ability (area under the curve = 0.946, p < 0.001). An S-PTD cut-off score >= 3 was associated with an 82.7% sensitivity, an 95.3% specificity, an 86.1% positive predictive value, a 94.0% negative predictive value, and a 92.0% overall diagnostic accuracy. These results were similar in both ICU and general ward patients. Conclusion: The S-PTD has excellent sensitivity and specificity in detecting delirium in both ICU and ward patients, even when compared with the gold-standard, a DSM-based neuropsychiatric examination.