A Study of the Psychometric Properties of the "Stanford Proxy Test for Delirium" (S-PTD): A New Screening Tool for the Detection of Delirium.

BACKGROUND: Delirium is a prevalent neuropsychiatric disorder associated with increased morbidity and mortality. Half the cases remain misdiagnosed. OBJECTIVE: Assess the effectiveness of the Stanford Proxy Test for Delirium (S-PTD) in detecting delirium in an inpatient setting. METHODS: This is a comparison study. Daily assessment with S-PTD, by the patient's nurse, and a neuropsychiatric assessment by a psychiatrist. Assessments were blinded. Inclusion criteria included 18 years or older. Exclusion criteria included patient's or surrogate's unwillingness to participate, inability to consent if a surrogate was not available, and inability to communicate in English or Spanish. A total of 309 patients were approached: 27 declined participation, 4 were excluded, and 278 subjects were followed up throughout their hospital stay. In the end, 78 were excluded for lack of neuropsychiatric assessment, S-PTD, or both. One was excluded for lack of demographic data. The sensitivity and specificity of the S-PTD in detecting delirium when compared with a neuropsychiatric assessment. RESULTS: Participants were on average 60.8 years old and 54.3% were male. Patients who developed delirium were, on average, older (15.12 y, confidence interval: 8.94-21.32). A total of 199 patients were analyzed; 43 patients (21.6%) met criteria for delirium. S-PTD detected 67 days with delirium (16.5%) of 405 hospital days, while neuropsychiatric evaluation identified 83 (20.5%). S-PTD had a sensitivity of 80.72% and a specificity of 90.37%. CONCLUSION: S-PTD is an effective, comprehensive, and simple screening tool for delirium, which is robust despite fluctuating symptoms and lack of cooperation. The use of S-PTD may enhance early diagnosis of delirium.