The relationship among frailty, delirium and attentional tests to detect delirium: a cohort study

Background: Few studies explored if frailty predisposes to delirium in hospitalized older patients. The aims of this study were to evaluate if frailty: 1) is independently associated with delirium, and 2) affects the patient's performance in three tests of attention used to detect delirium. Methods: Data are from a prospective cohort study of patients admitted to an Acute Geriatric Unit (AGU). Frailty was operationalized using the health deficit accumulation model (38-item Frailty Index). Delirium was screened using the 4AT, and the diagnosis confirmed with the DSM-5th criteria. During the first 7 days from the hospital admission, patients also underwent a double-blind assessment of attention using three ad hoc tests (i.e., Months of the year backwards, MOTYB; Days of the week backwards, DOWB; and Count backwards from 20 to 1, CB). Results: Eighty-nine patients were included (mean age 83.1 years, standard deviation 6.0). Forty-two (47.19%) patients were frail, and 37 (41.7%) had delirium. The likelihood of delirium was significantly higher in frail compared to the non-frail patients; it was also inversely associated with the three attention tests. Using the MOTYB test, the ability to discriminate delirium was similar in patients with (Area Under the Receiving Operator Characteristic [AUROC] 0.88, 95% Confidence Interval [CI] 0.82-0.92) and without frailty (AUROC 0.93, 95%CI 0.90-0.95) whilst was markedly different between the same groups using either DOWB and CB. Conclusions: Frailty is associated with delirium in hospitalized o