18A SERVICE EVALUATION OF THE PERFORMANCE OF THE 4-AT AS A COGNITIVE SCREENING TOOL IN AN ENGLISH UNIVERSITY HOSPITAL

Abstract

Introduction: Delirium is a common and serious disorder that is under-recognised in hospitalised patients. The 4-AT is a validated tool for detecting delirium. Since 2017, our hospital has routinely used the 4-AT under the national dementia CQUIN to screen nonelective admissions aged >75 without a known history of dementia. Our aim was to evaluate the feasibility of the 4-AT in our setting and study its association with patient characteristics and outcomes. Methods: We conducted a retrospective service evaluation in an English university hospital. We included all last admissions of non-elective patients aged >75 between 1st January 2017 and 2nd December 2017. Routinely measured patient characteristics included the 4-AT, demographics (age and sex), clinical frailty scale (CFS), acute illness severity (National Early Warning Score in the Emergency Department), Charlson Comorbidity Index (CCI), and discharge specialty (geriatric medicine, general medicine, surgery). Outcomes studied were: having an inpatient Older People's Mental Health (OPMH) team assessment, inpatient mortality, death within 30 days of discharge, new institutionalisation, length of stay (LOS) >7 days, delayed discharge, and readmission within 30 days. Statistical analyses were based on bivariate comparisons and logistic regression models. Results: Of 12,800 hospital episodes over the period, 8,307 were last admissions. 1,305 did not have a 4-AT as they had known dementia. Of the remaining, 3,633 had a 4-AT completed (52% of the eligible sample). Non-compliance seemed to be associated with higher frailty and acuity, and with shorter LOS. 84.0% had a 4-AT score of 0 (normal), 11.3% had a score of 1-3 (intermediate) and the remaining 4.7% a score of >4 (abnormal). There was an increasing proportion of inpatient OPMH assessment across 4AT groups (14.5%, 24.1% and 42.4%, respectively). After controlling for age, sex, CFS, CCI, acute illness severity and discharge specialty, the 4AT was an independent predictor of: inpatient mortality (OR 1.11, 95% CI 1.03-1.20, p = 0.005), 30-day post-discharge mortality (OR 1.15, 95% CI 1.04-1.28, p = 0.009), delayed discharge (OR 1.09, 95% CI 1.01-1.16, p = 0.018) and discharge to usual place of residence (OR 0.91, 95% CI 0.85-0.98, p = 0.013). Conclusions: Our evaluation suggests that the 4-AT is a robust tool to detect delirium in the acute hospital and has significant independent associations with hospital outcomes. Quality improvement work is needed to increase compliance rates.