P4‐291: BASELINE FLORBETAPIR AMYLOID PET STANDARD UPTAKE VALUE RATIO (SUVR) CAN PREDICT CLINICAL PROGRESSION IN PRODROMAL AD

delirium also have dementia. The 4AT (4“A”s Test: Alertness, Attention (Months of the Year Backwards), Abbreviated Mental Test-4 to test orientation; Acute change)is a short (<2 min) delirium assessment tool incorporating brief cognitive testing designed for routine clinical use which does not require special training: www.the4AT.com. Primary objective:diagnostic accuracy of the 4AT for delirium detection in acute patients aged>1⁄470. Secondary objectives: comparative performance of Confusion Assessment Method (CAM); assess performance of individual 4AT test items in the 4AT in detecting dementia; to determine if 4AT scores predict outcomes.Methods:This was a STARD-compliant, prospective, randomized, double-blind diagnostic test accuracy multi-site study of 785 patients aged >1⁄470 in the Emergency Department within 12 hours, or acute wards within 96 hours of attendance. Each patient underwent (1) DSM-IV reference standard delirium assessment informed by the Delirium Rating Scale-Revised-98, and (2) assessment with either 4AT or CAM (randomized). Results: Mean age was 81.4 (SD 6.4) years, 45% male, 9% known dementia diagnosis. 96 (11.7%) had reference standard delirium. The 4AT had an area under the receiver operating characteristic curve of 0.90. The 4AT had specificity of 95%(95% CI 92-97%) and sensitivity of 76%(95% CI 61-87%).The CAM had specificity of 100%(95% CI 98-100%) and sensitivity of 40%(95% CI 26-57%). Patients with positive 4AT had longer lengths of stay (median 5 days (IQR 2.0-14.0) than negative 4AT (median 2 days (IQR1.0 -6.0) and higher mortality. Cognitive test items of the 4AT were highly specific (AMT4 score 2: 97%(94-98%); attention score of 2: 98%(96-99%); but showed lower sensitivity (AMT4 score 2: 47% (32-62%; attention score of 2: 62% (36-83%) in detecting existing dementia. Conclusions: The 4AT is a rapid delirium assessment instrument which is feasible in routine care, including with patients with dementia, which has good diagnostic accuracy for delirium for acutely unwell older patients. Funding source: National Institute of Health Research Technology Assessment Programme (NIHR DTA) grant number 11/143/01.