1624-P: Cognitive Screening in Older Adults with Type 1 Diabetes: Which Tools Best Detect Potential Cognitive Decline?

Introduction and Objective: Cognitive changes in older adults with Type 1 diabetes (T1D) can impact daily self-management tasks for diabetes, but are not consistently detected. We aimed to compare performance on three cognitive screening tools against a standardised clinical neuropsychology assessment. We also evaluated whether subjective cognitive concerns could help identify individuals who need formal cognitive testing. Methods: Forty-two adults with T1D aged ≥55 years completed the Mini-Mental State Examination (MMSE), Montreal Cognitive Assessment (MoCA), Addenbrooke’s Cognitive Examination-III (ACE-III), Cognitive Change Index (CCI - self and informant versions) and neuropsychological tests, all via telehealth. Cognitive impairment was defined as ≥1.5 SD below expected levels on ≥2 neuropsychological tests, using either age-based or premorbid ability norms; using this definition as the reference standard, we performed receiver operating characteristic (ROC) analyses to evaluate screening tool performance and determine cut-off scores. Results: ACE-III performed best: ≤92/100 identified all impaired cases using age-norms. MoCA at its standard was less sensitive, and MMSE missed the most. For premorbid-norms, MoCA was most useful: ≤28/30 captured most impaired participants, whereas ACE-III required an almost perfect score for similar sensitivity, and MMSE at its standard ≤25/30 missed over 90% of those impaired. Under both impairment definitions, high CCI scores (both self and informant) showed moderate sensitivity, but required sacrificing specificity to detect subtle premorbid decline. Conclusion: ACE-III is the best initial screening tool for detecting age-normed cognitive impairment in T1D, whereas MoCA is more sensitive for detecting subtle, premorbid-normed decline. Subjective cognitive change reports are valuable for triaging further cognitive testing but cannot replace objective screening. Disclosure L. Li: None. D. Wong: None. C.A. Fisher: None. J.J. Conn: None. S. Fourlanos: Speaker's Bureau; Ended; Boehringer Ingelheim International GmbH, Novo Nordisk, AstraZeneca. Research Support; Current; Dexcom, Inc. P. Wraight: None. K. Pike: None.