Cut-off scores for mild and moderate dementia on the Addenbrooke's Cognitive Examination-III and the Mini-Addenbrooke's Cognitive Examination compared with the Mini-Mental State Examination.

Aims and method: We aimed to establish cut-off scores to stage dementia on the Addenbrooke's Cognitive Examination-III (ACE-III) and the Mini-Addenbrooke's Cognitive Examination (M-ACE) compared with scores traditionally used with the Mini-Mental State Examination (MMSE). Our cross-sectional study recruited 80 patients and carers from secondary care services in the UK.

Results: A score ≤76 on the ACE-III and ≤19 on the M-ACE correlated well with MMSE cut-offs for mild dementia, with a good fit on the receiver operating characteristic analysis for both the ACE-III and M-ACE. The cut-off for moderate dementia had lower sensitivity and specificity. There were low to moderate correlations between the cognitive scales and scales for everyday functioning and behaviour.

Clinical implications: Our findings allow an objective interpretation of scores on the ACE-III and the M-ACE relative to the MMSE, which may be helpful for clinical services and research trials.