Article Text

Download PDFPDF

038 MACE: optimal cut-offs for dementia and MCI
  1. AJ Larner
  1. Walton Centre, Liverpool

Abstract

Objective To determine optimal test cut-offs for diagnosis of dementia and MCI using the mini-Addenbrooke’s Cognitive Examination (MACE) and compare these with index study cut-offs (Dement Geriatr Cogn Disord 2015;39:1–11), and to calculate MACE predictive values across a range of disease prevalences.

Results Of 755 patients (F:M = 352:403, 47% female; median age 60 years), 114 received criterion diagnosis of dementia and 222 MCI. For diagnosis of dementia, optimal MACE cut-off determined by maximal Youden index was ≤20/30 (sensitivity 0.91, specificity 0.71) and by maximal correct classification accuracy was ≤14/30 (sensitivity 0.59, specificity 0.92). For MCI diagnosis, optimal MACE cut-off determined by maximal Youden index was ≤24/30 (sensitivity 0.90, specificity 0.57) and ≤19/30 (sensitivity 0.47, specificity 0.88) by maximal correct classification accuracy. These MACE cut-offs differed from the index study (≤25/30 high sensitivity, ≤21/30 high specificity). NPV was high (≥0.9) at all prevalences of dementia and MCI examined (range 0.05–0.4) suggesting normal MACE score effectively rules out dementia and MCI.

Conclusions In this large dataset, optimal MACE cut-offs for dementia and MCI differed from the index study reflecting the different casemix of the studies. Revision of MACE cut-offs in may be required in clinical practice to maximise test discriminative utility.

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.