MDS Abstracts

Abstracts from the International Congress of Parkinson’s and Movement Disorders.

MENU 
  • Home
  • Meetings Archive
    • All Meetings
    • 2026 International Congress
  • Keyword Index
  • Resources
  • Advanced Search

Clinical predictors of screen-defined dementia in early Parkinson’s

F. Baig, M. Lawton, M. Rolinski, C. Ruffmann, T. Barber, J. Klein, C. Lo, Y. Ben-Shlomo, M. Hu (Oxford, United Kingdom)

Meeting: 2018 International Congress

Abstract Number: 1256

Keywords: Parkinsonism dementia complex(PDC)

Session Information

Date: Monday, October 8, 2018

Session Title: Parkinson's Disease: Cognition

Session Time: 1:15pm-2:45pm

Location: Hall 3FG

Objective: Predicting early dementia in Parkinson’s disease (PD) has important implications for individual prognosis, designing clinical trials and targeting novel treatments, but there remains a lack of evidence in this area.

Background: This study examined which clinical factors predicted early dementia in a large cohort of early PD subjects.

Methods: Parkinson’s patients assessed within 3.5 years of diagnosis were recruited between 2010-2015 (the Discovery cohort, UK) and then re-assessed after 18 months. The Montreal cognitive assessment was used to assess cognition, using a score of <23 for screen-defined dementia. A broad spectrum of other motor and non-motor symptoms were also assessed. A logistic regression model with backwards stepwise selection was used to determine which baseline clinical assessments were independent predictors of dementia at 18 months.

Results: 61 of the 488 included PD patients developed new dementia at 18 month follow-up. Older age at diagnosis (OR 1.65 (95% CI 1.26-2.17), p<0.001) with poor performance on phonemic fluency (OR 4.03 (95% CI 1.87-8.70), p<0.001), cube copying (OR 1.94 (95% CI 1.31-2.86), p=0.001), and the Purdue assembly test (3.66 (95% CI 1.86-7.21), p<0.001) were all included in the final model as independent predictors of dementia (see figure 1). Figure 1 - Odds ratios of model for factors predicting dementia, with their 95% confidence intervals. Age at diagnosis was analysed as a 5 level ordinal variable, cube copying errors was analysed as a 3 level ordinal variable, and both phonemic fluency and purdue assembly were analysed as a dichotomous variables (poor performers <20th gentile). The area under the ROC curve for this model is estimated at 0.81 (95% CI 0.74-0.89) (see figure 2). Figure 2 - ROC curve depicting model of factors predicting dementia including age at diagnosis, Purdue assembly task, phonemic fluency and cube copying connection errors.

Conclusions: Poor performance on three simple clinical tests performed early in PD (the Purdue assembly task, phonemic fluency and cube copying) can be used to predict early dementia. This has implications for both clinical practice and clinical trials.

To cite this abstract in AMA style:

F. Baig, M. Lawton, M. Rolinski, C. Ruffmann, T. Barber, J. Klein, C. Lo, Y. Ben-Shlomo, M. Hu. Clinical predictors of screen-defined dementia in early Parkinson’s [abstract]. Mov Disord. 2018; 33 (suppl 2). https://www.mdsabstracts.org/abstract/clinical-predictors-of-screen-defined-dementia-in-early-parkinsons/. Accessed October 2, 2026.
  • Tweet
  • Email a link to a friend (Opens in new window) Email
  • Print (Opens in new window) Print

« Back to 2018 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/clinical-predictors-of-screen-defined-dementia-in-early-parkinsons/

Related Sites

International Parkinson and Movement Disorder Society

The Society that manages the annual International Congress »

International Congress

The official website for the International Congress of Parkinson’s and Movement Disorders® »

  • Help & Support
  • About Us
  • Cookies & Privacy
  • Wiley Job Network
  • Terms & Conditions
  • Advertisers & Agents
Copyright © 2026 International Parkinson and Movement Disorder Society. All Rights Reserved.
Wiley