Skip to main content

Table 3 The prediction results for one PD subject. Test data represent the predicted values for the outcome variables, with 95% credible intervals in parentheses

From: Predicting the multi-domain progression of Parkinson’s disease: a Bayesian multivariate generalized linear mixed-effect model

  18-month 36-month
MDS-UPDRS-II (motor)
 True value from subject 24 25
 Based on Baselinea 25.53 (20.08, 30.9) 27.63 (22.30, 33.10)
 Based on Baseline and 18-monthb   27.17 (23.18, 31.17)
MDS-UPDRS-I (non-motor)
 True value from subject 25 33
 Based on Baselinea 23.14 (17.47, 28.75) 25.86 (20.04, 31.76)
 Based on Baseline and 18-monthb   26.16 (21.32, 30.95)
Imbalance
 True value 1 0
 Based on Baselinea 1.00 (0.52, 1.00) 1.00 (0.35, 1.00)
 Based on Baseline and 18-monthb   1.00 (0.34, 1.00)
  1. a Each prediction for the 36-month outcome measure was based on baseline data alone
  2. b a combination of the baseline and 18-month data
Back to article page

BMC Medical Research Methodology

ISSN: 1471-2288

Contact us