Skip to main content

Table 4 Factors affecting long-term outcome with Alzheimer’s Disease Assessment Scale–Cognitive subscale as dependent variable, by age-at-onset group

From: Early- versus late-onset Alzheimer’s disease in clinical practice: cognitive and global outcomes over 3 years

 

EOAD

  

LOAD

  

Percentage of variance accounted for, all fixed terms

53.6%, p < 0.001

55.3%, p < 0.001

Significant predictors in final mixed models

β Value

95% CI

p Value

β Value

95% CI

p Value

Fixed terms

 Intercept

16.592

7.605, 25.579

< 0.001

−28.833

−47.874, −9.793

0.003

 Time in months from baseline

−0.352

−0.729, 0.026

0.067

−0.228

−0.397, −0.060

0.008

 ADAS-Cog score at baseline

0.708

0.567, 0.849

< 0.001

2.688

1.838, 3.538

< 0.001

 Time in months × ADAS-Cog score at baseline

0.022

0.013, 0.030

< 0.001

0.022

0.018, 0.026

< 0.001

 Background variables

 Sex (male = 0, female = 1)

  

NS

−2.177

−4.525, 0.171

0.069

 Sex × ADAS-Cog score at baseline

  

NS

0.150

0.043, 0.257

0.006

 Age at first assessment, years

  

NS

0.422

0.175, 0.668

0.001

 Age × ADAS-Cog score at baseline

  

NS

−0.026

−0.037, −0.015

< 0.001

 Education, years

−0.569

−1.091, −0.047

0.033

−0.073

−0.275, 0.129

0.478

 Time in months × education, years

0.041

0.009, 0.073

0.013

0.016

0.002, 0.030

0.029

 IADL score at baseline

  

NS

0.273

0.175, 0.371

< 0.001

 ChEI dosea

−0.097

−0.183, −0.011

0.027

−0.045

−0.070, −0.021

< 0.001

Random terms (variance)

 Intercept

32.457

21.150, 49.808

< 0.001

13.012

10.010, 16.915

< 0.001

 Time in months

0.145

0.096, 0.221

< 0.001

0.103

0.086, 0.124

< 0.001

  1. Abbreviations: AD Alzheimer’s disease, ADAS-Cog Alzheimer’s Disease Assessment Scale–Cognitive subscale, ChEI Cholinesterase inhibitor, EOAD Early-onset Alzheimer’s disease, IADL Instrumental Activities of Daily Living Scale, LOAD Late-onset Alzheimer’s disease, NS Not significant, PSMS Physical Self-Maintenance Scale
  2. Number of apolipoprotein E ε4 alleles, solitary living, duration of AD, PSMS score at baseline, number of medications and the specific concomitant medications used at baseline, as well as the variable comparing the ChEI agents, were not significant predictors in the models. β Values were unstandardised and are expressed per 1-unit increase for continuous variables and for the condition present in dichotomous variables
  3. aMean percentage of the maximum recommended dose, namely 10 mg for donepezil, 12 mg for rivastigmine and 24 mg for galantamine