Skip to main content

Table 3 Factors associated with participation and adherence (final multivariable logistic regression models)

From: Disparities in the participation and adherence of older adults in lifestyle-based multidomain dementia prevention and the motivational role of perceived disease risk and intervention benefits: an observational ancillary study to a randomised controlled trial

 

Factors associated with participation (N = 1267)

Factors associated with adherence (N = 1155)

OR

95%CI

p

OR

95% CI

p

Age (years)

0.94

0.91, 0.98

0.001

-

-

-

Household monthly income

  

< 0.001

-

-

-

 < 1000€

1.00

     

 1000–1999€

2.20

1.27, 3.80

0.005

   

 20002999€

2.56

1.47, 4.47

0.001

   

 30003999€

3.91

2.07, 7.38

< 0.001

   

 > 4000€

4.42

2.12, 9.19

< 0.001

   

Town popn. size > 200 000

-

-

-

0.71

0.55, 0.92

0.009

First source of information about trial

  

< 0.001

-

-

-

 Doctor

1.00

     

 Media

1.67

1.09, 2.57

0.019

   

 Intermediarya

2.15

1.45, 3.20

< 0.001

   

Family history of AD or related disordersb

  

0.036

.

.

.

 No

1.00

     

 Yes

1.60

1.10, 2.32

0.013

   

 Do not know

1.38

0.82, 2.24

0.228

   

Perceived risk of ADc

1.73

1.30, 2.29

< 0.001

-

-

-

Perceived social supportc

.

.

.

1.21

1.03, 1.43

0.024

Emotional stabilityc

0.61

0.47, 0.79

< 0.001

-

-

-

BMI, kg/m2

N/A

N/A

N/A

  

0.040

 18.524.9

   

1.00

  

 2529.9

   

0.75

0.57, 0.98

0.035

 ≥ 30

   

0.68

0.47, 0.97

0.036

Cognitive function (composite z score)

N/A

N/A

N/A

1.37

1.13, 1.67

0.001

Depressive symptoms (GDS)

N/A

N/A

N/A

0.94

0.90, 0.99

0.024

  1. The table includes variables that remained significant in either the multivariate ‘participation’ or the multivariate ‘adherence’ model after the backwards stepwise selection procedures
  2. ‘-’ denotes variables that were included in the multivariable models, but did not remain in the final model.
  3. Education, sex, marital status, internal locus of control, and belief that memory problems are a risk factor were additionally included in the multivariable participation models, but did not remain in the final model.
  4. Sex, impact of memory problems on everyday life, belief that memory problems are a risk factor, emotional stability, external locus of control (chance), subjective memory function, altruisim as a reason for participating, functional status, and frailty were additionally included in the multivariable adherence models, but did not remain in the final model
  5. ‘.’ denotes variables that were not included in the multivariable models (p ≥ 0.20 in bivariate analysis).
  6. ‘N/A’ denotes variables that were not assessed as predictors of participation (as they were not available for non-participants)
  7. BMI, Body mass index, GDS, Geriatric Depression Scale
  8. aConferences organized by pension fund organisations, word of mouth, participants from previous studies, and via organisations such as local Alzheimer’s associations, University of the 3rd Age, sports clubs and home-help organisations; b blood relative with memory problems, AD or ‘senility’; c score/4; higher scores indicate, respectively: greater perceived risk of Alzheimer’s disease, more social support, less emotional stability (i.e. more anxious)