Skip to main content

Table 2 Results of the multilevel logistic regression analysis

From: Cognitive decline and poor social relationship in older adults during COVID-19 pandemic: can information and communications technology (ICT) use helps?

 

70–79 age group

 ≥ 80 age group

 

Model 1

Model 2

Model 3

Model 4

Model 1

Model 2

Model 3

Model 4

 

Odds ratio [95% confidence intervals]

Odds ratio [95% confidence intervals]

Loneliness Score

1.28*

1.28*

1.41*

1.28*

1.30*

1.32*

0.86

1.35*

 

[1.01,1.63]

[1.00,1.62]

[1.05,1.89]

[1.00,1.62]

[1.01,1.69]

[1.01,1.73]

[0.52,1.45]

[1.03,1.78]

Social Isolation Score

0.98

0.98

0.98

0.98

1.44*

1.41 † 

1.44 † 

0.76

 

[0.71,1.36]

[0.70,1.35]

[0.71,1.35]

[0.65,1.47]

[1.00,2.06]

[0.98,2.02]

[0.99,2.10]

[0.37,1.56]

Non ICT-user

 

1.21

3.79

1.23

 

2.25*

0.17

0.36

(ref: ICT-user)

 

[0.59,2.47]

[0.51,28.35]

[0.29,5.27]

 

[1.01,5.01]

[0.01,2.02]

[0.06,2.29]

Non ICT-user × Loneliness

  

0.75

   

2.01*

 

(ref.: ICT-user × Loneliness)

  

[0.46,1.21]

   

[1.05,3.85]

 

Non ICT-user × Social isolation

   

0.99

   

2.45*

(ref.: ICT-user × Social isolation)

   

[0.52,1.90]

   

[1.05,5.75]

N

484

484

484

484

222

222

222

222

 − 2 LLR

278.18

277.92

276.52

277.92

194.24

190.2

185.08

185.5

△ − 2 LLRa

 

-0.26

-1.4

0

 

-4.04*

-5.12*

-4.7*

  1.  † p < 0.1, * p < 0.05. ICT Information and communications technology, ref Reference, LLR Likelihood ratio
  2. Models were all adjusted for gender, education, economic status, coronavirus anxiety, depression, weekly physical activities, smoking status, and comorbidities. Model 2 = Model 1 + ICT use. Model 3 = Model 2 + ICT use × Loneliness. Model 4 = Model 3 + ICT use × Social Isolation
  3. aresults of likelihood ratio tests. Differences of -2LLR between Model 1&2, Model 2&3, Model 2&4