Skip to main content
. 2023 Jan 30;25:e40858. doi: 10.2196/40858

Table 2.

Comparison of demographic characteristics of participants in the mild cognitive impairment and normal cognition function subsets in the training set (579/827, 70% of the data set).

Characteristics Normal cognition function subset (n=402) Mild cognitive impairment subset (n=177) χ2 (df)/Z P value
Age (year)a 64.5 (60.00-70.00) 66.00 (62.00-71.00) 1.974 (1) .048
Sex, n (%)

0.170 (1) .68

Male 148 (36.8) 62 (35.0)

Female 254 (63.2) 115 (65.0)

Education (year)a 12.00 (9.00-14.00) 9.00 (9.00-12.00) –6.818 (1) <.001
BMI (kg/m2)a 23.30 (21.48-25.59) 23.74 (21.70-25.29) 0.137 (1) .89
Concomitant chronic disease, n (%)

4.130 (2) .12

Nil 205 (51.0) 78 (44.1)

n=1 140 (34.8) 63 (35.6)
n≥2 57 (14.2) 36 (20.3)
Smoking, n (%)

0.176 (1) .68

Yes 30 (7.5) 15 (8. 5)

No 372 (92.5) 162 (91.5)
Drinking, n (%)

0.743 (1) .39

Yes 78 (19.4) 29 (16.4)

No 324 (80.6) 148 (83.6)
Proficiency using electronic devices, n (%)b

23.156 (1) <.001

Yes 334 (83.1) 115 (65.0)

No 68 (16.9) 62 (35.0)

aA 2-sample rank test was used because the data did not conform to a normal distribution. Median values (25th percentile-75th percentile) instead of mean values were used to describe results.

bThe cutoff for yes/no was the average time of using an electronic device (eg, smartphone) for 1 hour or more/day.