Skip to main content
. 2020 Dec 21;22(12):e19767. doi: 10.2196/19767

Table 6.

Logistic regression analysis across different risk groups (P<.001).

Characteristic Medium-risk versus high-riska, ORb (95% CI) Medium- and low-risk versus high-riskc, OR (95% CI)
Education level (ref: junior high school or less)d


Senior high school 4.65 (1.54-13.99) 4.50 (1.50-13.52)

University/college 13.22 (4.24-41.23) 13.46 (4.32-41.93)

Graduate school 18.40 (5.03-67.36) 19.11 (5.23-69.90)
Income (New Taiwan $) (ref: $0)e

≤$23,800 0.90 (0.37-2.20) 0.93 (0.38-2.28)

$23,801-$30,000 4.06 (1.44-11.45) 3.86 (1.37-10.87)

$30,001-$50,000 4.08 (1.72-9.69) 3.78 (1.59-8.95)

$50,001-$70,000 4.23 (1.56-11.47) 3.92 (1.45-10.59)

$70,001-$100,000 2.18 (0.75-6.33) 2.16 (0.75-6.23)

≥$100,001 1.76 (0.58-5.36) 1.62 (0.54-4.90)
Level of internet use (ref: never/rarely)

Several times per week/month 9.19 (2.61-32.37) 9.05 (2.57-31.84)

Several times per day 3.80 (1.71-8.46) 3.95 (1.77-8.79)
Health status (ref: poor)

Normal 1.93 (0.90-4.14) 2.00 (0.93-4.31)

Good 2.98 (1.30-6.83) 3.22 (1.40-7.39)

a–2 log-likelihood: 564.251; Nagelkerke R2: 0.311.

bOR: odds ratio.

c–2 log-likelihood: 576.339; Nagelkerke R2: 0.309.

dref: reference.

eA currency exchange rate of NT $1=US $0.035 is applicable.