Skip to main content
. 2022 May 4;10:881074. doi: 10.3389/fpubh.2022.881074

Table 3.

Stratified logistic regression analyses of PSU and mental health symptoms across education modes.

Dependent variable Subgroup No. of participants No. of symptoms Prevalence Crude ORsa Adjusted ORsb P for
(%) (95% CI) (95% CI) interaction
Depressive symptoms Total sample 1629 423 25.97
PSU 953 334 35.05 3.37(2.45-4.64) 3.14(2.26–4.37) 0.015
Non-PSU 676 89 13.17 1 1
Online group 788 236 29.95
PSU 495 205 41.41 4.31(2.64-5.67) 4.66(2.83–6.75)
Non-PSU 293 31 10.58 1 1
Face-to-face group 841 187 22.24
PSU 458 129 28.17 2.39(2.38-4.85) 2.33(2.31–4.82)
Non-PSU 383 58 15.14 1 1
Anxiety symptoms Total sample 1629 357 21.92
PSU 953 264 27.70 3.51(2.42-5.07) 3.73(2.13–4.59) 0.021
Non-PSU 676 93 13.76 1
Online group 788 198 25.13
PSU 495 163 32.93 5.67(3.46-9.09) 6.05(3.49–9.67)
Non-PSU 293 35 11.95 1 1
Face-to-face group 841 159 18.91
PSU 458 101 22.05 3.09(2.04-4.67) 2.94(1.93–4.49)
Non-PSU 383 58 15.14 1 1
Insomnia symptoms Total sample 1629 1189 72.99
PSU 953 761 79.85 2.97(2.27-3.88) 2.96(2.23–3.92) 0.589
Non-PSU 676 428 63.31 1 1
Online group 788 598 75.89
PSU 495 415 83.84 3.24(2.87-5.35) 3.29(2.87–5.72)
Non-PSU 293 183 62.46 1 1
Face-to-face group 841 591 70.27
PSU 458 346 75.55 2.87(2.13-3.87) 2.86(2.11–3.89)
Non-PSU 383 245 63.97 1 1

PSU, problematic smartphone use; OR, odds ratio.

a

Non-adjusted model.

b

Adjusted for age, sex, grade, educational level, residence, satisfaction, and exercise habits.

Participants with no PSU comprised the reference groups.