Skip to main content
. 2020 Jul;140:105687. doi: 10.1016/j.envint.2020.105687

Table 4.

Effects of mobile phone use at baseline (weekly minutes of conversation) in GSM and UMTS systems (first base station) on different sleep outcomes at follow-up.

Sleep disturbance Sleep adequacy Daytime somnolence Insomnia1 Sleep latency > 30 min1
Exposure indicator2 β (95% CI)3 β (95% CI)3 β (95% CI)3 OR (95% CI)3 OR (95% CI)3
All calls, UMTS call time divided by 150
<24 min 0.0 (Ref) 0.0 (Ref) 0.0 (Ref) 1.0 (Ref) 1.0 (Ref)
24–84 min −0.63 (−1.15, −0.10) −0.19 (−0.91, 0.54) 0.12 (−0.34, 0.58) 1.17 (1.01–1.36) 0.91 (0.79–1.04)
85–189 min −0.58 (−1.20, 0.04) −0.33 (−1.19, 0.54) 0.33 (−0.22, 0.88) 1.18 (0.99–1.40) 0.91 (0.78–1.07)
190 + min −1.22 (−1.95, −0.48) −0.18 (−1.20, 0.83) −0.06 (−0.71, 0.59) 1.09 (0.89–1.33) 0.88 (0.73–1.07)



Time on GSM network4
<24 min 0.0 (Ref) 0.0 (Ref) 0.0 (Ref) 1.0 (Ref) 1.0 (Ref)
24–84 min −0.64 (−1.17, −0.11) −0.16 (−0.90, 0.57) 0.12 (−0.35, 0.59) 1.17 (1.01–1.36) 0.91 (0.80–1.04)
85–189 min −0.55 (−1.18, 0.07) −0.44 (−1.30, 0.42) 0.34 (−0.21, 0.89) 1.20 (1.00–1.42) 0.93 (0.79–1.08)
190 + min −1.24 (−1.99, −0.50) −0.35 (−1.39, 0.68) −0.03 (−0.69, 0.64) 1.14 (0.93–1.40) 0.89 (0.74–1.08)



Time on UMTS network4
0 min 0.0 (Ref) 0.0 (Ref) 0.0 (Ref) 1.0 (Ref) 1.0 (Ref)
1–34 min −0.07 (−0.60, 0.46) −0.37 (−1.10, 0.36) −0.21 (−0.67, 0.26) 1.06 (0.91–1.23) 1.12 (0.98–1.29)
35–121 min 0.02 (−0.62, 0.66) −0.85 (−1.74, 0.03) 0.19 (−0.37, 0.76) 1.19 (1.00–1.42) 0.96 (0.81–1.13)
122 + min 0.21 (−0.54, 0.95) −0.87 (−1.90, 0.17) 0.45 (−0.22, 1.11) 1.21 (0.98–1.48) 1.19 (0.99–1.44)



Restricted to participants with only GSM call time5
<24 min 0.0 (Ref) 0.0 (Ref) 0.0 (Ref) 1.0 (Ref) 1.0 (Ref)
24–84 min −0.71 (−1.54, 0.11) 0.12 (−1.03, 1.27) 0.06 (−0.67, 0.80) 1.20 (0.94–1.54) 0.95 (0.76–1.18)
85–189 min −0.54 (−1.42, 0.35) −0.24 (−1.47, 1.00) 0.25 (−0.54, 1.04) 1.12 (0.86–1.46) 0.99 (0.78–1.25)
190 + min −1.94 (−2.88, −1.00) 0.00 (−1.30, 1.31) −0.11 (−0.95, 0.73) 1.08 (0.82–1.42) 0.84 (0.65–1.08)
1

Restricted to individuals who did not report the outcome at baseline

2

Categorized according to the 50th, 75th, and 90th percentiles.

3

All analyses were adjusted for age, gender, country, sleep outcome at baseline, current smoking, alcohol consumption, body mass index, educational level, weekly headache, mental and physical health score (SF-12), and diagnosis of depression.

4

GSM and UMTS calls were included in the same regression model.

5

n = 11,676.