Skip to main content
. 2015 Nov 9;50(20):1252–1258. doi: 10.1136/bjsports-2015-095084

Table 2.

Sensitivity analysis of screen time and risk of depression in juveniles

Variable Number of reports OR (95% CI) P Value I2, p Value*
No study being excluded 21 1.12 (1.03 to 1.22) 0.007 82.5, <0.001
 CS studies 17 1.20 (1.10 to 1.30) <0.001 73.8, <0.001
 LS studies 4 0.88 (0.67 to 1.14) 0.327 90.4, <0.001
Exclude the converted studies 16 1.19 (1.11 to 1.29) <0.001 76.5, <0.001
 CS studies 15 1.21 (1.11 to 1.32) <0.001 76.2, <0.001
 LS studies 1 1.07 (1.01 to 1.14) 0.023
Exclude studies with no covariate adjustment 18 1.16 (1.10 to 1.26) <0.001 80.3, <0.001
 CS studies 15 1.21 (1.11 to 1.32) <0.001 76.2, <0.001
 LS studies 3 0.94 (0.71 to 1.26) 0.695 86.8, 0.001
Exclude studies with no adjustment for family or socioeconomic status† 16 1.19 (1.10 to 1.29) <0.001 76.3, <0.001
 CS studies 14 1.22 (1.11 to 1.34) <0.001 77.8, <0.001
 LS studies 2 1.07 (1.01 to 1.14) 0.021 0.0, 0.870
Exclude studies with no adjustment for health risk behaviour‡ 8 1.25 (1.12 to 1.40) <0.001 77.7, <0.001
 CS studies 8 1.25 (1.12 to 1.40) <0.001 77.7, <0.001
 LS studies 0

*p Value for heterogeneity.

†Family and socioeconomic status including: race/ethnicity, family structure, parental education, presence of parents, whether the family lives together, parents’ job status, socioeconomic status/household income.

‡Health risk behaviour including: physical activity, body mass index, smoking, drinking, drug abuse, fruit and vegetable intake, sleep time, other sedentary behaviours.

CS, cross-sectional; LS, longitudinal study.