Table 1.

Multivariate regression model on pathological Internet use (PIU) with age, gender, psychopathological scores and self-destructive behaviour categories as explaining variables

Psychopathology and self-destructive behaviours	Pathological Internet use
	Coefficient	95 % CI	Standardized coefficienta	p value
Age	−0.014	−0.048 to 0.020	−0.009	0.426
Female gender	−0.086	−0.141 to −0.032	–	0.002
Depression (BDI-score)	0.032	0.027 to 0.037	0.166	<0.001
Anxiety (Z-SAS-score)	0.033	0.028 to 0.038	0.169	<0.001
Emotional symptoms (SDQ-subscore)	−0.002	−0.017 to 0.013	−0.003	0.825
Conduct problems (SDQ-subscore)	0.059	0.041 to 0.077	0.064	<0.001
Hyperactivity and/or inattention (SDQ-subscore)	0.057	0.044 to 0.070	0.084	<0.001
Peer relationship problems (SDQ-subscore)	−0.007	−0.024 to 0.009	−0.008	0.375
Pro-social behaviour (SDQ-subscore)	−0.011	−0.026 to 0.003	−0.015	0.121
Self-injurious behaviour	0.132	0.027 to 0.237	–	0.014
Suicidal ideation	0.314	0.241 to 0.387	–	<0.001
Suicide attempts	0.530	0.185 to 0.875	–	0.003
Random-effects parameters
Country (variance of intercept)	0.044	0.177 to 0.111
School (variance of intercept)	0.032	0.022 to 0.049

Young’s Diagnostic Questionnaire (YDQ)

Model Wald χ ²(12) = 2,605.9, p < 0.0001; LR test versus linear regression χ ²(2) = 349.1, p < 0.0001, BIC = 37,976,94

^aThe standardized coefficient of continuous predictor x was calculated as coef(x) × SD(x)/SD(y)