Table 3.

Multivariate associations between non-drinking as the dependent variable and computer games when divided into weekdays only, weekends only and both weekdays and weekends

	Year 2008	Year 2010	Year 2012
	OR (CI)	OR (CI)	OR (CI)
Gender (boys as reference)	0.98 (0.78–1.23)	0.81 (0.64–1.03)	0.84 (0.64–1.12)
Social media/chatting	0.57 (0.45–0.73)	0.57 (0.44–0.72)	0.53 (0.41–0.67)
Computer games weekdays only	0.54 (0.27–1.08)	0.78 (0.48–1.28)	1.14 (0.66–1.95)
Computer games weekends only	1.07 (0.76–1.53)	1.14 (0.83–1.57)	1.74 (1.13–2.69)
Computer games weekdays and weekends	1.23 (0.95–1.58)	1.12 (0.86–1.46)	1.82 (1.31–2.54)

Note. Chatting ≥ 2 hours.day. Playing computer games is divided into four groups: ≥2 hours/day on weekdays or weekends; ≥2 hours/day on weekdays only, ≥ 2 hours on weekends only, or ≥ 2 hours on both weekdays and weekends. The models are adjusted for covariates including country of birth, family substance problems, conflicts between parents, physical abuse, number of friends, status among friends, psychosomatic problems, sleeping problems, use of cannabis and use of other drugs.