Table 4.

Logistic regression analyses of associations between ordinal levels of alcohol consumption^a and negative school-related outcomes.

	Low GPAb		High number of days missed from schoolb		High number of hours missed from schoolb
	OR	95% CI	OR	95% CI	OR	95% CI
CRUDE MODEL
Never consumed alcohol	(Base)		(Base)		(Base)
Non-consumption	1.42**	1.14, 1.76	1.61**	1.19, 2.19	1.74**	1.25, 2.42
0.1–19.9th percentile	1.21	0.96, 1.52	1.59**	1.14, 2.21	1.88***	1.32, 2.67
20–49.9th percentile	1.99***	1.59, 2.48	1.97***	1.49, 2.62	2.64***	1.97, 3.55
50–79.9th percentile	3.24***	2.50, 4.21	2.77***	2.12, 3.64	4.03***	3.04, 5.34
80–89.9th percentile	3.28***	2.15, 5.01	3.77***	2.70, 5.27	5.09***	3.61, 7.18
90–100th percentile	2.97***	2.01, 4.40	3.05***	2.17, 4.29	6.05***	4.37, 8.40
FULLY ADJUSTEDc
Never consumed alcohol	(Base)		(Base)		(Base)
Non-consumption	1.35**	1.08, 1.69	1.36	0.99, 1.88	1.49*	1.06, 2.10
0.1–19.9th percentile	1.34*	1.05, 1.72	1.30	0.91, 1.84	1.60*	1.12, 2.30
20–49.9th percentile	2.01***	1.58, 2.56	1.50**	1.11, 2.03	1.98***	1.45, 2.72
50–79.9th percentile	3.14***	2.37, 4.17	1.83***	1.36, 2.46	2.83***	2.09, 3.83
80–89.9th percentile	2.82***	1.82, 4.38	2.50***	1.74, 3.59	3.50***	2.43, 5.04
90–100th percentile	2.62***	1.75, 3.94	2.06***	1.42, 2.98	4.01***	2.81, 5.71

N = 7,874 (girls n = 4,121, boys, n = 3,753).

^aPresented alcohol level consumption percentiles are calculated among those adolescents who report to have an actual alcohol consumption.

^bp-value for trend in the association between alcohol variable and school-related variable: all p < 0.001.

^cAdjusted for the confounding of age, gender, SES and mental health problems.

Bold fonts denotes statistically significant associations:

^***p < 0.001,

^**p < 0.01,

^*p < 0.05.