Table 3.

Final model of the multilevel multiple linear regression analysis with the natural logarithm of average alcohol consumption at T2 as outcome variable among alcohol users only (N=73).^a

Fixed effects		Estimated variance	B	SE	95% CI	P
	Health counselingb		–1.11	0.92	–2.94, 0.72	.23
	Computer tailoringb		6.41	1.75	2.92, 9.90	<.001
	Not smokingc		–1.23	0.40	–2.03, –0.43	.003
	Alcohol use before pregnancy		0.00	0.03	–0.06, 0.06	.95
	Social support		0.16	0.16	–0.16, 0.47	.34
	Health counselingb * alcohol use before pregnancy		–0.05	0.05	–0.15, 0.05	.32
	Computer tailoringb * alcohol use before pregnancy		–0.43	0.12	–0.67, –0.18	.001
	Health counselingb * social support		0.39	0.24	–0.08, 0.87	.10
	Computer tailoringb * social support		–1.38	0.35	–2.08, –0.67	<.001
Random effect		0		0

^a 2 respondents were not included because they had not reported the amount of alcohol use; 6 respondents were lost because they had not filled in the question about social support.

^b Usual care is the reference category.

^c Smoking is the reference category.