Examining the Relationship between Heavy Alcohol Use and Assaults: With Adjustment for the Effects of Unmeasured Confounders

Table 2

Associations between heavy alcohol use and two outcome variables (assault and driving under the influence of alcohol). Estimates from standard logistic regression plus additional control for the effects of unmeasured confounders.

	Violent behaviour			Driving while under the influence of alcohol
	Odds ratio	95% CI^∧		Odds ratio	95% CI^∧
Days of having 5+ drinks last month
0	1.00			1.00
1	1.08	0.85	1.38	2.22^*	2.01	2.44
2	1.06	0.78	1.43	2.44^*	2.19	2.72
3 to 4	0.95	0.76	1.19	3.01^*	2.70	3.35
5 to 8	1.43^*	1.15	1.79	4.18^*	3.73	4.69
9 to 30	1.63^*	1.31	2.03	4.50^*	3.99	5.08

Model controlled for the same potential confounding factors listed in Table 1 plus additional control for the effects of unmeasured confounders derived from proxy models.
95% confidence interval.
.