. Author manuscript; available in PMC: 2020 Jun 1.

Published in final edited form as: Addiction. 2019 Mar 3;114(6):957–967. doi: 10.1111/add.14578

Table 2.

Effect Size and Heterogeneity Estimates from Random Effects Models by Alcohol-related behaviors

Alcohol-Related Behaviors	k	r (95% CI)	Q	I²
Alcohol Consumption	9	.12 (0.08, 0.17)^***	15.79^*	49.35
Heavy/binge Drinking	12	.06 (0.02, 0.10)^**	15.03	26.80
At-Risk Drinking	4	.14 (0.06, 0.23)^**	2.43	0.00
AUD	3	.10 (−0.01, 0.20)	20.76^***	90.37
Negative Drinking Consequences	5	.25 (0.09, 0.42)^**	65.99^***	94.00

Note. The Alcohol Consumption variable combines all effect sizes of alcohol use quantity, alcohol use frequency, and quantity*frequency measures. The At-Risk Drinking variable combines all effect sizes of the CAGE and AUDIT-C measures. A leave-one-out sensitivity analysis [35] was performed by iteratively removing 1 study at a time to confirm that our findings were not driven by any single study. All effect sizes remained stable across variables, with one exception; the association of AUD DSM-IV Criteria with racial discrimination became significant (r = .15 [0.10, 0.20], p < .05) upon removal of Hunte & Barry, 2012. CI: confidence interval.

p < .05.

^**

p < .01.

^***

p < .001.