Our findings for physical abuse are more difficult to interpret. In this case, the direct positive association between physical abuse and alcohol dependence severity is dampened by a negative indirect effect resulting from an inverse association between physical abuse and withdrawal anxiety symptoms. The commonly held belief is that childhood physical abuse increases risk for anxiety and depression. However, many of the studies linking physical abuse and anxiety fail to account for important confounders such demographics, parental history of anxiety, more current stressors, and other comorbid psychiatric disorders (Cougle et al., 2010; Fuller-Thomson et al., 2012). Even though we control for some of these factors in our analyses, the finding of a negative association between physical abuse and withdrawal anxiety was unexpected. Anxiety during acute withdrawal is a heavily state dependent measure that is influenced by a number of factors, including perhaps overall withdrawal severity. However, including the maximum CIWA score for each individual obtained within the first couple of days of acute withdrawal as a potential mediator did not change the results (data not shown). Furthermore, analyzing anxiety and
