The majority of AAF for chronic conditions are calculated by ARDI on the basis of relative risk estimates from meta-analyses and the prevalence of alcohol use at specified risk levels (7,8). Self-reported alcohol use from the Behavioral Risk Factor Surveillance System (BRFSS) (9) was used to capture drinking at levels specified by the meta-analyses, which use slightly higher cut-points for risky drinking than those more commonly used in the United States. For the majority of acute conditions (ie, injuries), ARDI includes a direct estimate of the AAF. AAF for these conditions is based on studies assessing the proportion of deaths from a particular condition that occurred at or above a blood alcohol level of 0.10 g/dL (10). In addition, certain conditions (eg, alcoholic cirrhosis of the liver) are by definition 100% alcohol-attributable and therefore did not need to be estimated. To calculate YPLL attributable to excessive alcohol consumption, the age- and sex-specific AAD estimates for each cause were multiplied by the corresponding estimate of life expectancy based on the age and sex of the decedent.
