Alcohol availability and cirrhosis mortality rates by gender and race.

This study test whether the availability of alcoholic beverages is a simple integrated dimension as implied by certain policy models and in its treatment by researchers. Factor analysis reveals two independent availability factors: on-premise and retail availability. A correlation analysis found that on-premise availability was related to cirrhosis mortality rates for the total population, White males, non-White males, and White females. It was not related to non-White female cirrhosis mortality. In contrast, retail availability was not related to any of cirrhosis mortality rates. Examination of the states with extremes of high and low on-premise availability indicates that this type of availability is not a manipulable control variable but an index of extant norms toward drinking. It is recommended that differential prevention strategies be adopted rather than a uniform policy prevention model.