Formal examination of the relationship between state price per pack and SFA policy scores with state per capita alcohol consumption over time was performed using linear regression with state and year fixed effects, adjusting for state unemployment rate, per capita income, age distributions, and alcohol excise taxes (Table 2). For total per capita alcohol consumption, the addition of both tobacco policies significantly improved the model compared to a model that included only the other covariates (R2reduced =0.9481, R2full=0.9532, F(2,49)=9.24, p<.001). A one percent increase in price per pack was associated with a 0.083% reduction in per capita total alcohol consumption (95% confidence interval [CI] 0.0002% to 0.166%, p=.050). A one point increase in SFA score (on a 6 point scale) was associated with a 1.1% reduction in total alcohol consumption (95% CI 0.4% to 1.7%, p=.001).
