Similarly, the joint effects of the tobacco policies significantly improved the model predicting per capita beer consumption (R2reduced =0.9290, R2full=0.9334, F(2,49)=6.65, p=.003). A one percent increase in price per pack was associated with a 0.106% reduction in per capita beer consumption (95% CI 0.027% to 0.186%, p=.010) and a one point increase in SFA score was associated with a 0.7% reduction in per capita beer consumption, (95% CI 0.1% to 1.2%, p=.019). For per capita wine consumption, the addition of the two tobacco policies did not significantly improved the variance explained by the model (R2reduced =0.9689, R2full=0.9695, F(2,49)=2.19, df 2, p=.124). Estimates for the individual policies were not significant, though both were in the expected direction. Finally, the joint effect of the tobacco policies significantly improved the model predicting per capita spirits consumption (R2reduced =0.9534, R2full=0.9580, F(2,49)=18.97, p<.001). A one point increase in SFA score was associated with a 1.9% reduction in per capita spirits consumption (95% CI 1.1% to 2.6%, p<.001), while the estimate for price per pack was non-significant (−0.083%; 95% CI –0.197% to 0.031%; p=.152).
