Estimates of the Mean Alcohol Concentration of the Spirits, Wine, and Beer Sold in the United States and Per Capita Consumption: 1950 to 2002
Level 4 - case-series / case-control
Non-clinical aggregate historical time-series and market data modeling (by design analogy)
OpenAlex W2072834900 · doi:10.1111/j.1530-0277.2006.00190.x
What was done
The authors estimated the mean percent alcohol by volume (%ABV) of spirits for each US state and year from 1950 to 2002 using brand-level %ABV and sales figures by spirit type. Mean %ABVs for beer and wine across this period were extrapolated from previous estimates. These adjusted %ABVs were applied to national beverage sales figures to calculate revised yearly per capita consumption estimates of beer, wine, spirits, and total pure alcohol for the US population aged 15 and older.
What was found
The abstract reports no numerical figures or confidence intervals. It found that the mean %ABV for spirits was lower than previous estimates and varied substantially over time and across states. Applying these revised concentrations resulted in per capita consumption estimates indicating that more alcohol was consumed from beer and less from wine and spirits than previously calculated.
Why it matters
Aggregate alcohol research and econometric models depend on accurate conversions from beverage volume to absolute ethanol. Updating historical %ABV estimates by beverage category improves the accuracy and precision of aggregate time-series and cross-sectional models examining alcohol-related outcomes.
Limits
The abstract provides no quantitative figures, effect estimates, or uncertainty metrics. The %ABV data for beer and wine were extrapolated rather than directly measured from detailed historical brand sales. In addition, aggregate sales data do not capture unrecorded consumption, spillage, home production, or individual-level distribution of drinking.
Cited by
- contradicts Alcohol consumption in the United States has dropped by about 50% to its lowest level in approximately 90 years.