Distribution of Alcohol Consumption and Expenditures and the Impact of Improved Measurement on Coverage of Alcohol Sales in the 2000 National Alcohol Survey
Level 4 - case-series / case-control
Cross-sectional survey validation against aggregate sales data (design analogy).
OpenAlex W2000655659 · doi:10.1111/j.1530-0277.2007.00467.x
What was done
Researchers evaluated improved alcohol volume estimates and new expenditure questions from the 2000 US National Alcohol Survey (a random digit dialed telephone survey of 7,612 respondents, including 323 recontacted for drink ethanol measurement) by comparing aggregated survey estimates to national alcohol sales data. They used beverage-specific graduated frequency measures, empirical drink ethanol content adjustments, and past-month beverage-specific spending reports to assess the distribution of alcohol intake, heavy drinking days, and spending across drinker volume deciles.
What was found
Survey measures accounted for 52.3% of alcohol sales for consumption and 59.3% for expenditures. Wine coverage was highest (92.1% of sales), whereas spirits coverage improved from 37.2% to 55.2% after applying empirical drink ethanol adjustments. Alcohol consumption was heavily concentrated: the top 10% of drinkers consumed 55.3% of all alcohol, accounted for 61.6% of all 5+ drink days, and accounted for nearly 80% of all 12+ drink days. For spirits, the top decile consumed 62.9% of total volume. The top decile accounted for 33% of total expenditures, paying an average of $0.79 per drink compared to $4.75 per drink among the bottom 50% of drinkers.
Why it matters
The findings demonstrate extreme concentration of alcohol intake and heavy drinking days among a small fraction of drinkers who disproportionately purchase cheaper alcohol. They also show that while empirical ethanol adjustments improve survey capture of spirits consumption, self-report surveys still miss roughly 40% to 48% of total alcohol sales.
Limits
Alcohol intake and spending relied on retrospective self-report, resulting in substantial undercoverage (capturing only 52.3% of sales volume and 59.3% of expenditures). The subsample used to calibrate drink ethanol content was small (n = 323). The cross-sectional telephone design excludes institutionalized and non-telephone populations.
Cited by
- supports Approximately 10% of the population in the United States consumes about 50% of all alcohol sold.