Esser · Journal of studies on alcohol and drugs 2022 · population-based cross-sectional modeling study · n=2,198,089

Improving Estimates of Alcohol-Attributable Deaths in the United States: Impact of Adjusting for the Underreporting of Alcohol Consumption.

Cited 28 times in the scientific literature.

Level 4 - case-series / case-control

Cross-sectional survey analysis combined with population-attributable fraction modeling.

PubMed 35040769 · doi:10.15288/jsad.2022.83.134 · record verified 2026-08-26

What was done

Researchers evaluated six methodological approaches to adjust self-reported average daily alcohol consumption (ADC) among U.S. adults from the 2011–2015 Behavioral Risk Factor Surveillance System (BRFSS; N = 2,198,089). Self-reported consumption was adjusted using data from the National Alcohol Survey, per capita alcohol sales data from the Alcohol Epidemiologic Data System, or both. These prevalence estimates across low, medium, and high ADC categories were then used to calculate annual alcohol-attributable deaths using a population-attributable fraction method.

What was found

Unadjusted BRFSS self-reported ADC covered 31.3% of per capita alcohol sales, rising to 36.1% using indexed-BRFSS data and 44.3% with National Alcohol Survey adjustments. Adjusting for per capita sales shifted population distributions away from low ADC and toward medium and high ADC categories. Consequently, estimated annual alcohol-attributable deaths increased from approximately 91,200 per year using unadjusted BRFSS data (Method 1) up to 125,200 per year when adjusted to 100% of per capita sales (Method 6).

Why it matters

Standard surveys substantially underreport alcohol intake compared with sales data, leading to systematic underestimation of excessive drinking and associated mortality. Incorporating sales-based adjustment models provides more realistic assessments of alcohol-related harms to inform public health policy.

Limits

The analysis relies on self-reported survey responses subject to recall and social desirability biases. Applying aggregate per capita sales data to individual consumption profiles requires modeling assumptions about the distribution of underreported intake that cannot be directly verified at the individual level.

Cited by