Educational differences in alcohol consumption and heavy drinking: Age-Period-Cohort Perspective

Background: Reducing healthcare disparities remains a public health priority in the U.S. While there have been significant reductions in overall mortality and morbidity, disparities have widened over time pointing to significant cohort differences in health. An inverse SES-health relationship is typically found but the SES-alcohol relationship is mixed. To further our understanding of social disparities in health, this study examines educational differences in alcohol consumption and heavy drinking from an age-period-cohort (APC) perspective.

Methods: Data from seven cross-sectional National Alcohol Surveys (NAS) covering 1979 to 2010 are used to assess APC trends in alcohol consumption and heavy drinking for those with less than high school, high school degree and more than high school education levels. Past year total alcohol volume is based on combined frequency (from never to every day) and quantity (1-2, 3-4, or 5-6 drinks). Past year heavy use is total days of 5+ drinks. Gender and education-stratified models were conducted using multilevel random effects model with age clustered within periods and cohorts. Models accounted for race, income, employment, marital status, religion, and  U.S. region.

Results: Significant APC trends by education are apparent and the relationships vary by gender. The effect of education converges over time for males, but widens for females. Cohort-education effects also widen among younger cohorts with a positive effect of education on heavy drinking for both males and females.

Conclusion: Study findings enhance health disparities research by substantiating that alcohol-related disparities have increased over time and identifying which SES groups may benefit from targeted interventions.