Aims: This study aimed to test whether the alcohol harm paradox (AHP) is observed in Brazil by investigating (i) the association between educational attainment and alcohol-related consequences (ARC) and (ii) the contribution of average alcohol volume consumed (AVC), past-month heavy episodic drinking (HED), smoking, body mass index (BMI), and depression in accounting for the disparities in ARC.
Methods: We analysed data from the 2019 Brazilian National Health Survey, a nationally representative household survey. The composite ARC outcome was considered present when an individual reported a past-year episode of activity failure, amnesia, and concern by others due to alcohol consumption. Adjusted binary logistic regression models were fitted using a hierarchical approach to calculate the odds ratios (OR) and respective 95% confidence intervals (CI), and to assess the contribution of each set of variables in attenuating the educational differences in ARC.
Results: Those from the lowest educational strata (incomplete elementary school) exhibited higher odds of ARC than their counterparts (OR: 2.03; 95% CI: 1.73-2.37). Although smoking, BMI, and depression attenuated the educational gradient (i.e. reduced the difference between reference and riskier categories) in ARC by ~13%, the adjustment for AVC and HED amplified inequalities by 0.3% and 5.7%, respectively.
Conclusion: We found evidence of the AHP in Brazil. Educational inequalities in ARC were scarcely attenuated by behavioural factors, and a suppression effect was noted when adjusting for AVC and HED.
Keywords: alcohol drinking; alcoholism; socioeconomic factors.
© The Author(s) 2024. Medical Council on Alcohol and Oxford University Press. All rights reserved. For permissions, please email: journals.permissions@oup.com.