Abstract
BACKGROUND: The alcohol harm paradox (APH) posits that alcohol harms are more prevalent in low socioeconomic position (SEP) groups compared to high SEP groups, when adjusted for alcohol consumption volume. AIM: We explored whether APH is valid for self-reported alcohol harms and whether SEP differences in hazardous drinking patterns may explain SEP differences in alcohol harms, accounting for consumption volume. DATA AND METHODS: We applied cross-sectional data from national population surveys in Norway, restricted to a subsample of past month drinkers aged 25-79 years (n = 8 171). Two binary outcome measures of self-reported alcohol-related harms were constructed from the AUDIT-10 items 4 through 10; alcohol dependence symptoms and alcohol related consequences. We separated two SEP groups based on education level: low versus high. Employing logistic regression models, we examined the extent to which SEP differences in alcohol-related harms were attributable to a more hazardous drinking pattern in terms of: (i) heavy episodic drinking and (ii) proportion of consumption volume by spirits, adjusting for consumption volume and demographic characteristics. RESULTS: In the low SEP group, alcohol-related harms were significantly more prevalent as compared to the high SEP group when adjusted for gender and age, and more so after adjustment for consumption volume. Measures of hazardous drinking pattern and having a spouse could only to a minor extent account for the elevated risk of alcohol-related harm in the low SEP group. The findings were robust across various sensitivity analyses. CONCLUSION: Low SEP was associated with alcohol-related harm and a more hazardous drinking pattern but also with lower consumption volume. The findings support the validity of the alcohol harm paradox for self-reported alcohol harms.