Background:
Moderate alcohol consumers have a reduced risk for cardiovascular and all-cause mortality. Alcohol intake improves iron absorption and also has a profound effect on iron metabolism and thus could in part explain the observed inverse association between moderate alcohol intake and mortality. We sought to investigate whether moderate alcohol confers mortality benefits in part through improvement in iron status.
Methods:
Publicly available data from two consecutive National Health and Nutrition Examination Survey (NHANES) cycles (1999/2000 and 2001/2002) were obtained and linked to public data on all-cause mortality. Study participants were grouped as never, past, moderate (≤2 drinks/day for men, ≤ 1 drink/day for women) and heavy drinkers (>2 drinks/day for men, >1 drink/day for women). To assess the quality of alcohol data in NHANES, we tested whether self-reported alcohol consumption was associated with biomarkers of alcohol intake (HDL-C and γ-glutamyl transferase (GGT)). Cox-models, weighted using four-year sampling weights, were fitted to determine whether alcohol intake was associated with all-cause mortality. Measures of iron status, particularly serum ferritin, transferrin saturation, hemoglobin and free erythrocyte protoporphyrin were evaluated as potential mediators of the association between alcohol consumption and risk of all-cause mortality.
Results:
Among 7,532 men and women with complete data, 17% were never drinkers, 20% were past drinkers, 30% were moderate drinkers and 33% were heavy drinkers. We found an increase in HDL-C and GGT with increased alcohol intake, suggesting that self-reported alcohol intake is reliable in this population. In the weighted analysis, the hazard ratio (95% CI) for all-cause mortality among moderate alcohol users compared to never users was 0.56 (0.37-0.85), in models adjusted for age, race, smoking, statin use and history of diabetes, among other variables. Adjustment for transferrin saturation as a measure of iron status attenuated the benefit from moderate alcohol (HR = 0.69; 95% CI: 0.41-1.14), suggesting that moderate alcohol consumption may in part confer a benefit on mortality through improvement in iron status. These results warrant further evaluation in rigorous formal mediation analyses.
Conclusions:
Moderate alcohol consumption was associated with higher transferrin saturation and with reduced risk of mortality. Adjusting for iron status attenuated the association between moderate alcohol consumption and all-cause mortality suggesting that the effects of moderate alcohol on mortality may in part be via improvement in iron status.