Purpose: A U-shaped relation between alcohol use and mortality may be due to biological hormesis, differential response (i.e., effect modification) by health status or changes in alcohol use with ill-health and aging (i.e., reverse causality).We examined whether alcohol had the same association with mortality in healthy and unhealthy older people.
Methods: We used Cox regression analysis to examine the association of alcohol with mortality by health status in a population-based cohort of 56,167 people (65+ years), enrolled during July 1998 to December 2000 in Hong Kong.
Results: After a mean follow-up of 4.1 years, there were 3,819 deaths in 54,087 subjects. Adjusted for age, socioeconomic position and lifestyle, both occasional and moderate types of alcohol use were associated with lower mortality compared to never-drinkers, relative risk (RR) 0.72 (95% confidence interval [CI] 0.63-0.83) and 0.73 (95% CI 0.59-0.90) in men and 0.77 (95% CI 0.64-0.94) and 0.54 (95% CI 0.29-1.01) in women. However, these associations were not maintained in those with good health status: RR 1.02 (95% CI 0.74-1.39) and 1.09 (95% CI 0.71-1.68) in men and 0.63 (95% CI 0.36-1.12) and 1.27 (95% CI 0.40-4.01) in women.
Conclusions: Moderate alcohol use may only be beneficial for older adults in poor health.