Large study shows link between alcohol consumption in later life and mortality

Reviewed by Kate Anderton, B.Sc. (Editor)Jul 5 2019

Alcohol consumption in later life has increased over the past decade. Although moderate alcohol intake in older adults has been previously linked to a lower risk of cardiovascular disease and death, recent studies have suggested little ─if any ─health benefit to alcohol. Assessing the relationship between alcohol intake and mortality is extremely challenging, partly because of the need to disentangle the effect of alcohol from that of other factors that influence health, and also because people's drinking habits often change over time. However, research methodology and data quality continue to improve. A new report in the journal Alcoholism: Clinical & Experimental Research presents a 16-year follow-up of the Health and Retirement Study (HRS) ─one of the largest and most rigorous US studies of the relationship between alcohol consumption and mortality to date.

The study involves almost 8000 older adults (born between 1931 and 1941) who have provided information on their drinking habits since 1992, with biannual interviews from 1998 to 2014. This contrasts with many previous studies that assessed alcohol intake at just a single point in time. At each assessment timepoint, participants were categorized into one of five groups for analysis: lifetime abstainers, current abstainers, heavy drinkers, moderate drinkers, and occasional drinkers. By also tracking any deaths that occurred, the researches could examine the relationship between alcohol intake and mortality over the assessment timeframe. Importantly, the analyses accounted for a wide range of additional factors (termed 'confounders') that can influence health outcomes ─including those that vary over time (such as level of household assets, smoking, BMI, health/functioning, depression and chronic disease) as well as static factors such as age, education, sex and race.

The researchers found that, overall, moderate or occasional drinkers had lower mortality rates than people who abstain from alcohol, after accounting for confounding factors, but assessments of the potential for residual bias suggest we should still be cautious in our conclusions. Indeed, 'current abstainers' (who had reported drinking in the past) had the highest mortality rate, and this was probably indicative of a 'reverse causation' effect whereby people stop drinking at the onset of ill health. Further, mortality among moderate/occasional drinkers was also generally lower than among lifetime alcohol abstainers, at least in women. As might be expected, the mortality benefit of drinking was lower among smokers than non-smokers, and the mortality reduction was also lower for men than women.

Despite finding associations between moderate/occasional drinking and reduced mortality, the researchers urge caution, given that even high-quality observational studies such as the HRS can be subject to biases, errors in measurement. and the risk of as yet unknown confounding factors. As such, they conclude that continued research, with additional and further improved data sources, is required to characterize the changing patterns of drinking among older adults and the role of alcohol in promoting or harming health in later life.

Lifetime abstainers: less than 12 drinks in their lifetime. Current abstainers did not drink in the current assessment period, but drank (or may have drank) in the past. Heavy drinkers were defined as consuming more than three (men) or two (women) drinks per day or binging more than five (men) or four (women) drinks in a single day. Moderate drinkers were defined as drinking one to two (women) or one to three (men) drinks one or more days per week, and not binging more than five (men) or four (women) drinks in a single day. Occasional drinkers were defined as drinking less than one day per week, not binging more than five (men) or four (women) drinks in a single day, and drinking a maximum of three (men) or two (women) drinks per day.