Researchers say the positive association between alcohol use and mortality found in previous studies "may be attributable to an inappropriate selection of a referent group and weak adjustment for confounders."
More than half of us are regular drinkers, defined by the Centers for Disease Control and Prevention (CDC) as consuming at least 12 alcoholic beverages in the past year.
The health implications of heavy drinking are well documented. It has been associated with high blood pressure, stroke, liver diseases and cancer, among other conditions. But what about alcohol consumption in moderation? Is the odd glass of wine at the end of a hard day's work bad for us?
Many studies suggest not. Last month, for example, research published in the European Heart Journal claimed drinking up to seven alcoholic drinks a week could protect against heart failure, while a study published in October 2014 found light alcohol use later in life may improve memory.
According to the background of this latest research, however, many of the studies associating alcohol use with health benefits are "contentious," and the protective effects of alcohol may be confounded by categorizing former drinkers and never drinkers - who are often used as control participants - into one group.
"Specifically, former drinkers have been found to exhibit poorer self-reported health, higher levels of depression and increased risk of mortality than never drinkers," note the authors, including Craig S. Knott of University College London in the UK.
"As such, protective associations identified among light drinkers may be less a consequence of a beneficial biological mechanism and more a statistical artifact resulting from the application of a pooled non-drinking category."
In addition, the researchers note that very few studies that hail the benefits of alcohol consumption have included older participants.
With these points in mind, Knott and colleagues set out to gain a better understanding of the effects of alcohol consumption on mortality among two age groups: 50-64 and 65 and older.
Selection bias, poor controls 'may explain positive link between alcohol and mortality'
Using 1998-2008 data from the Health Survey for England - which provided information on participants' self-reported alcohol consumption and mortality - the researchers assessed 18,368 adults aged 50-64 years and 34,523 adults over the age of 65.
Participants who reported consuming alcohol at some point over the past 12 months - "occasional drinkers" - were divided into groups based on their weekly alcohol consumption. Adults who reported consuming alcohol in the past week - deemed "current drinkers" - were split into groups based on their daily alcohol consumption.
From unadjusted models, the researchers identified reduced mortality risk across a wide range of alcohol usage among men and women in both age groups.
But after excluding former drinkers and accounting for other influential factors - such as socioeconomic status and lifestyle - the researchers only identified significant mortality benefits among men aged 50-64 who consumed 15-20 units of alcohol a week or up to 1.5 units on the heaviest day, and women aged 65 and over who drank less than 10 units a week or up to 4.5 units on the heaviest day.
The researchers say their results "may have better isolated the true effect of alcohol consumption on mortality." They add:
"The findings from this study suggest that beneficial associations typically identified between low-intensity alcohol consumption and all-cause mortality may, in part, be attributable to an inappropriate selection of a referent group and weak adjustment for confounders.
The effect of such biases should therefore be borne in mind when evaluating findings from alcohol health studies - particularly when seeking to extrapolate results to the population level."
In an editorial linked to the study, Prof. Mike Daube, of Curtin University in Australia, says there are a number of conclusions that should be drawn from this research.
He says the results suggest health professionals should "discourage" findings indicating that alcohol use - even at low levels - has benefits for cardiovascular disease and mortality. Furthermore, he says the alcohol industry should refrain from using such findings to promote their products.
"Globally, more than 3 million deaths each year are attributable to alcohol," he notes. "The real mortality benefits will come from determined action at the political level, not outdated advice and wishful thinking."
In November 2014, another study questioned the benefits of moderate alcohol consumption after finding that - contrary to some previous studies - it only protects against coronary heart disease in 15% of the population.