“Sick quitters” versus “under-reporters”
The association between the intake of alcoholic beverages and cardiovascular disease in epidemiological studies could have 2 confounding factors that distort the risk, thus, the abstainers as a reference group as well as underreporting of the actual amount of alcoholic beverages consumed. New methods are available to determine the ethanol consumption and the “sick quitters” are controlled for.
Meta-analyses of cohort studies show a J-shaped relationship between the consumption of alcoholic beverages and cardiovascular as well as all-cause mortality, with the lowest risk for light to moderate intake.
However, some experts argue that moderate consumption may only appear healthy because the “abstainers” as reference group to whom moderate drinkers are compared, may include many individuals who may have stopped drinking for health reasons (“sick quitters”).
They contend that when this bias is controlled for, the protective effect of drinking on all-cause mortality is almost completely eliminated and only applies to very low levels of drinking.
Yet, there are several observational studies that accounted for the “former drinker” problem and a protective effect of light to moderate drinking on cardiovascular and all-cause mortality was still observed. In addition, a great number of randomized controlled trials suggest that moderate drinking has significant benefits on various cardio-metabolic risk factors, thus, demonstrating a biologically plausible causal interaction.
Rarely discussed in this context, is the phenomenon of systematic underreporting of the consumed alcoholic beverages in these observational studies. It is known that self-reported intake of alcoholic beverages in population surveys only accounts for approximately 40–60% of alcoholic beverages sold. As a consequence, underreporting could affect the nadir (amount of alcohol where the health benefit is the greatest) of the J-shaped curve when relating the daily intake of alcoholic beverages and the risk of morbidity and mortality. A general trend towards underreporting in epidemiological studies could indicate that the reported risk reduction and risk increase actually occur at higher daily alcohol intakes than those reported in the scientific literature and public health reports.
To settle the discussions on the question about the optimal/beneficial versus detrimental amount of alcohol and to determine the association between alcohol intake and disease outcomes more reliably, realistic data on the intake of alcoholic beverages are required. Instead of interviews or questionnaires, measurements of specific biomarkers for alcohol, for example, the continuous transcutaneous ethanol measurements, might be a better option to provide better information on actual consumption levels in future observational studies and give public health advice. This could provide better information on consumption levels associated with beneficial as well as detrimental effects as the basis for public health advice.
Mørland J. Alcohol and the heart. J Intern Med. 2016 Apr;279(4):362-4.
For more information about this article, read the scientific abstract here