This study looks at the impacts of initiating and ceasing alcohol consumption on serum cholesterol levels (in particular, the relative levels of low- to high-density lipoprotein cholesterol) in two categories of patients attending ‘annual health checks’ in Japan over a 10 year period. The reason for the study interest was that lipid profiles have an important role in cardiovascular health, with a greater proportion of cardiovascular events associated with higher proportions of low density low (relative to high) density lipoprotein cholesterol, and that alcohol has long been associated with the poor cardiovascular outcomes. After statistically controlling for a range of lifestyle factors, the authors report that low density lipoprotein cholesterol increased (and high density decreased) at greater levels among those who did not commence drinking alcohol than among those who did start drinking over the time period. At the same time, low density lipoprotein cholesterol increased over the 10 years in those who gave up alcohol consumption (with lower high density lipid proportions) relative to those who continued drinking. Presumably, this lipid profile suggests that alcohol consumption may decrease risk of cardiovascular disease, at least at low levels of consumption, which is something that alcohol promoters have been very keen to suggest for many years and many in the population would be also be very relieved to hear!

There are a few things to think about here, however, before celebrating. This study included a large number of participants but excluded a relatively large number as well. Those who already had diagnosed high cholesterol were among the large number excluded, as were those who were regular drinkers - including those who had increased or decreased their consumption in the 10 year study period. The coverage of the population attending the annual checks was also not discussed, so we are left with some uncertainty about the representativeness of the final cohorts included in the study. Further, as the authors point out, the alcohol consumption data were self-reported (always a bit of an issue in observational studies) and there were (necessarily) limitations in the range of lifestyle and cardiovascular health outcome information collected from participants. I would also add that we have no idea about the full range of health or social conditions of the participants that may or may not have contributed to alcohol consumption patterns but which may also be associated with lipid profile.

All in all, the study provides interesting information about the potential influence of alcohol consumption and certainly underscores the need for further study on this very important issue. I just wouldn’t be popping the champagne corks just yet!!

Alcohol is a colourful thread that is woven into the fabric of Australian society. A range of health claims have been made about alcohol, often stemming from research that has been sponsored by the alcohol industry. Many of these claims, such as the idea that wine could have cardiovascular benefits, have now been disproven, with subsequent research demonstrating that such findings are often quirks of the methodology employed in the research.
 
The findings of the most recent study appear to share some of these methodological issues. For example, some people who ceased drinking might have been heavy drinkers compared to the moderate drinkers who seem to have benefited from alcohol consumption. Like the research into the apparent cardiovascular benefits of alcohol, if people who have been heavy drinkers but ceased consuming alcohol are excluded from the analysis, then the data might show there is actually no benefit from consuming alcohol on cholesterol levels. Indeed, the researchers note that their finding that people who drink moderate levels of alcohol have better cholesterol might not translate to any reduction in cardiovascular disease. Further, genetic differences in people from Southeast Asia can affect the way that alcohol is metabolised by the body such that these findings might not been seen in a Caucasian population. Finally, the differences in cholesterol between the different drinking groups are quite small and the finding might reflect liberal statistical analysis.
 
People should not start drinking or increase their alcohol consumption as a result of these findings.

This new research should be interpreted with some caution – particularly as it looked at people who chose themselves to start or stop drinking. These decisions often accompany other major lifestyle or health changes which could be responsible for the observed cholesterol changes. The phenomenon of ‘sick quitting’ is particularly well known, in which people stop drinking after their health has already started to decline.

That said, the findings in this new paper are consistent with past studies, which do generally find a link between moderate alcohol consumption and increased LDL-C (considered harmful for heart disease) as well as reduced HDL-C (associated with good heart health).

However, even if low-level drinking does have modest benefits for cholesterol levels, recent research has shown this doesn’t necessarily translate into protection against death by cardiovascular disease. Moreover, for other serious conditions like cancer, we know that every drink increases risk. For these reasons, it’s really important that people don’t interpret this new research as a reason to start drinking or to maintain unhealthy drinking patterns.

The relationships between alcohol intake, blood lipids (fats) and heart disease are complex.  While some older studies suggested potential cardiovascular benefits from light-to-moderate drinking, more recent research indicates that alcohol consumption at all levels may increase cardiovascular risk. The relationship between alcohol and cardiovascular mortality is complicated, with the risk increasing substantially at higher levels of consumption. Current evidence does not support recommending alcohol consumption for cardiovascular health benefits.

Mild-moderate alcohol intake is associated with a small increase in HDL-cholesterol that may be considered beneficial, consistent with this study. Effects on LDL-cholesterol found here have also been reported, but less consistently. Triglycerides increase in association with alcohol intake.

However, these changes depend not only on the amount of alcohol consumed, but also on the pattern of consumption.  Binge drinking has adverse effects whereas small regular amounts seem less harmful.  There are also genetic factors known to influence the relationship between alcohol intake and lipid metabolism. 

This observational study in a large Japanese population should not encourage anyone to increase their alcohol intake. Alcohol intake was self-reported. This is known to be fraught. People may have changed their diet once they stopped drinking alcohol which may account for some of the changes in blood lipids along with many other confounding factors that could contribute to the findings. There are also differences in patterns of drinking between Japanese and some other populations as well as the pattern of genes associated with lipid changes and alcohol.

Finding a potential beneficial change in biochemistry does not change the weight of present evidence that alcohol intake beyond the national guidelines is associated with adverse health outcomes, not only heart disease, but cancer and many other serious health problems.