"A daily pint or glass of wine can slash the chances of a suffering heart attack by a third," reports The Sun.
Researchers found that people who drank alcohol within moderate drinking guidelines were less likely to have a first episode of a range of heart and vascular diseases than those who never drank alcohol.
This four-year study looked at health records of almost 2 million adults without any cardiovascular disease at the start of the study.
It found that non-drinkers were more likely to need treatment for many diseases such as heart attack, heart failure and angina, compared to people who drank alcohol within the previous recommended guidelines, which were 21 units per week for men and 14 units for women.
There was less difference between the groups for circulation diseases such as stroke and bleeding in the brain.
However, heavy drinkers, consuming above guideline limits, were also at higher risk compared with moderate drinkers. Former and occasional drinkers also had increased risk of several outcomes.
Along with other study limitations, like the possible influence of other health and lifestyle factors, we can't be certain that moderate drinking directly decreases risk.
And at the risk of sounding like killjoys, there are far healthier and more effective methods of reducing cardiovascular disease, such as regular exercise. Regular drinking, can increase your risks of a number of cancers.
Alcohol guidelines changed at the start of 2016 to recommend that both men and women should drink no more than 14 units per week. This was to reflect the point that there is no such thing as a "safe amount" of alcohol.
The study was carried out by researchers from Cambridge University and University College London and was funded by organisations including the National Institute for Health Research, Wellcome Trust and Medical Research Council.
The study was received with enthusiasm by the UK media. The Sun's exhortation to readers to drink "a pint a day," accompanied by a photograph of a man sinking a beer, was typical of the tone of much of the coverage. However, the headline oversimplifies the study.
The Daily Mirror does a more balanced job, warning readers that "There's a catch" and quoting experts warning of the link between alcohol and cancer.
The Mirror also carries a statement from Dave Roberts, director general of the Alcohol Information Partnership, who claims that "the anti-alcohol campaigners' mantra that there is no safe limit just doesn't stack up".
But as the Alcohol Information Partnership is funded by drinks firms including Diageo, Pernod Ricard, Campari and Bacardi (as the Mirror helpfully points out) there may be a potential conflict of interest.
The media reporting also fails to point out that this study was basing the definition of moderate drinking on the old, pre-2016, recommendations (21 units per week for a man, 14 per week for a woman).
This was a cohort study using population-based records. Researchers wanted to see how alcohol consumption at different levels was linked to a wide range of cardiovascular conditions.
Cohort studies can show links between factors, such as alcohol consumption and cardiovascular disease risk. But they can't show that one factor causes another. Confounding factors (such as diet and physical activity) might distort the results.
Researchers used anonymised electronic patient records from a GP database, which included people's reported alcohol consumption. They included 1,937,360 patients aged 30 or over, and tracked their illnesses, hospital admissions and deaths over an average six years.
They divided people into groups based on their drinking, then (after adjusting for confounding factors) looked to see what their chances were of having had one of 12 cardiovascular conditions, or having died from any cause.
The researchers only looked at people's first record of a cardiovascular disease. So, for example, someone might have had treatment for unstable angina, then later go on to have a heart attack, but only the unstable angina would be recorded.
The researchers used three linked databases, to give them a better chance of including all necessary detail. As well as the GP database they used the Myocardial Ischaemia National Audit Registry Project, hospital episode statistics and the Office of National Statistics.
The researchers divided people into five groups: non-drinkers (who never drank alcohol), former drinkers, occasional drinkers, moderate drinkers (who drank within the then-current guidelines of 21 units per week for men and 14 units for women) and heavy drinkers (who exceeded this).
Potential confounding factors included in the analysis were:
About 5% of people in the study had a first diagnosis of a cardiovascular disease during the study. As in previous studies, this was more common among non-drinkers, former drinkers, occasional drinkers and heavy drinkers, compared to moderate drinkers.
Compared to moderate drinkers, non-drinkers had a higher risk of a first report of:
There was no significantly increased risk of bleeding in the brain, transient ischaemic attack ("mini-stroke"), or sudden cardiac death.
Heavy drinkers also had an increased risk of death from any cause or from heart disease, of cardiac arrest, heart failure, stroke from blood clot or bleed and peripheral artery disease, with risk increases ranging between 11% and 50%.
Former drinkers and occasional drinks also had increased risk of most outcomes compared with moderate drinkers.
The researchers also noted that all non-drinkers were more likely to belong to the most deprived socioeconomic group, to have diabetes, and to be obese.
The results were similar for women, although there was less difference in risk levels between non-drinkers and moderate drinkers.
The researchers say their study showed that "moderate alcohol consumption is associated with a lower risk of initially presenting with several, but not all, cardiovascular diseases." They go on to say that "heavy drinking is differentially associated with a range of diseases."
While the research found that heavy drinkers were less likely to have a heart attack as a first presentation, the researchers warn that could be because "they die from other causes before they are able to develop a cardiovascular disease."
This study paints a more complicated picture than the "Pint a day keeps the doctor away" story proffered by The Sun.
It seems to confirm the findings of other studies, which have shown that non-drinkers tend to have a higher risk of cardiovascular diseases than people who drink moderately.
It suggests that some cardiovascular diseases (mainly those directly affecting the heart) seem to have a stronger link to a possible protective effect from alcohol than other vascular diseases, such as mini-strokes and bleeding in the brain. However, this can't be concluded with certainty due to the study design.
We need to remember that cohort studies like this cannot prove that alcohol consumption or lack of it is a direct cause of cardiovascular disease. Many health and lifestyle factors may be having an influence. For example, non-drinkers were more likely to be from deprived areas, to have diabetes or be obese, factors which the analysis didn't adjust for.
We also have no information about other factors such as diet or exercise, which could also affect the results.
Also, the researchers' decision only to include people's first diagnosis of a cardiovascular disease complicates matters. For example, if a person had a Transient ischaemic attack (TIA) (also known as a "mini-stroke") and then went onto have a full stroke, only the TIA would be recorded. Therefore it is difficult to be certain of a person's overall cardiovascular disease status. We can't be sure that the figures around how much a person's risk of a particular disease outcome is increased by a particular consumption level are accurate.
As a result, we really shouldn't conclude, for example, that people who drink heavily are less likely to have a heart attack than those who don't drink. They may have a stroke first, and then a heart attack, or die of another cause.
The study isn't a green light for people to drink more alcohol, without worrying about it. However, it does suggest that drinking alcohol within the lower-risk drinking guidelines may not raise the risk of cardiovascular disease, and may lower it. Remember that alcohol does contribute to other diseases.
Check whether you're drinking within low-risk levels with our introduction to alcohol units.