"Moderate tipplers have the best health and are less likely to miss work through illness," reports the Mail Online.
A study of 47,520 people from Britain, Finland and France found those who drank alcohol in moderation were less likely than teetotallers to take sickness absence for a range of illnesses.
But the results don't mean that drinking alcohol makes you healthier.
One obvious explanation may be that people with some health problems avoid alcohol because it makes their condition worse, or because they're on treatments that can't be taken with alcohol.
The study also found that people who didn't drink alcohol at all were more likely to be from poorer backgrounds, which can increase people's chances of ill health.
In addition, the study showed that people who drank above recommended limits were also more likely than moderate drinkers to need time off.
But in the case of heavy drinkers it was because of external causes, including injury or poisoning, rather than illness.
In the UK, men and women are advised to drink no more than 14 units a week.
Fourteen units is equivalent to 6 pints of average-strength beer or 10 small glasses of low-strength wine.
The researchers who carried out the study were from the Finnish Institute of Occupational Health, the University of Helsinki and the University of Turku in Finland, University College London in the UK, and the National Institute of Health and Medical Research (INSERM) and Université Paris Descartes in France.
It was funded by the Academy of Finland, the Nordic Programme on Health and Welfare, and the Economic and Social Research Council.
It was published in the peer-reviewed journal Addiction on an open access basis, so it's free to read online.
The study attracted widespread coverage in the UK media. This was reasonably accurate and balanced, with most coverage explaining that the results didn't show that drinking alcohol is healthier than not drinking.
This was a meta-analysis of data from 4 cohort studies in 3 countries.
Each of these provided information about people's alcohol consumption at 2 time points, and about their health-related absences from work in a follow-up period.
This type of study is useful for looking at links between factors – such as between alcohol consumption and sickness absence – but can't tell us whether one factor causes the other.
Researchers used information from 4 cohort studies: 2 from Finland (35,683 people in total), 1 from the UK (3,730 people) and 1 from France (8,107 people).
People were questioned about their alcohol use at time points between 2 and 6 years apart.
Researchers then looked at their employment sickness records for the following 4 to 7 years.
Researchers categorised people into:
They categorised illness absences according to the type of diagnosis:
Using low-risk people as the baseline, researchers looked at whether other patterns of alcohol use affected people's chances of taking time off for any of these 6 types of health problem.
They adjusted their figures to help take account of people's age, sex, socioeconomic status, smoking habits and body mass index, which can all affect a person's health.
Compared with low-risk drinkers:
Former at-risk and new at-risk drinkers didn't show any substantial differences in sickness leave from low-risk drinkers.
The researchers said their results showed "new evidence underlying the U-shaped association between alcohol use and sickness absence".
In particular, they said it demonstrated "different diagnostic patterns" for sickness absence, depending on whether people were non-drinkers or heavy drinkers.
They found more long-term mental and physical health conditions associated with abstinence, while heavy drinking was associated with injury or poisoning.
They added: "Our results can help occupational health care and facilitate early interventions/auditing for at risk alcohol use when accumulating sickness absence due to external causes [such as accidental injury] are observed."
The study supports what researchers have previously noted – a "U-shaped curve" whereby people who don't drink alcohol at all, or drink heavily, have poorer health than those who drink in moderation.
The study adds some information about the different reasons people who either don't drink or drink heavily take time off for sickness.
It's important to note that the study doesn't provide any evidence that drinking alcohol in moderation is therefore healthy, or that up to 17 units for women or 34 units for men would be considered "low risk". These cut-off points were based on Finnish guidelines, not UK ones.
It's unsurprising – and in line with previous research – that people who drink heavily are more at risk of needing time off for accidental injuries or even alcohol poisoning.
The explanation for non-drinkers taking time off work is less clear, but it's likely that some people who abstain from alcohol altogether do so because they have a long-term health condition, and either their condition or their medication means they can't drink alcohol.
The study has some other limitations:
The study supports sticking to recommended alcohol limits, which are no more than 14 units a week for men or women in the UK.