People who drink alcohol three to four times a week have a lower risk of diabetes than those who don't drink at all, while a having a heavier consumption is associated with a risk greater than or equal to that of alcohol abstainers.
Many epidemiological studies have suggested that light to moderate alcohol consumption is associated with a lower risk of diabetes compared with abstention, in both men and women.
However, there has been little work done to investigate the impact of drinking patterns, such as drinking frequency or binge drinking, on the risk of diabetes. Some research had implied that drinking frequency could be more important than the amount of alcohol that people drink when it comes to assessing the risk of diabetes.
In a large study now published in the journal Diabetologia, scientists have investigated this issue further, examining the effects of drinking frequency on type 2 diabetes risk for 28,704 men and 41,847 women in Denmark.
"What we knew was that there was a U-shaped relationship between alcohol consumption and risk of diabetes. People who drink a moderate amount of alcohol appear to have the lowest risk of diabetes," lead author Janne Tolstrup from the National Institute of Public Health of the University of Southern Denmark, told IBTimes UK.
"Here, rather than studying the link between number of drinks and diabetes, we wanted to study the influence of drinking patterns – in other words the number of times a week they were drinking."
Drinking patterns and alcohol types
The researchers analysed data from the Danish Health Examination Survey from 2007-2008, in which Danish citizens aged 18 and over filled in a self-reporting questionnaire which included questions about their lifestyle and health status. They also gathered follow up information until 2012 from Danish nationwide registries.
To conduct their analyses, the researchers had to classify the participants into different groups - those who abstained from alcohol, those who were drinking alcohol less than 1 day a week, 1 or 2 days a week, 3 to 4 days a week and 5 to 7 days a week.
They also looked at how much people were binge drinking and documented the types of beverage people drank, as well as their average weekly alcohol amounts. Participants were also asked whether their alcohol consumption had increased, decreased or remained stable over the previous 5 years.
Finally, the scientists controlled for factors such as age, sex, level of education, body mass index, smoking status, diet, leisure time activity, hypertension and family history of diabetes.
They found that during the follow up, 859 men and 887 women were diagnosed with type 2 diabetes. When they looked at the amount that people drank, their results were similar to those of previous studies - individuals with the lowest risk of developing diabetes were those who were consuming moderate amounts of alcohol.
But the new findings were those regarding drinking frequency. The scientists showed that consuming alcohol 3 to 4 days a week was associated with the lowest risk of diabetes - a 27% lower risk in men and a 32% lower risk in women - when compared to individuals who were drinking less than one day per week or 5 to 7 days a week. The idea is that spreading out drinking on three to four days during the week could be beneficial.
"For a given amount, taking it in on three or four separate occasion reduces the risk more than taking the same amount in one day," Tolstrup explained.
It is not clear how a moderate and frequent alcohol consumption reduces the risk of diabetes. Previous studies have suggested that alcohol increases insulin sensitivity. However, more experimental studies now need to be done in the lab to find out more.
That being said, the message of this study is not that people should rush to drink a lot of alcohol at different points during the week to reduce their risk. It merely confirms a correlation between moderate alcohol consumption spread out over the week and beneficial health impact.
"Type 2 diabetes risk is complex. Several factors contribute to it, including family history, ethnic background, age and being overweight. While these findings are interesting, we wouldn't recommend people see them as a green light to drink in excess of the existing NHS guidelines. Especially as the impact of regular alcohol consumption on the risk of Type 2 will be different from one person to the next," said Emily Burns, Head of Research Communications at Diabetes UK.