Monday September 2 2013
The researchers advised that if you don't drink, you shouldn't start
Drinking a glass of wine every day may be good for mental health, report The Daily Telegraph and The Guardian.
The story is based on a Spanish study which followed 5,505 people aged 55 to 80 for over seven years. It found that those who drank two to seven glasses of wine a week were about a third less likely to develop depression than non-drinkers.
However, the study found that those who drank heavily (over five units of alcohol a day) tended to be at greater risk of developing depression, but it could not rule out the possibility that this finding occurred by chance.
The authors acknowledge that their findings contrast with a number of other studies which have found alcohol consumption to be associated with an increased risk of depression. They say that this may be because the people in these other studies drank different types of alcohol or had different patterns of consumption (for instance, binge drinking versus regular consumption).
The study has a number of limitations, including that factors other than alcohol may be having an effect, such as significant life events. Also, the researchers relied on people telling them that they had been diagnosed with depression or that they were taking antidepressants, which may not be a completely accurate way of identifying those with the condition.
Overall, it would not be wise to take up drinking just to reduce depression risk based on this study, as it’s not possible to say that this will have the desired effect. However, it is in line with current advice that if you do drink alcohol, you should do so in moderation.
Where did the story come from?
The study was carried out by researchers from the University of Navarra and other research centres in Spain. Sources of funding for the study were not reported, but the first author on the paper was supported by the Spanish Government.
The researchers declared various potential competing interests, for example, one reported serving on the board of the Research Foundation on Wine and Nutrition, the Beer and Health Foundation, and the European Foundation for Alcohol Research.
The study was published in the peer-reviewed open access journal BMC Medicine.
The media has covered this story relatively uncritically, but the Guardian does include the important note from one of the study authors that "If you are not a drinker, please don't start drinking".
What kind of research was this?
This was a cohort analysis assessing the relationship between alcohol intake and risk of developing depression. The individuals being assessed were taking part in a randomised controlled trial called the PREDIMED study.
This study assessed the cardiovascular effects of the Mediterranean diet supplemented with either extra virgin olive oil or mixed nuts, or a control diet. However, alcohol consumption was not allocated randomly, instead people decided on their own alcohol intake. As this was the case, the main inherent limitation to the study design is that people who chose to drink either more or less may differ in other characteristics from those who make different choices.
These other differences (called confounders) may be affecting depression risk rather than alcohol intake. Researchers can try to take any known differences between the groups into account, but there may be unknown differences having an effect.
What did the research involve?
Men and women aged from 55 to 80 were enrolled, and their alcohol intake was assessed at the start of the study and every year subsequently. The researchers assessed which individuals developed depression during up to seven years’ follow-up, and analysed whether a person’s alcohol intake was related to their risk of developing the condition.
To be eligible for the study, individuals had to be free from cardiovascular disease at enrolment, but had to have either type 2 diabetes or three or more coronary heart disease risk factors. People with problematic alcohol use were excluded from the study.
For the current analysis, individuals reporting depression currently or in the past, or use of antidepressants were excluded. People with missing alcohol data, lost to follow-up, or with unfeasibly high or low reported calorie intakes were also excluded. This left 5,505 people for analysis.
Alcohol consumption and intake of other drinks and food were assessed using a food frequency questionnaire. Nine questions on alcoholic beverages were included, addressing different types of wine, beer and spirits. Participants were divided into four groups according to their alcohol intake:
- no alcohol intake (abstainers)
- less than 5g of alcohol per day (for reference, a UK unit contains just under 8g of alcohol, so this would be less than one UK unit a day)
- between 5g and 15g of alcohol per day (about one to two UK units a day)
- more than 15g of alcohol per day (more than two UK units a day).
When looking at the effect of wine, participants were divided into five groups based on their intake of wine:
- abstainers (those who did not drink wine but did drink other alcoholic drinks were excluded)
- less than one drink a week
- one to less than two drinks a week
- two to seven drinks a week
- more than seven drinks a week.
Individuals who reported being diagnosed with depression by a physician in their annual assessment interview were considered to have the condition, as were individuals who reported habitually taking antidepressant drugs. Only a person’s first episode of depression was considered in the analyses.
Researchers looked at whether risk of developing depression differed in people with different alcohol intakes. They took into account potential confounders including age, gender, smoking, physical activity, total energy intake, body mass index, marital status, which group they were in in the randomised controlled trials, education, living alone, and where the person was recruited.
What were the basic results?
At the start of the study:
- 33% did not drink alcohol
- 25% drank less than 5g of alcohol a day (for reference, a UK unit contains just under 8g of alcohol)
- 23% drank between 5g and 15g of alcohol per day
- 19% drank over 15g of alcohol per day.
During follow-up, 443 people (8%) had an episode of depression.
After taking into account potential confounders, those with low to moderate alcohol intake (between 5g and 15g of alcohol per day) at the start of the study, were about 28% less likely to develop depression during follow-up than those who drank no alcohol (hazard ratio (HR) 0.72, 95% confidence interval (CI) 0.53 to 0.98).
Those who drank less than this (up to 5g alcohol per day) or more than this (more than 15g per day) at the start of the study did not differ significantly in their risk of developing depression from abstainers. There was a trend for heavy drinkers (more than 40g alcohol per day, about five UK units) to be at increased risk of depression, but this did not reach statistical significance, possibly because there were only a small number of heavy drinkers in the study (HR 1.34, 95% CI 0.69 to 2.59).
If the analyses took into account changes in alcohol consumption during the study, there were similar results, although in these analyses light drinkers (up to 5g alcohol per day) were also less likely to develop depression in these analyses.
When looking specifically at wine consumption, those who drank two to seven drinks of wine a week at the start of the study were at about 32% less likely to develop depression than those who drank no alcohol (HR 0.68, 95% CI 0.47 to 0.98).
If the researchers carried out analyses excluding people who developed depression shortly after their alcohol intake was assessed (for example those who may have already had depression but had not been diagnosed), this did not differ largely from their main analyses. This was also true if they excluded former drinkers from the “abstainers” group.
How did the researchers interpret the results?
The researchers concluded that low to moderate alcohol consumption (about one to two UK units a day on average), and moderate wine consumption specifically, may reduce the risk of depression. However, heavy drinking (more than five UK units a day) may increase the risk. They say that further cohort studies are needed to confirm these results.
This study has found an association between low to moderate wine consumption in Spain and a reduced risk of developing depression. The authors note that this contrasts with other studies that have found alcohol consumption to be associated with an increased risk of depression. They say that this may be because the people in these other studies drank different types of alcohol (for example less wine and more of other drinks) or had different patterns of consumption (such as binge drinking as opposed to regular low to moderate consumption).
The researchers did attempt to reduce other potential problems with alcohol studies, by collecting data prospectively and at multiple time points, and by testing the effect of removing those who were most likely to have had undiagnosed depression at the time their alcohol intake was assessed, and those who could have given up alcohol intake for health reasons.
As with all studies of this type, the main limitation is that people who chose to drink low to moderate amounts of alcohol may differ in other characteristics from those who make different choices. These other differences (called confounders) may be affecting depression risk rather than alcohol intake. Researchers tried to take some differences between the groups into account, but it is difficult to completely remove their influence. There may be other unmeasured differences having an effect – for example, significant life events.
Other limitations are that people may not report their intake very accurately. Also, the analyses of the effect of wine were based on number of ‘drinks’, which does not tell us how large these drinks were. Therefore it is difficult to interpret the results on wine intake. The study was in older individuals (average age 67 years), and results may not apply to younger people.
Also, the researchers did not assess people for depression themselves; they relied on people telling them they had been diagnosed or that they were taking antidepressants. This way of measuring diagnosis may not be very accurate, for example, if people feel that there is stigma attached to their diagnosis and do not report it to researchers. Also, antidepressants are used to treat some conditions other than depression, so this may also cause some misclassification.
Overall, it would not be wise to take up drinking just to reduce your depression risk based on this study, as it is not possible to say that this will have the desired effect. However, the study is in line with current alcohol guidelines, which is that if you do drink, you should do so in moderation.
It’s also important to note that the study also doesn’t apply to those who already have depression, who are generally advised not to drink alcohol.
Analysis by NHS Choices. Follow Behind the Headlines on Twitter.