HONcode Certified

This website is certified by Health On the Net Foundation. Click to verify.

This site complies with the HONcode standard for trustworthy health information: verify here.

Light drinking in pregnancy tells us more about the socioeconomic status of the mother than the development of the child

Ami Banerjee
Last edited 7th October 2010

We have known since case reports from the early 1970s that heavy drinking in pregnant mothers leads to problems with growth and development in the foetus. The “foetal alcohol syndrome” has been well-characterised since then. What is not so clear is whether “light drinking” is as harmful during foetal development and later childhood. Print and web media have been obsessing all day about whether pregnant mothers are alright to have the odd gin-and-tonic or glass of wine. It is a comment on the prominent place of alcohol in UK culture that all the commentaries are calling for relaxation of current recommendations which advise abstinence during pregnancy, rather than calling for a cautious re-examination of data.

The furore is due to findings based on the UK Millenium Cohort Study, which looked at over 11 000 infants
born across the UK, selected “such that disadvantaged residential areas are over-represented”. These infants were seen at 9 months, 3 years and 5 years and detailed data about the health and behaviours of mother and child were recorded. So immediately, we see that the authors can only comment about childhood upto 5 years of age and not beyond. This study has previously claimed that at 3 years of age, children of mothers who drink 1-2 drinks of alcohol per week are not at increased risk of behavioural or learning difficulties.

In the current study, the authors compared behavioural and learning difficulties in children of mothers in light, moderate and high drinking categories with children of mothers who were always tee-total. They adjusted their findings for possible confounding factors:e.g. child's age, birth weight, mother's age at time of birth, number of children in the household, mother smoked during pregnancy and other socioeconomic factors. The p-values for odds ratios are not reported in the paper, but the implication is that children of mothers who stopped drinking during pregnancy were worse off than children of mothers who were light drinkers. They did find a much higher prevalence of learning and behavioural difficulties in children of mothers who drank heavily though, and so they refer to a “U-shaped relationship” between alcohol consumption and risk of disease, which has previously been documented in adults. That is to say that a bit of alcohol is good for you, in fact it is better than none at all, but a lot of it is bad for you.

Importantly, the authors found that “light drinkers were more socioeconomically advantaged compared with mothers in all other categories. The socioeconomic profile of mothers in the ‘not-in-pregnancy’ group was more advantaged than the ‘never-drinker’ group but less advantaged than the ‘light’ drinking group”. The authors themselves admit that the effects may be explainable by socioeconomic status, i.e. mothers who are light drinkers have other beneficial environmental factors which may explain any differences in development of the children more plausibly than alcohol. Unsurprisingly, this was not reported by any of the blogs or newspapers I read today. Thankfully the Department of Health is being a bit slower to open the floodgates to heavy boozing during pregnancy. Surely, longer follow-up and more studies are required before we recommend all mothers to drink during pregnancy.

Need for clear, consistent public health messages

The evidence in your quoted document states:
"In summary, for most outcomes, there was no consistent evidence of adverse effects from low-to-moderate prenatal alcohol consumption. Nevertheless, the evidence is probably not strong enough to rule out any risk. There was some evidence of adverse effects on neurodevelopment of binge drinking during pregnancy."

So we cannot at the moment rule out risk. Evidence-based policy faces a dilemma in translation because until the evidence is unequivocal that low/moderate alcohol intake does not affect the foetus, relaxation of drinking guidance may send the public health message that heavy drinking is OK in pregnancy.

Without a doubt, alcohol has health risks and costs which tend to be underestimated compared to other factors. An analysis by Allender et al showed that the cost per year due to alcohol-related illness to the NHS is £3 billion compared with £5 billion due to smoking. I meant to say that in the UK, we have for too long ignored the consequences of high alcohol intake, which may be an opinion but I can back it up with plenty of evidence.

Clear messages

Thanks for the reply.

The evidence in your quoted document states:

Well, no. "The summary states..." would be accurate. The evidence is, well, the evidence. The summary is an interpretation of that evidence. This is not just playing with semantics, as surely the point of public health medicine is correct interpretation of evidence and then conveying an appropriate message to the public.

So we cannot at the moment rule out risk

Of course, that is true. Isn't it also true to say that at the moment we cannot demonstrate any harm from mild/mod intake? Isn't that important?

Without a doubt, alcohol has health risks and costs

Absolutely, but there is a dose/response relationship - as the evidence here appears to demonstrate. Health risks occur with excess drinking, and these effects are serious. I think we agree on this point.

So, what message do you then send out? You can recommend complete abstinence, but I do not see how this is supported by the evidence, and I think it's counter-productive. I believe people are clever enough to realise that low quantities of alcohol are unlikely to do harm, and they will characterise advise that all intake is to be avoided as erroneous. It is then much more difficult to reach these people with the message that binge or excessive drinking is definitely associated with harm, as they are no longer listening.

I think this is a situation where we would be better off letting people decide for themselves after a realistic representation of the risks. It is my opinion that our audience is sophisticated enough to understand the points we have discussed if they are presented properly.

Finally, if alcohol is perceived as something that authority wishes to keep you away from, and there is poor engagement/lack of trust between the public and those in authority, it's attractiveness to those we find difficult to reach may continue to rise...


It is a comment on the prominent place of alcohol in UK culture that all the commentaries are calling for relaxation of current recommendations which advise abstinence during pregnancy

Is it? Really? Or perhaps it is a comment on the puritanical nature of this public health advice which is completely unsupported by any evidence? Yes, alcohol in excess causes problems, but the evidence suggests mild/mod intake does not cause any problems. At all.


There are obvious drawbacks with the some of the studies reviewed in this document, but the message remains: mild/mod intake does not appear to cause any problems to the foetus.

Advice to the contrary is not based in EBM, so therefore is personal prejudice, isn't it?

Twitter TrustTheEvidence.net


Search the TRIP Database

TRIP Database


Recent Comments