“A glass of wine every day in pregnancy could be good for your baby,” is the entirely incorrect headline in The Daily Telegraph today. Other newspapers reported that drinking while pregnant does ‘no harm’, these claims are also misleading.
The news is based on a new study investigating the link between alcohol exposure before birth and childhood balance, which is considered an important sign of children’s development. Previous research has found that alcohol consumption during pregnancy is linked to poorer outcomes on several markers of neurodevelopment, but the effect on balance is uncertain.
Researchers found no evidence of an adverse effect of low-to-moderate maternal alcohol consumption on childhood balance. They also found moderate alcohol exposure seemed to have a beneficial effect compared to no alcohol. However, the researchers say that this positive effect is possibly due to them not being able to fully adjust for the fact that higher alcohol use was linked to social advantage.
This study adds to knowledge about the effects of alcohol in pregnancy on children’s balance. However, uncertainty remains over whether there is a ‘safe’ level of alcohol consumption during pregnancy. For this reason, current guidance suggests that women avoid alcohol completely during the first three months of pregnancy. If women choose to drink after this time, they should not drink more than one to two units of alcohol once or twice per week, and avoid binge drinking altogether. Contrary to media suggestions, this study does not change this advice.
The study was carried out by researchers from the University of Bristol and University Hospital Bristol NHS Foundation Trust and was funded by the UK Medical Research Council, the Wellcome Trust, the University of Bristol and the Alcohol Education and Research Council (AERC). The study was published in the peer-reviewed, open access medical journal, BMJ Open.
The Telegraph’s headline was incorrect and potentially dangerous. While the researchers found a positive effect of alcohol in one measure, they clearly and categorically said that this likely to be a statistical blip. This headline also ignores the fact that the research was into just one aspect of children’s development and that alcohol consumption was measured at only one point in time.
Because of these limitations, women should stick to the existing advice on alcohol in pregnancy. Fortunately, in its online edition, the Telegraph used a more accurate headline. All other newspapers’ headlines suggested that drinking in pregnancy was “OK”, or does “no harm” – and are also misleading.
This was a prospective cohort study that aimed to determine whether there was a link between alcohol exposure during pregnancy and balance in 10-year old children. Although this is the ideal study design to address this question, cohort studies can only show association, and cannot prove a cause-and-effect relationship. This is because other factors (confounders) may be responsible for any association seen.
This problem is demonstrated by this study. Despite the fact that the researchers collected information on a number of socioeconomic factors, and adjusted for them in their analyses, they conclude that the associations seen are probably due to not fully being able to adjust for social advantage.
The researchers used information on 6,915 children and their parents, who were participating in the Avon Longitudinal Study of Parents and Children.
This study used children who were born singly (who were not twins or another multiple birth) between April 1991 and December 1992, who had undergone balance assessment at 10 years of age and for whom they had information on the mother’s alcohol intake.
The balance assessment evaluated three types of balance:
Children were said to have ‘good balance’ if they were in the top 25% fastest times for crossing the balance beam (good dynamic balance), if they maintained the static balances with their eyes open for 20 seconds (good static balance with eyes open), and if they were in the top 25% longest times for holding the static balances with eyes closed (good static balance with eyes closed).
Alcohol exposure was measured by asking mothers and fathers to self-report their alcohol intake at 18 weeks of pregnancy.
At 18 weeks of pregnancy, mothers reported both their current consumption and their consumption prior to pregnancy. For each time point, the mothers reported the total number of glasses (defined as a pub measure of spirits, half a pint of larger or cider, a small glass of wine) consumed per week, categorised into none (0 glasses), low (1-2 glasses), moderate (3-7 glasses), and high (more than 7 glasses) consumption. Mothers were also asked how many days in the previous month they had drunk the equivalent of at least four units of alcohol (binge drinking). Similarly, fathers reported alcohol consumption and binge drinking.
Information on other factors that could explain any association seen (confounders) was also collected. These included marital status, crowding index (number of people in the household and number of rooms), home ownership, parity (the number of previous children the mother has), maternal education, ethnicity, maternal age, maternal social class, smoking, cannabis use, caffeine consumption, number of stressful maternal life events during pregnancy, and maternal depression.
The researchers looked to see whether there was a link between alcohol exposure during pregnancy and the balance ability of 10-year old children after adjusting for these potential confounders.
Few mothers reported drinking heavily during pregnancy, with 95.5% of mothers reporting no alcohol consumption to moderate alcohol consumption.
In general, higher total levels of maternal alcohol consumption were associated with higher socioeconomic status and higher maternal age, whereas higher levels of binge drinking were associated with lower socioeconomic status and lower maternal age.
No evidence was found of an adverse effect of maternal alcohol consumption on childhood balance.
Regular heavy maternal binge drinking (more than 10 times per month) was also associated with good static balance (eyes shut) in children. There was no significant association between any other level of binge drinking, or any other measure of balance.
Paternal drinking during the first three months was associated with good static balance (eyes open) in children, with fathers that reported drinking less than one glass per week, at least one glass per week and at least one glass per day having children with better static balance than those that reported never drinking.
The researchers then analysed the data differently, using a technique called “Mendelian randomisation”. This approach is based on the assumption that a person’s DNA is not linked to socioeconomic status.
It is known from previous research that particular variations in a gene that codes for alcohol dehydrogenase (an enzyme that breaks down alcohol) predisposes people to lower alcohol consumption. The researchers looked at this variant. Mothers carrying this variant consumed less alcohol before, during and after pregnancy.
There was no evidence that mothers carrying this variant had children with poorer balance, which is not what would be expected if alcohol exposure improves balance. The researchers use this result to suggest that the previous association between maternal alcohol consumption and balance outcomes may have been due to the fact that the current analysis could not completely adjust for socioeconomic status.
The researchers suggest that the most correct interpretation of their results is that they, “provide no strong evidence of an effect, either beneficial or detrimental, of moderate maternal alcohol use during pregnancy on offspring balance.”
This large, well-designed prospective study has found no evidence that moderate maternal alcohol consumption at 18 weeks of pregnancy has an adverse effect on offspring balance at age 10.
The study is limited by the fact that, as a cohort study, it cannot show a cause-and-effect relationship. This is because other confounding factors may be responsible for any association seen.
Despite the fact that the researchers collected information on a number of socioeconomic factors, and adjusted for them in their analyses, they conclude that the small benefits seen for some outcomes with some drinking patterns are probably due to not fully being able to adjust for social advantage.
Also, maternal and paternal alcohol use was self-reported and alcohol use during pregnancy was assessed at only one point in time, which could be subject to bias. The researchers also reported that the balance measures used had low test-retest reliability.
Although the results of this study will add to knowledge about the effects of alcohol in pregnancy on one particular developmental outcome, uncertainty remains over what is a ‘safe’ level of alcohol consumption during pregnancy.
Current guidance suggests that women avoid alcohol completely during the first three months of pregnancy due to increased risk of miscarriage, and if they choose to drink after this time, should not drink more than one to two units of alcohol once or twice per week, and avoid binge drinking altogether. This study does not change this advice.