“Women who binge drink for the first three months of pregnancy have probably left it too late to benefit from giving up”, claims The Independent. The newspaper says a study shows that “the damaging effect of alcohol on pregnancy occurs during the first trimester, and cannot be reversed by stopping at that point”. The Daily Mail says that drinking in early pregnancy made an early birth 2.3 times more likely.
These stories are based on a study that questioned 4,700 new mothers on their drinking habits in pregnancy, and looked for links to prematurity and low birth weight. While the study did suggest drinking alcohol increased risk of prematurity, there are several limitations to the study. Almost all of the results were non-significant, including the 2.3-fold risk increase quoted above. The study also found that links were weakened after taking into account smoking and other known medical and pregnancy-related risk factors.
Expectant mothers should never consider it too late to reduce their alcohol, as suggested by media coverage. Instead, they should stick to NICE recommendations for drinking during pregnancy, regardless of prior drinking alcohol intake. NICE recommends that women avoid drinking during the first three months of pregnancy. Beyond that, they should drink no more than one or two units once or twice a week, and they should avoid binge drinking.
Where did the story come from?
This research was carried out by Colleen O’Leary and colleagues of the Centre for Child Health Research, the University of Western Australia, the National Perinatal Epidemiology Unit and the University of Oxford.
The study was funded by Healthway (a Western Australian Health Promotion Foundation) and the Australian National Health and Medical Research Council. It was published in the peer-reviewed medical journal the British Journal of Obstetrics and Gynaecology.
What kind of scientific study was this?
This was a retrospective cohort study investigating the relationship between alcohol consumption during pregnancy and foetal growth/premature birth. This study used data from a population-based cohort of women in Western Australia (WA), assessing the effects of quantity of alcohol, frequency of consumption and overall consumption during each trimester on foetal growth and preterm birth.
While there has been a lot of research on alcohol consumption and pregnancy, evidence on the effects of alcohol on growth and birth is considered by some to be inconclusive, particularly evidence on binge drinking.
Ten per cent of all women who gave birth in Western Australia between 1995 and 1997 were invited to complete a survey on their health habits during and after pregnancy. This was a postal survey, given when their baby was 12 weeks old. Mothers who had stillbirths or had adopted their baby were excluded.
A total of 4,861 women completed the questionnaires (81% response rate), and all but one were linked with their corresponding birth information on the WA Midwives’ Notification Scheme. After excluding multiple births and mothers from Australia’s indigenous people, 4,719 were available for analysis.
Women were asked about frequency of alcohol intake and quantity of various types of drink consumed during each three-month trimester of pregnancy. Frequency of alcohol intake was grouped as either five or more days per week; three to four days per week; one to two days per week; once or twice per month; less than once per month; or never).
Levels of alcohol consumption were grouped as:
- Low (<weekly to <daily, ≤20g per occasion, ≤60g per week).
- Moderate (<weekly to daily, 20-50g per occasion, ≤70g per week).
- Binge (classified as ‘less than weekly up to twice per week’ 50+g per occasion, ≥5g per week).
- Heavy (2+ days per week >10-50g per week, ≥68g per week, including binge drinkers).
The researchers defined one standard drink as containing the equivalent of 10g of pure alcohol. The ‘low’ category is said to be in line with the Australian National Health and Medical Research Council alcohol guidelines, which recommend no more than seven standard drinks in one week, and no more than two standard drinks in any one day.
The main outcome of this study was the effect of reported drinking on foetal growth and premature birth. The researchers calculated proportion of optimal birthweight (POBW), a measure of a baby’s actual weight compared to their ideal birthweight. In their calculations, the researchers took into account infant sex, duration of pregnancy, maternal height, and number of children mothers already had already. Premature birth was defined as infants born at less than 37 weeks of pregnancy.
Researchers assessed effects of drinking during each trimester on foetal growth and premature birth using the maximum alcohol consumed during that period. The analysis adjusted for some potential confounders, including: smoking, recreational drug use, maternal age, number of prior children, ethnicity, marital status, income, maternal medical complications, pregnancy complications, and procedures.
What were the results of the study?
Overall, about 57% of women in the sample abstained from alcohol in both the first and second trimesters, and 53% abstained in the third. Across trimesters, 28-35% of the sample drank low quantities of alcohol; 8-10% moderate quantities; 1-3% binged; and 1-2% were heavy drinkers.
The researchers found that higher levels of alcohol consumed during pregnancy were associated with both increased percentage of premature births and babies that were small for the length of time pregnancy lasted. However, adjustment for smoking weakened this association.
There were no significant associations between any level of drinking in any trimester of pregnancy and reduced birth weight. There was only one significant association between alcohol consumption and preterm birth: that was found when researchers combined the categories of moderate, binge and heavy drinking levels during the first trimester with later abstinence. This resulted in a borderline significant increased risk of premature birth compared with women who abstained throughout pregnancy (adjusted odds ratio 1.73; 95% confidence interval 1.01 to 3.14).
What interpretations did the researchers draw from these results?
The authors conclude that high levels of alcohol intake – particularly heavy and binge drinking – are associated with increased risk of premature birth even when confined to only the first trimester. However, they say that this finding requires further investigation into the quantity and timing of alcohol consumption during pregnancy, and any potential link to premature birth.
What does the NHS Knowledge Service make of this study?
The researchers’ conclusions may be overstated given that results were not statistically significant and that evidence of a trend towards increasing risk is questionable (given the small sample size used and the resultingly imprecise estimates). However, as the authors acknowledge, their study was small, and more research is needed.
The main limitation of this study was the reliance on the women’s self-reporting of alcohol consumption. There are a number of potential biases associated with this:
- Women having to recall alcohol consumption from up to a year before.
- Possible difficulty in estimating the size of a measure of alcohol.
- Possible intentional underestimation of alcohol intake.
- The possibility that women with a particular birth outcome may report their consumption differently from women with other birth outcomes.
- The researchers had to estimate grams of alcohol content in drinks and place them into a particular classification group.
There are other points that should be kept in mind when interpreting these results:
- The findings have also been over-interpreted by the news, as all but one of the adjusted results were not statistically significant. These results had been adjusted to account for the effects of smoking and other known medical and pregnancy-related issues that can affect prematurity.
- The authors say that although their sample was representative of mothers in Western Australia, there was under-representation of mothers with low birth weight infants (overall estimate 5.3% vs. 4.7% prevalence among study respondents), and mothers aged under 20 years (6.0% overall vs. 2.5% representation in this sample). This may have affected the results if women with lower-birth-weight infants and younger mothers had different drinking habits.
Messages to expectant mothers on how much alcohol they can consume may seem mixed and confusing. For now, current NICE recommendations should be followed:
- Pregnant women should avoid drinking alcohol in the first three months of pregnancy because it may be associated with an increased risk of miscarriage.
- Women who choose to drink alcohol during pregnancy should drink no more than one to two UK units once or twice a week.
- Although there is uncertainty regarding a safe level of alcohol consumption in pregnancy, at this low level there is no evidence of harm to the unborn baby.
- Getting drunk or binge drinking during pregnancy (defined as more than five standard drinks or 7.5 UK units on a single occasion) may be harmful to the unborn baby.
Analysis by Bazian
Edited by NHS Website
Links to the headlines
The Guardian, 21 January 2009
Daily Mail, 21 January 2009
BBC online, 21 January 2009
The Daily Telegraph, 21 January 2009
The Independent, 21 January 2009
Links to the science
BJOG 2009, January 21 [advanced online publication]