Article Text

Download PDFPDF

“You can’t handle the truth”; medical paternalism and prenatal alcohol use
  1. C Gavaghan
  1. Colin Gavaghan, School of Law, University of Glasgow, Glasgow G12 8QQ, UK; C.Gavaghan{at}


The publication of the latest contribution to the alcohol-in-pregnancy debate, and the now customary flurry of media attention it generated, have precipitated the renewal of a series of ongoing debates about safe levels of consumption and responsible prenatal conduct. The University College London (UCL) study’s finding that low levels of alcohol did not contribute to adverse behavioural outcomes—and may indeed have made a positive contribution in some cases—is unlikely to be the last word on the subject. Proving a negative correlation is notoriously difficult (technically, impossible), and other studies have offered alternative claims. The author is not an epidemiologist, and the purpose of this article is not to evaluate the competing empirical claims. However, the question of what information and advice healthcare practitioners ought to present to pregnant women, or prospectively or potentially pregnant women, in a situation of uncertainty is one to which healthcare ethicists may have a contribution to make. In this article, it is argued that the total abstinence policy advocated by the UK’s Department of Health, and even more stridently by the British Medical Association, sits uneasily with recent data and is far from ethically unproblematic. In particular, the “precautionary” approach advocated by these bodies displays both scant regard for the autonomy of pregnant and prospectively pregnant women and a confused grasp of the principles of beneficence and non-maleficence.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

The publication of the latest contribution to the alcohol-in-pregnancy debate, and the now customary flurry of media attention it generated, have precipitated the renewal of a series of ongoing debates about safe levels of consumption and responsible prenatal conduct. The University College London (UCL) study’s finding that low levels of alcohol did not contribute to adverse behavioural outcomes—and may indeed have made a positive contribution in some cases—is unlikely to be the last word on the subject. Proving a negative correlation is notoriously difficult (technically, impossible), and other studies have offered alternative claims.

I am not an epidemiologist, and the purpose of this article is not to evaluate the competing empirical claims. However, the question of what information and advice healthcare practitioners ought to present to pregnant women (which should for these purposes be taken to include prospectively and potentially pregnant women) in a situation of uncertainty is one to which healthcare ethicists may have a contribution to make.


The causal relationship between heavy alcohol consumption during pregnancy and a range of developmental disorders is widely recognised. Considerably more doubt, however, continues to surround the effects of light and moderate consumption during pregnancy. Until recently, the official position of the UK’s Department of Health (DH) was that pregnant women should ideally not drink alcohol at all, but failing that, should limit themselves to 1 to 2 units of alcohol once or twice a week.1 In mid 2007, however, the DH revised its guidelines, embracing instead a recommendation of total abstinence. As deputy chief medical officer, Dr Fiona Adshead explained:

We have strengthened our advice to women to help ensure that no-one underestimates the risk to the developing foetus of drinking above the recommended safe levels. Our advice is simple: avoid alcohol if pregnant or trying to conceive. This advice could also be included on alcohol packaging or labels. The advice now reflects the fact the many women give up drinking alcohol completely during pregnancy. It is now straightforward and stresses that it is better to avoid drinking alcohol altogether.2

The DH acknowledged that “the revised advice is not a result of new scientific evidence”, but said that it had “revised the advice to make it easier to understand and to provide consistent advice across the UK”.

Shortly thereafter, the British Medical Association (BMA) published guidance that went even further.3 While the DH advised abstinence as the ideal but acknowledged the 1- to 2-drink limit as the next best alternative, the BMA offered an even more strident, “abstinence-only” message. Indeed, its recommendations went beyond advising pregnant women not to drink, arguing instead that “the only sure way” to prevent fetal alcohol disorders is “to stop alcohol consumption during pregnancy” (p31),3 and advocating increased taxation on alcohol as a means to achieving this.

To complicate matters further, the National Institute for Health and Clinical Excellence (NICE) issued guidance in March 2008 recommending:

If women choose to drink alcohol during pregnancy they should be advised to drink no more than 1 to 2 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.4


It is in this slightly ambiguous context that the research paper by Kelly and colleagues appeared.5 The study of 18 553 households aimed to determine if a correlation could be found between light alcohol consumption during pregnancy and cognitive and behavioural deficits in children at age 3 years. The study’s main finding was that no such correlation could be found:

… we have shown that at 3 years of age children born to mothers who drank not more than 1–2 drinks per week or per occasion during pregnancy were not at increased risk of clinically relevant behavioural problems or cognitive deficits compared with children whose mothers did not drink (p134).5

This should perhaps not have come as a total surprise. One year earlier, a meta-analysis of publications on the subject between 1970 and 2005 found that “at relatively low amounts of alcohol and infrequent occasions of binge-drinking, there is no consistent evidence of adverse effects.”6 And even some publications that have advocated a total abstinence policy have conceded that “there have been no known cases of damage to the fetus from a few drinks in early pregnancy.”7

The more surprising feature of the UCL study, however, was that the children of light-drinking mothers tended to have better outcomes than the children of total abstainers:

Boys born to mothers who had up to 1–2 drinks per week or per occasion were less likely to have conduct problems … and hyperactivity … These effects remained in fully adjusted models. Girls were less likely to have emotional symptoms … and peer problems … compared with those born to abstainers. These effects were attenuated in fully adjusted models. Boys born to light drinkers had higher cognitive ability test scores compared with boys born to abstainers (p129).5

Correlation, of course, does not imply causation, and the authors of the UCL study were quick to point out possible alternative explanations for these results:

Children’s social and emotional behaviours and cognitive abilities are heavily influenced by the social environment, and in this study population light alcohol consumption is a marker of relative socio-economic advantage. Therefore, it might be that these social circumstances, rather than the direct physico-chemical impact of ethanol, may be responsible for the relatively low rates of subsequent behavioural difficulties and cognitive advantage in children whose mothers were light drinkers (p138).5

It is far from clear, then, that it is alcohol consumption per se that resulted in these better outcomes. However, while the UCL study did not attempt to explain the correlation between light drinking and better cognitive and behavioural outcomes, it did at least identify such a correlation; on the basis of this research, there appears to be no evidence whatever of a correlation between light drinking and bad outcomes.

One of the most interesting aspects of the study’s publication was the near instantaneous reaction it drew from the office-bearers of the BMA. On the day that the study was published, Dr Vivienne Nathanson, head of science and ethics, was quoted on BBC Online, saying:

We are concerned that the findings from the UCL study may lull women into a false sense of security and give them the green light that there is no problem with drinking during pregnancy. This is not the case. So-called “heavy” and “moderate” drinking harm the unborn baby. Very light drinking may or may not. The BMA believes the simplest and safest advice is for women not to drink alcohol during pregnancy.8

This approach—that, while matters remain uncertain, it is best to advise abstinence—is in keeping with the stance taken by the Department of Health. Furthermore, it has frequently been advocated in the medical literature. (See, for example, Bailey and Sokol7 and Mukherjee and Turk.9) In the face of uncertainty as to whether a safe drinking level exists, and what it might be, there is an obvious appeal to an approach that appears to err on the side of safety. There are, however, a number of reasons why such an approach should give us cause for concern.


First, it may be seen to pay scant respect for the autonomy of pregnant women. It is doubtless trite to observe that recent decades have seen a shift in medical ethics from beneficent paternalism to a prioritisation of autonomy and choice, but it is worth remembering that an indispensable aspect of this respect for choice and autonomy involves conveying sufficient information to a patient to render any choice meaningful.10 The General Medical Council’s 2008 consent guidelines state, “For a relationship between doctor and patient to be effective, it should be a partnership based on openness, trust and good communication.”11 And the BMA’s own advice is that “[d]octors should respond honestly to direct questions from patients and, as far as possible, answer questions as fully as patients wish.”12

Where the available evidence does not permit of a straightforward, unambiguous answer to a question like “how much is it safe to drink during pregnancy?”, an honest response would be one that communicates that uncertainty. Vivienne Nathanson’s response to the UCL study was to note that “So-called “heavy” and “moderate” drinking harm the unborn baby. Very light drinking may or may not.” So what is the problem with telling pregnant women this? If this is the current state of knowledge, then a commitment to honesty and accuracy involves explaining this to the patient.

A “partnership based on openness, trust and good communication” does not involve substituting a true, complicated account of risk with a more simplistic, but less accurate, one. As one doctor has argued, “If we in the medical and midwifery professions have failed to communicate clearly to women the meaning of safe limits, then we need to put this right—not take the easy option (for us) and ban alcohol completely.”13 The days when doctors routinely withheld information—about risks, alternatives, side effects or prognoses—on the grounds that patients would become confused and make bad decisions are, supposedly, consigned to history. It is far from clear why a paternalistic exception is permitted in the case of pregnant women.


Respect for autonomy may be the pre-eminent bioethical principle of our age, but it is not the only principle. Is it possible that some sacrifice of autonomy is justified by the demands of beneficence or non-maleficence? The problem here is that a closer examination of the DH/BMA approach seems to cast doubt on the extent to which either of these principles is truly furthered.

As news of the UCL study circulated in the British press, Vivienne Nathanson responded by reiterating the BMA’s previous line: “So-called “heavy” and “moderate” drinking harm the unborn baby. Very light drinking may or may not.” Conspicuously absent from this statement was any recognition of the UCL study’s finding that “Boys born to light drinking mothers were less likely to have conduct and hyperactivity problems and … had higher scores on cognitive ability assessments, and for the test on colours, shapes, numbers and letters” (my emphasis) (p134).5

Is it possible, then, that low levels of alcohol exposure could actually benefit the resulting child? Other possible explanations may seem intuitively more likely; specifically, as noted above, it may then be that it is the socioeconomic class of the light-drinking mothers that were responsible for the higher performance among this cohort of children, rather than the exposure to low levels of alcohol.

There again, those mothers who had drunk heavily, and whose children had worse outcomes, were disproportionately likely to be from lower socioeconomic groups, rendering it difficult to disentangle the variables in those cases where heavy drinking was thought to have made matters worse. Once again, the unavoidable conclusion to which all of this leads is that not enough is known to be sure. Some (tentative) evidence, from previous studies, seems to suggest that low levels of alcohol exposure may cause harm, but other (tentative) evidence seems to suggest the opposite, that it may in fact benefit the resulting child. Faced with such contrasting possibilities, it cannot plausibly be argued that advocating total abstinence is erring on the side of caution; if the UCL study is right, pregnant women cutting out their one or two drinks a week may be making things worse.

There may be those who would argue that failing to benefit someone is not ethically synonymous with harming them, and that therefore the risk of harming an otherwise healthy child should be prioritised over the risk of failing to benefit that child. This is not the place for a detailed examination of whether beneficence and non-maleficence are, in general, equally compelling ethical imperatives, but it is at least plausible to argue as much in the context of a parental duty of care. If a mother-to-be owes a duty to her child to minimise its chances of cognitive and behavioural problems, then it seems that she is subject to a prima facie obligation to follow whichever course of conduct—either stopping drinking altogether, or continuing to drink a small amount—is most likely to bring about that outcome.

Thus, if it turns out that low-level alcohol exposure bestows a positive benefit—perhaps by lowering maternal stress levels?—then, in the same manner as there is now a societal expectation that prospectively pregnant women will take folic acid supplements, so too might we come to expect them to take “alcohol supplements”. Of course, as even the researchers acknowledge, we are a long way from being able to conclude any such thing. The point is, however, that where

  • choice X may lead either to good or bad consequences,

  • there is no good reason to suppose either outcome is more likely than the other and

  • the good and bad consequences are about equally weighty,

the principle of precaution cannot plausibly be invoked either in favour of, or against, choice X.


In the preceding section, I referred quite consciously to prima facie duties owed by pregnant women to their future children. Of course, the welfare of those children is not the only valid ethical concern in such matters, and the pregnant woman herself must not be overlooked. Is a pregnant woman’s desire to imbibe alcohol the sort of interest to which weight should be attached? I would argue that an action that gives pleasure with no countervailing harm must be one that should at least be considered in any ethical reckoning. For many people, an appreciation of alcohol—fine wines, rare malt whiskys, real ales—transcends a mere means to inebriation, and constitutes a real source of aesthetic and intellectual richness in their lives. For others, the Friday night glass of shiraz or gin and tonic may just be a relaxing ritual that signals the boundary between the working week and the leisure of the weekend.

For an increasing number of people, however, it seems as though the quality of life of the pregnant woman is so far behind the interests of the future child in any scale of ethical relevance as to be beneath consideration altogether. As one respondent to an article on the subject put it, “It is time society grew up and put the health of future generations before profiteering and hedonism.”14

It is certainly not being suggested here that pregnant women should take a serious risk with the health of future children, rather than give up a merely pleasurable activity for a few months. When the risk is non-existent, however, the sacrifice becomes pointless. Furthermore, the sacrifice begins to look less than trivial when considered alongside the ever-widening range of foods, drinks and activities—coffee, tea, chocolate, shellfish, unpasteurised cheeses, liver and sunbathing, to name but a sample—that pregnant women are instructed to cut down on or cut out.

It should also be remembered that the advice regarding abstinence from alcohol applies to women who know they are pregnant and also to those who are attempting to become pregnant—a process that may take many months or even several years. Indeed, this has led some to argue that all women of childbearing age should avoid alcohol; the United States surgeon general has issued the following advice:

Recognizing that nearly half of all births in the United States are unplanned, women of child-bearing age should consult their physician and take steps to reduce the possibility of prenatal alcohol exposure … Health professionals should inquire routinely about alcohol consumption by women of childbearing age, inform them of the risks of alcohol consumption during pregnancy, and advise them not to drink alcoholic beverages during pregnancy.15

Is singling out one sex for particular monitoring and lecturing from healthcare professionals a legitimate cause for concern? It may be thought that, at the very least, we should require some reasonably compelling ethical justification for what is, on the face of it, a straightforwardly sexist policy. To date, the evidence suggests that we have no such justification.


One final potential problem of the BMA’s “total abstinence” position is that it may backfire, serving to erode the trust the public places in medical advice. The BMA expresses concern that anything less than a recommendation of total abstinence risks being interpreted as a “green light” for heavier drinking. But the opposite possibility must also be considered—that if they are seen to be exaggerating risks that recent (and well-publicised) studies have shown to be negligible, their advice on genuine risks will carry less authority. It may, in short, be important for bodies like the DH and BMA to consider whether they are in danger of “crying wolf”, with the attendant risk that genuine warnings of real dangers will be treated less seriously.

It is also worth considering whether constant additions to the list of do’s and don’ts may result in a sort of “advice fatigue”. As the list of behaviours that pregnant women are advised to avoid continues to grow, so too will the likelihood that more pregnant women will end up, as one pregnant GP has put it, “feeling rather picked on”.16 The danger, perhaps, is that, in the face of unduly onerous requirements, and what are coming to be seen as unwelcome intrusions into previously private matters, many of them will abandon reliance on medical advice and epidemiological research altogether and simply rely on anecdote or instinct.

Again, it is not being suggested here that evidence-based advice about significant risks should be diluted or withheld because of the risk that pregnant women may feel uncomfortable; such an approach would also be incompatible with respect for their autonomy. The point is simply that burdening pregnant women with ever more lifestyle obligations may not be without an ethical cost.


The communication of uncertain risks presents a significant challenge for the healthcare profession. If that profession’s commitment to patient autonomy is at all serious, however, it is a challenge that cannot be ducked. If the best evidence is that low-level alcohol consumption presents negligible risk to the fetus, then respect for autonomy means that such information must be communicated to pregnant women. If uncertainty persists, then that uncertainty must itself be communicated. If evidence suggests that women are misunderstanding the information, it is incumbent on medical professionals to consider better mechanisms for getting accurate information across; it is not reasonable to replace more accurate information with less accurate merely because it is simpler to communicate.

To continue preaching total abstinence because of a fear that women will misunderstand the truth, or regard a reassuring message about low-level consumption as a “green light” for unrestrained overindulgence, is patronising and paternalistic to a degree that is hard to reconcile with any real respect for autonomy and informed decision-making. Furthermore, it risks alienating or worrying women who are at very low risk, while having a negligible impact on high-risk drinkers who ignored the previous guidelines anyway. And—contrary to what appears to be widespread medical opinion—attempting to “guilt-trip” women out of a harmless and pleasurable activity is not entirely ethically unproblematic, especially in the context of an already stressful time when they already feel besieged by demands to modify their behaviour.

Finally, where the evidence is that a behaviour may bestow some small benefit or inflict some small harm (but will probably do neither), it is not ethically coherent to “err on the side of caution” by advising abstinence, as to do so would be to regard non-maleficence as automatically more important than beneficence. For a parent, at least, failing to do good for a child is not relevantly distinct from avoiding harm.



  • Competing interests: None declared.

  • Provenance and peer review: Not commissioned; externally peer reviewed.

Other content recommended for you