There is uncertainty as to whether there is a safe threshold for drinking alcohol during pregnancy. We explored pregnant women’s attitudes towards drinking alcohol in pregnancy and their attitudes towards sources of information about drinking in pregnancy following recent changes in UK government guidance.
Methods: A qualitative study involving individual, semi-structured interviews with 20 pregnant women recruited from community organisations in the UK.
Interview transcripts were analysed qualitatively using thematic analysis.
Results: Most women found information and advice about safe levels of drinking in pregnancy confusing and lacking in evidence and detail. Although most women considered that there were risks involved with drinking in pregnancy and these perceptions influenced their behaviour, only six women reported abstinence.
Women reported being influenced by advice from family and friends and their experiences of previous pregnancies. Many had received no individual advice from general practitioners or midwives relating to drinking during pregnancy.
In order to do so, they require clear and consistent advice about safe levels of drinking from policy makers and health professionals. This is an important issue as women might drink socially during their pregnancy.
Alcohol can readily cross the placenta and interfere with fetal development. High levels of alcohol consumption in pregnancy are associated with harmful effects such as fetal alcohol syndrome . In recent years, the debate has focused on the safety of light drinking during pregnancy although the research findings are inconclusive [2-8]. Some studies [6,7] suggest an association between light drinking and childhood behavioural problems but a systematic review found no consistent evidence of adverse effects across a range of outcomes . More recently, a large epidemiological study also found no evidence of harmful effects on child behaviour or learning . Hence, a safe threshold for drinking in pregnancy has not been conclusively established [9-11]. The proportion of women of child-bearing age in the UK who drink over 14 units of alcohol per week (one unit is equivalent to one glass and contains 8 grams of alcohol) has increased in recent years . Consequently, the fetus is most likely to be exposed to alcohol in the first trimester, before pregnancy recognition . These findings suggest that alcohol consumption during pregnancy is an important public health issue.
Government advice regarding alcohol consumption in pregnancy varies internationally. In many countries, including the US, Canada and France, complete abstinence is recommended. Until recently, the UK Department of Health guidance  was that pregnant women should limit themselves to 1-2 ‘units’ of alcohol once or twice a week and avoid getting drunk. In May 2007, the guidance was adjusted to recommend abstinence . This primarily reflected the greater clarity of a ‘no drinking’ message rather than the research evidence . However, the guidance also stated that, if pregnant women choose to drink, they could drink up to the previously recommended limits. The implications that this guidance will have on pregnant women’s drinking behaviour are unclear.
Further uncertainty is created by the differing guidance provided by other UK health organisations. The Royal College of Obstetricians and Gynaecologists  have recommended a limit identical to the previous Department of Health guidance. In contrast, the British Medical Association  expressed concern that government guidance and drinking limits could be misinterpreted and advised pregnant women to avoid alcohol. This uncertainty was highlighted when National Institute of Clinical Excellence (NICE) draft clinical guidelines for antenatal care initially suggested that pregnant women should limit their alcohol intake to less than 1.5 units per day . However, following consultation, the recommendation was revised to maintain greater consistency with the Department of Health guidance . These discrepancies in advice as well as research findings have generated vast media coverage, often involving conflicting opinions. Pregnant women are likely to receive information from sources such as news reports, magazine articles, online information and advice from health professionals, family, and friends. Inconsistent reporting of information could lead to considerable uncertainty and anxiety  for pregnant women about whether it is safe to drink and which guidelines to follow.
Most research exploring pregnant women’s attitudes towards drinking in pregnancy has been quantitative [20,21]. These studies have found that pregnant women are aware that alcohol can harm their unborn babies but most believed that some alcohol intake during pregnancy was acceptable [20,21]. Such research suggests women may be reluctant to follow guidance recommending complete abstinence.
A qualitative approach could provide a richer understanding of women’s attitudes towards alcohol consumption during pregnancy and can give insight into previously unexplored areas . A better understanding of women’s attitudes towards sources of information and government guidance is essential if health information is to be made clinically relevant and credible for them. In this exploratory study, we aimed to explore pregnant women’s attitudes towards alcohol consumption during pregnancy and their attitudes towards sources of information and advice about drinking in pregnancy.
Sample and Procedure
In late 2007, a few months after the Department of Health guidance had changed, we purposively sampled a group of women who engaged in antenatal care in order to explore their range of knowledge and attitudes. We anticipated that some of these women might drink socially during their pregnancy [6,8]. Pregnant women in Nottingham and London were recruited from a range of community organisations including Sure Start Children Centres (antenatal input is also provided here), National Childbirth Trust antenatal groups and mother and toddler groups. Details about the study were sent to these community organisations. Co-ordinators of the organisations that agreed to take part distributed information packs about the study to pregnant women. These packs included the study Information Sheet, Consent Form and contact details form. All women were given the option of individual telephone or face-to-face interviews. We used individual interviews as this allowed participants anonymity, confidentiality, and the opportunity to express themselves freely in a non-threatening forum. Ethics approval was received from the University of Nottingham Medical School Research
Twenty women agreed to participate within the study period and all requested telephone interviews. These semi-structured interviews were carried out by the same interviewer and lasted between 20-40 minutes. Informed consent was obtained prior to the interview and participants returned their signed consent forms by post. The interviews were audio recorded and transcribed verbatim. Women were asked open questions using a topic guide covering the following areas: information received about drinking during pregnancy; the influence of health professionals, friends, family, and the media on their drinking; their attitudes towards government and other advice; and their views about drinking during pregnancy and available health information. This provided standardisation between the individual interviews. During the interview, women were also asked some questions about their demographic status to establish sample characteristics and some closed questions about their current and prepregnancy levels of alcohol consumption and their knowledge of Government advice. Responses to the closed questions are reported descriptively in the Participant Characteristics.
The interview transcriptions were analysed using thematic analysis . Theme analysis was chosen for its flexibility which allows a full exploration of the data without the constraints of methods such as interpretative phenomenological analysis or grounded theory analysis which are more theoretically bounded . With the collected data, theme analysis enabled the identification and description of barriers and facilitators to drinking in pregnancy. Theme analysis allows themes to be produced both deductively where they are generated by the researcher based on previous theory and research literature before analysis and inductively, from the raw data itself . Initially, the transcripts were thoroughly read by N.R. to become familiarised with the ideas and attitudes expressed. The transcripts were then systematically searched using NVivo7 (QSR International) to identify all codeable moments. These codeable moments were grouped together to create themes which were reviewed, debated and refined methodically by three researchers (N.R., C.S., K.S.) resulting in eight final themes. Four themes were identified inductively and four were generated deductively based on past theory and research.
Once the themes were identified, a thematic code for each was developed, Each thematic code satisfied Boyatzis’ five elements including: a conceptually meaningful label, a definition, a description of how to know when the theme occurs, a description of any qualifications or exclusions to the application of the theme, and examples of positively and negatively coded extracts from the data . These thematic codes were put together to form a code book. In order to establish the reliability of the themes, an experienced independent researcher used the code book to code a sample of 21 extracts selected to represent all eight themes. Inter-rater agreement was high with concordance for 18/21 (86%) extracts.