Alcohol brief interventions in Scottish antenatal care: a qualitative study of midwives’ attitudes and practices

NHS care in Scotland is provided through 14 regional Health Boards, which plan and deliver health services within their geographical areas, with overall policy directed by the Scottish Government Health Directorates [24]. Participants for this study were recruited from one Scottish Health Board, NHS Lothian, which has two consultant units and about 130 community midwives overall. Our sample of community midwives was drawn from the 34 midwives based at one of the consultant units. Within NHS Scotland, community based midwives provide the majority of care to women during the antenatal and post-natal periods, involving the women’s General Practitioner or Obstetrician as appropriate based on the woman’s needs and local care pathways [25]. As such, they are well placed to screen and deliver ABIs to pregnant women. At the time of data collection in mid to late 2010, the midwives who participated in this study informed us that nearly all midwives in NHS Lothian, particularly community midwives, were trained to screen and deliver ABIs. The training involved half a day face-to-face screening and ABI skill training. All community midwives involved in alcohol screening and delivery of ABIs were considered for inclusion. This study purposively sampled midwives with a range of roles - community midwives (middle grade clinical midwives) and their team leaders, as well as consultant midwives (senior clinical midwives with a remit for professional leadership, evaluation and education). Details of the study in the form of information packs were delivered to three midwifery team leaders to be given to all midwives within their areas. The information packs contained an invitation letter, information sheet (including details of incentives), an expression of interest form and self-addressed stamped envelope. All midwives were given the option of individual interviews or focus group to encourage participation. Informed consent was obtained prior to interview or focus group. Participants were compensated for their time and effort with either lunch or £20 in high street store vouchers.

Semi-structured interviews were used and the topic guide (available on request) covered a range of areas including, attitudes to alcohol use in pregnancy and opinions about the appropriate timing for screening and delivering ABIs. Midwives were encouraged to describe their practice in relation to screening and delivering of ABIs and to identify barriers and facilitators. Their views about training, support and the implementation process were also sought. All interviews and the focus group were conducted and digitally recorded by the first author, who also transcribed the data. The second and third authors reviewed the transcripts. Anonymous unique codes were used to represent midwives who participated in the study.

The interview transcripts were analysed using a thematic analysis which employed a hybrid approach of inductive and deductive coding and theme development [26]. This combined the data-driven inductive approach [27] and deductive a priori code template [28]. Data analysis proceeded in six main stages as follows: development of a code manual; test reliability of codes; identification of initial themes emerging from the data; utilization of the code manual to apply codes to entire transcripts whilst noting emerging codes; connection of codes into themes and corroboration of themes through interpretation of analysis [26].

For a code to be credible, it must capture the qualitative richness of the phenomenon [27]. Using NVivo [29] coding was undertaken by selecting the appropriate segments of text and coding them appropriately. Several codes were generated. New codes were devised as new ideas emerged from the data. When new codes were identified, previous transcripts were re-read to determine if the new codes were applicable to the texts. Thus, the code manual was continually revised. A final code manual was then produced. The reliability of coding is enhanced if two or more researchers independently code a qualitative transcript [30]. For that reason, the first author coded all transcripts and the second and third authors coded half of the transcripts. Generally, inter-coder agreements were high, with over 80% agreement on coded extracts. Minor disagreements were resolved through discussion between the authors. Through the connection of similar codes, themes emerged. Agreements and disagreements in opinions between segments of the data were informative at this stage. It was also important to ensure that themes were representative of the original data [26]. Consequently, transcripts were re-read ensuring that themes had appropriately captured their phenomena. With the emergence of patterns of meaning, similarities and differences within units of the data were explored before the analysis progressed to the interpretative phase.

The West of Scotland Research Ethics Committee 2 approved this study.

Midwives ages ranged between 24 and 56 years. All were clinical midwives, eight were midwifery team leaders and two were consultant midwives. Their years of midwifery practice ranged between 3 and 33.

Seven major themes were identified. Individual interview and focus group data are presented together.

Prompted by previous research findings that there were uncertainties regarding the effects of low-moderate levels of drinking on the fetus, we explored how this message had influenced midwives’ attitudes to prenatal alcohol consumption. We anticipated that midwives’ attitudes may either directly or inadvertently influence their decisions about the advice they offered to women, or the manner in which they delivered ABIs. It was apparent that uncertainties regarding the threshold at which alcohol could harm the baby prompted many of the midwives to take a cautious stance in relation to advising abstinence from alcohol during pregnancy (see Table 1). During the interview, midwives sometimes spoke about their own alcohol use to emphasize a point or convey an opinion and this was used to understand their attitudes towards the issue. Most of the midwives who indicated that they did not drink alcohol had strong views against drinking in pregnancy while a few of the midwives who indicated that they did drink alcohol appeared sceptical regarding the current push for abstinence in pregnancy.

We hypothesized that midwives’ knowledge and understanding of risk could have an influence on the level of priority they accorded to the issue of drinking in pregnancy in the midst of other competing work pressures. Generally, midwives showed a good understanding of the effects of alcohol on the fetus. All the participants knew that heavy sustained drinking is associated with fetal alcohol syndrome. However, there were divided opinions regarding the effects of low-moderate level of drinking on the fetus. Drinking at all stages in pregnancy poses varying risks for the mother and the unborn child [31, 32]. We further explored midwives views on the impact of timing of alcohol consumption during pregnancy on subsequent developmental outcomes. It was clear that most considered drinking in the first trimester to be more risky, which is congruent with the evidence [33, 34]. However, a few thought that drinking in the first trimesters possibly posed the least risk to the fetus (see Table 2). They reasoned that quite a lot of women had consumed alcohol during the early stages of pregnancy while they were unaware that they were pregnant, yet it appeared that their children were rarely harmed.

The midwives mentioned that most of the pregnant women who drank alcohol did so during the period when pregnancy was unconfirmed. Once pregnancy was confirmed, they discontinued taking alcohol. They all agreed that pre-conception alcohol prevention strategies would be more beneficial to the fetus than alcohol interventions delivered in antenatal care. However, midwives recognised that without a specific target audience such interventions might struggle to have an impact. Some participants expressed the opinion that pregnant women who planned their pregnancies were more likely to benefit from pre-pregnancy interventions (see Table 3). Planning of pregnancy could mean that women would have been more likely to have explored ways to maximise their chances of having a healthy pregnancy. For example, this could involve contacts with family planning clinics regarding advice on discontinuing contraceptive use or contacts with fertility clinics. Midwives believed that some of these pre-pregnancy opportunities would have been more appropriate ‘teachable moments’ because of their emphasis on health issues, particularly those of the unborn child.

Almost all the midwives indicated that tackling alcohol use in pregnancy was part of their role. Nevertheless, some mentioned that the screening and ABI program had been added to their already considerable workload. One of the rationales for the program was to train midwives to intervene with women who drank at risky levels, but for whom drinking had not reached the dependence level that might require alcohol specialist’s attention. However, a few midwives misconstrued this rationale (see Table 4). There was also a sense that midwives felt that their role was to advise women to be abstinent. This is not entirely in line with the motivational interviewing approach of delivering ABIs, which does not imply a pre-determined decision about what is best for the patient.

At the time of data collection most of the midwives had completed their screening and ABIs training at least six months previously, but it was clear that most of them had not had enough opportunities to practice their skills because they perceived that few women admitted to alcohol consumption in pregnancy. However, it appeared that in some instances the opportunity to deliver ABIs at the antenatal appointment was missed.

The midwives felt that the assessment to tease out the nature and quantity of alcohol consumed by women was beneficial. Previously, no such detailed assessment was conducted and some midwives believed that the program had improved the quality of identification. Further, midwives felt that the additional support and resources provided by the implementing organisation (NHS Health Scotland) had had positive impact. In addition to the ABI skill training, midwives were also provided with an ABI antenatal professional pack. The pack included information about the evidence base of ABIs, screening tools, prompt cards and a drink calculator (for estimating alcohol units).

Midwives believed that they were well positioned strategically to deliver health information or ABIs, and that the type of services they provide meant women were more likely to respond to them in terms of accepting health behaviour interventions. However, they believed that they had to be careful in order not to alienate women when discussing sensitive alcohol issues.

Most midwives felt that screening and delivery of ABIs at the first antenatal appointment may not be appropriate because of the potential impact it could have on midwife-pregnant woman relationships. They expressed the feeling that there could be issues with trust at the first appointment and that it would be difficult for women to divulge sensitive issues at that point. One midwife likened the issue of trying to identify pregnant woman who have been drinking at risky levels at the first antenatal visit, to that of identifying a domestic violence victim (see Table 5). In spite of this, midwives felt that in the context of antenatal care, the first appointment was the best opportunity to deliver ABIs.

Midwives reported that the program had several positive aspects, for not only women and their unborn child but also for the value it had added to their practice. Midwives’ overall view was that screening and ABIs had generated increased awareness about alcohol use in pregnancy. Some viewed their place in antenatal care as appropriate because pregnancy is a time when women are more motivated to change negative health behaviours. However, others indicated that offering ABIs in antenatal care was a bit late. A few midwives suggested that ABIs may not necessarily be of benefit to the current pregnancy but might benefit subsequent pregnancies and future drinking behaviour. One participant suggested that providing ABIs, particularly to women who consumed excessive amounts of alcohol before pregnancy might be helpful in terms of equipping them with knowledge that could facilitate healthy drinking behaviour after they had had their baby (see Table 6).

They also highlighted the potential benefits of screening and ABIs to the child, as well as financial benefits to the family. All midwives recognised the added benefits of the program to their practice in relation to their confidence in discussing alcohol-related issues.

Midwives were asked to identify the main challenges that the implementation of the program had had on their practice or some of the difficulties they had experienced since its inception. It was clear that workload pressure and time constraints were the most significant barriers. Midwives indicated that in the midst of competing workload priorities, alcohol interventions remained a low priority. The difficulty of converting different types of alcoholic drinks into standard units was also seen as a problem and in the focus group all participants agreed with that view (see Table 7).

Some midwives were worried that the amount of information provided at the first antenatal appointment may adversely influence pregnant women’s capacity to assimilate all relevant information, in particular when a behaviour change intervention is involved. The social intricacies surrounding the issue of drinking in pregnancy were also stated as a barrier. A few midwives suggested that because drinking in pregnancy is generally considered undesirable in the Scottish culture, this probably discourages women from disclosing their true consumption levels.

Public health interventions are generally complex and context-dependent and their transferability between countries, localities or settings requires careful consideration. What has worked for one group may not necessarily work for another because people’s values, behaviours and priorities differ [47, 48]. Although, screening and ABIs have been evaluated extensively in primary care, their utility is limited in antenatal care and emerging evidence of their usefulness in this setting is often from countries outside the UK. Ongoing evaluation would be required to improve the program in the Scottish antenatal setting in terms of raising the priority of ABIs in the first antenatal appointment. Also, this study has showed that when examining outcomes, for instance, uptake of ABIs by midwives, it is important to consider issues in the process of implementing the intervention.

Pregnancy is a period where women are generally motivated to change risky behaviours. Therefore, pregnant women who still continue to drink at risky levels possibly require more extensive interventions, for example extended brief interventions to build their confidence to change their drinking behaviour [4, 10].

Alcohol brief interventions do not only mean giving advice about drinking. They have structure and may require a ‘motivational style’ drawn from motivational interviewing. The practice of motivational interviewing requires a high level of skills and extensive training [49]. Effective practical assessment tools that can measure the integrity of practitioners’ motivational interviewing skill, for example, the motivational interview treatment integrity code could be a useful tool to assess and improve midwives’ motivational interviewing practice [50]. Good motivational interviewing or person-centred communication skills might also help to overcome barriers such as lack of midwife-pregnant woman rapport and facilitate screening and ABIs delivery at the first antenatal appointment.

Knowledge of risks (for example, fetal alcohol syndrome) is known to induce drinking behaviour change in pregnancy [51]. When training midwives to screen and deliver ABIs emphasis should be placed on the severity of some of these adverse fetal conditions. This may enhance the urgency of alcohol intervention activities in antenatal care. Currently, there is consistent evidence of effects of heavy drinking on the fetus. However, the evidence relating to low-moderate drinking is still inconclusive and our study has highlighted that there may be still scepticism among some midwives. There is therefore a need for more robust studies to establish the threshold and specific fetal risks that could results from these levels of drinking and to enhance communication of research findings to health practitioners.

This study was conducted in only one Health Board in Scotland and represented the views of a relatively small number of midwives. The findings of the study although not generalisable to all midwives delivering ABIs, may still be of interest to those involved in implementing similar interventions in antenatal care settings in any country. Considering the approach for recruiting participants, it was possible that midwives who felt knowledgeable about screening and ABIs were more likely to participate in the study. Therefore, midwives who did not participate may have had different views.

Both individual interviews and focus groups were conducted. Individual interviews offered participants the privacy of sharing opinions that could not be easily discussed in a group setting, especially as the program was a policy initiative. However, the group setting facilitated the development of ideas and debate and moved the discussion into dimensions which individual interviews were unable to achieve. We were keen to carry out more focus groups but unfortunately, by the nature of their work, community midwives are geographically dispersed and it was logistically challenging to conduct more focus groups. Concerning the analysis, because only one focus group was carried out, we felt it was inappropriate to analyse the data separately from the individual interview data. As a result, the two sets of data were analysed together, retaining the individual quotes from the focus group.