Healthcare professionals’ views on two computer-based decision aids for women choosing mode of delivery after previous caesarean section: a qualitative study

Authors


Dr KM Rees, Academic Unit of Primary Health Care, Bristol University, 25-27 Belgrave Rd, Clifton, Bristol BS8 2AA, UK.
Email kateyrees@hotmail.com

Abstract

Objective  To explore healthcare professionals’ views about decision aids, developed by the DiAMOND study group, for women choosing mode of delivery after a previous caesarean section.

Design/Methods  A qualitative focus group study. Data were analysed thematically.

Setting  Two city maternity units, surrounding community midwife units and general practitioner (GP) practices in southwest England.

Sample  Twenty-eight healthcare professionals, comprising obstetricians, hospital and community midwives and GPs, who participated in six focus groups.

Results  Participants were generally positive about the decision aids. Most thought they should be implemented during early pregnancy in the community, but should be accessible throughout pregnancy, with any arising questions discussed with an obstetrician nearer to term. Perceived barriers to implementation included service issues (e.g. time pressure, cost and access), computer issues (e.g. computer literacy) and people issues (e.g. women’s prior delivery preferences and clinician preference). Facilitators to implementation included access to more standardised and reliable information and empowerment of the user. Self-accessing the aids, increased awareness of decision aids among healthcare professionals and incorporation of aids into usual care were suggested as possible ways to improve implementation success.

Conclusions  This study gives insight into healthcare professionals’ views on the role of decision aids for women choosing a mode of delivery after a prior caesarean section. It highlights potential obstacles to their implementation and ways to address these. Such aids could be a useful adjunct to current antenatal care.

Introduction

In recent years, greater emphasis has been placed on women’s choice and involvement in decision making during pregnancy and childbirth.1,2 The National Institute of Clinical Excellence Guideline for Caesarean Section published in 2004, states that decisions about mode of birth after caesarean should consider maternal preferences and priorities and include discussion of risks.3 A recent study suggested that 60% of women having a repeat section may be unaware of alternative delivery options.4 Qualitative research has found that women want to be involved in the decision-making process and wish for tailored information about delivery options,5 but many experience decisional conflict and anxiety when making the decision.6

There is evidence that decision aids are effective in many clinical contexts.7 A recent randomised controlled trial, the DiAMOND study, found that two decision aids for mode of delivery after a previous caesarean section reduced women’s decisional conflict and anxiety, and increased knowledge of delivery options and their associated risks and benefits.8 Details of the content of the two aids are outlined in Box 1.

Table Box 1..   Content of the information program and decision analysis program decision aids
Both decision aidsInformation programDecision analysis program
Descriptive information about possible risks and benefits associated with each type of delivery (elective caesarean section, emergency caesarean section and vaginal delivery). This included possible physical complications for mother and baby, social and emotional factorsExplicit probability information
Website access
Explicit quantification of user’s views on outcomes
Suggested a preferred method of delivery

However, less is known about implementing decision aids into practice.9–11 Healthcare professionals have been found to be key in the delivery of decision aids.10 Implementation strategies for decision aids should involve both healthcare professionals and patients.12 While there is some research on users’ views on how aids might be used in practice,13 views on healthcare professionals are rarely sought.10,14,15 Furthermore, their views have only been sought once the decision aid has been implemented into practice. We are not aware of any qualitative studies of healthcare professionals’ views on aids at an earlier stage, when they are still adaptable to being changed or improved.

This qualitative study was undertaken at the end of the DiAMOND study, a randomised controlled trial of two decision aids for women choosing mode of delivery after a previous caesarean section.8 see Box 1. The aim was to explore healthcare professionals’ views about the decision aids and their implementation into practice.

Methods

We undertook six focus groups with health professionals involved in the care of pregnant women in the Bristol area. Ethical approval for the study was gained from the South West MultiCentre Research Ethics Committee.

Setting

Participants were recruited from two hospitals in Bristol, UK and from surrounding community sites, including a community midwife team base and five general practices. The hospitals deliver 4914 and 4922 babies per year and have total caesarean rates of 26 and 23% respectively. Delivery of care to women with a previous caesarean section is similar across locations, with women being referred to a hospital outpatient clinic to discuss delivery with an obstetrician at least once during their pregnancy.

Participants

Trainee and consultant obstetricians, hospital and community midwives and general practitioners were approached for inclusion in the study. Sampling was purposeful and used a maximum variation strategy16 to include a range of professionals involved in the care of pregnant women with a previous caesarean section history. We invited all obstetric medical staff in each hospital to participate, through a nominated clinical contact. For midwifery staff, we asked team leaders to nominate staff, requesting that a range of seniority and experience be represented. General practitioners were recruited by a snowball sampling technique16 through networks of contacts in local general practices. We formally invited those who agreed to be contacted. In total, written invitations were mailed to 50 prospective participants, with background information about the study, consent forms and a questionnaire requesting information on demographics and participant’s availability. Respondents were invited to participate in a focus group.

Data collection

Focus groups were chosen to enable us to explore healthcare professionals’ views at individual and group levels, to explore areas of consensus and disagreement both across and between groups and to create a comfortable environment for participants to share their views in the presence of others from similar professional backgrounds.17

Six uni-professional focus groups, each including between four and five participants, were conducted to enable saturation of themes to occur.18 The groups comprised: two groups of hospital midwives, one group of community midwives, two groups of obstetricians and one group of general practitioners. They took place at participants’ workplaces between June and November 2007 and were facilitated by KR, a healthcare professional with no personal link to the development of the decision aids or the preceding randomised controlled trial. A co-facilitator (KB) was also present at three of the groups. The discussions lasted approximately an hour and followed a demonstration of and participant familiarisation with the decision aids, with opportunities for questions.

The group discussions followed a topic guide, which was informed by the literature and the process and outcomes of the preceding trial (see Box 2: Topic guide for focus groups). The guide was used flexibly, enabling all key topics to be covered across all of the groups, while not necessarily in the same order (following the natural progression of the conversations) and allowing group participants to explore emerging issues as appropriate. All focus groups were audio-recorded and transcribed verbatim by a professional transcriber. Written notes on the process and content of the discussion were also taken by the co-facilitator or facilitator to aid recognition of participants for transcription purposes.19

Table Box 2..   Topic guide for focus groups
Presentation of decision aids with opportunities for questions.
Attitudes towards mode of delivery after a previous caesarean section and usual advice given to women.
Views on content of decision aids and how they might be used by women.
Views on implementation of aids
 Are they worth using?
 How could you see them being used within usual care?
 How would they impact on usual care?
 Any obstacles to decision aid use within usual care?
 How could any obstacles be overcome?
Any other issues.

Data analysis

Analysis of the focus group transcripts was thematic and drew on the constant comparative method.20 Following reading and re-reading of the transcripts, ‘open’ codes were initially applied to represent the significance of sections of text. Through continued comparison across transcripts, the codes were developed and grouped into categories and over-arching themes were eventually applied to all the transcripts, assisted by ATLAS.Ti software (ATLAS.Ti, Berlin, Germany).21 Data analysis was led by KR. KB and AS independently read and provisionally coded a subset of the transcripts to ensure the incorporation of different perspectives and to check the credibility of the developing coding framework. The codes were discussed by all team members and the final coding framework applied to the transcripts was agreed by all. An audit trail of the data collection process and development of the coding framework was maintained in a reflexive diary, kept by KR to aid transparency.22

Results

Of the 50 invited healthcare professionals, 34 agreed to participate in the study and 16 did not reply. Six were unable to attend focus groups because of time constraints, so 28 professionals finally took part. The participants’ characteristics are detailed in Table 1. They varied in age, gender, level of experience, country of training, number of own children and method of their children’s delivery. Three participants had seen the decision aids prior to the focus group. Focus group participants may have had women under their care who had used the decision aids during the RCT as they were working in the same locality.

Table 1.   Background information on focus group participants
Focus groupUnique identifierAgeGenderExperience level (years)Country qualifiedNumber of childrenMethod of delivery
  1. GP, general practitioners.

  2. *Seen decision aids before focus group through involvement in prior RCT.8

Midwives hospital AMA140–50F15–20UK2Vaginal
MA220–30F5–10UK0 
MA3*30–40F5–10UK0 
MA420–30F<5UK0 
MA520–30F5–10UK0 
Midwives hospital BMB1*40–50F15–20UK3Vaginal
MB2*50–60F>20UK2Vaginal
MB330–40F5–10UK3Vaginal
MB450–60F15–20UK1Vaginal
Midwives, communityMC140–50F10–15UK2Mix
MC240–50F>20UK2Vaginal
MC350–60F10–15UK2Mix
MC440–50F5–10UK3Vaginal
Doctors hospital ADA120–30F<5UK0 
DA240–50F15–20UK2Vaginal
DA320–30M<5UK0 
DA430–40M10–15NZ1CS
DA520–30F<5UK0 
Doctors hospital BDB130–40F10–15Pakistan0 
DB220–30F<5India0 
DB340–50F15–20India2CS
DB420–30F5–10UK0 
DB520–30F<5UK0 
Doctors, community (GPs)DC130–40M10–15UK1Vaginal
DC230–40F5–10UK0 
DC330–40M5–10UK1Vaginal
DC430–40F5–10Australia0 
DC520–30F5–10Ireland1Vaginal

Three recurrent themes relating to decision aid implementation were identified through the analysis. Verbatim quotes are used to illustrate the themes and the quotes presented reflect the full range of views expressed in the focus groups. To maintain anonymity, the quotes are identified by the participant’s professional role and unique identifier.

Detailed comparison and appraisal of the content and presentation of the two decision aids is outside the scope of this study. Further details of the aids can be found elsewhere.8,23 The two aids were often referred to synonymously by participants in this study. The findings we present here focus on the professionals’ views on implementation of these decision aids into the routine care of women choosing mode of delivery after a prior caesarean section.

In general terms, participants spoke positively about the decision aids. Many preferred the simpler information website to the more complex decision analysis program.

I think they’re a very good resource (Obstetrician, DB4)

I think the idea of having a help, a decision aid is really good because you often get women asking about VBACs [vaginal birth after caesarean] and you don’t have time to go into it in great depths with them…you don’t have the figures at hand…like you would on those decision aids. (Community midwife, MC3)

Practicalities of decision aid implementation: when, who and how?

Most participants felt that information about the existence of the decision aids should be offered as early in pregnancy as possible, ideally close to the booking visit (when women are around 12 weeks pregnant). However, some thought the aids should be accessible throughout, reflecting the view that decision making occurs over time and can change during the pregnancy.

I think the booking visit, as we said before, because that’s when the midwife takes the whole history, sees how many sections the woman’s had before, and then identifies the patient. (Obstetrician, DA5)

There was consensus within and across the focus groups that community midwives were the most appropriate healthcare professionals to first introduce the aids to women, particularly as they would have more personal knowledge of the woman and her circumstances. However, many felt that any questions generated should be discussed with an obstetrician nearer to term.

I think it would be best if it were the community midwife because it’s the community midwife that has got to know them and knows their individual circumstances, and might know the circumstances of their previous delivery, so I think that’s the most relevant person to do that. Am I just talking myself into more work? (Community midwife, MC4)

I think their community midwives can very easily tell them that there’s a, you know, there’s an aid around that can help them make those decisions which will then be discussed then when they come to see the doctor. (Hospital midwife, MA4)

Professionals other than community midwives gave reasons, such as lack of knowledge or not being available at the right time during pregnancy, as reasons why they were not the best professional group to first introduce the aids to women.

We don’t know that much about all these complications. (GP, DC2)

I think it depends when, when you’d offer it, ‘cause we don’t often see women until later on. (Obstetrician, DA1)

The majority of professionals felt that all women with a previous caesarean section should be offered a decision aid. However, others felt that there should be some targeting, notably towards ‘those interested’ or ‘those undecided’ and some stated that women with more than one previous section should be excluded. There seemed to be a hospital/community divide in terms of views on which groups of women may benefit most from the aids. Several hospital professionals suggested that ethnic minorities and deprived groups should be targeted, whereas some community staff identified educated and higher socio-economic groups as those most likely to benefit.

It may be very important, more useful towards…populations where you just can’t get across the information, ‘cause of the language barrier. (Obstetrician, DA4)

DC3: I think by definition you’re going to use it on an educated population, because...they are the people who are most likely to benefit from it.

DC5: Yeah and to want it and to be worried about it.

DC2: It’s your white middle-class interested people.

DC4: Yeah, it’s like the worried well isn’t it? (GPs)

Participants also discussed how the aids could be implemented. The venues highlighted as the best placed for implementation included women’s homes (via their personal computer), antenatal clinics and public facilities, such as libraries or evening classes. While feeling that community midwives were best placed to introduce the aids to women, participants tended to agree that self-access would be the best option for actually viewing and reading the program contents. This would enable more women to receive the information and pose fewer barriers to access than if use of the aids needed to be guided by clinicians (see ‘barriers’ below).

MC1: Yeah, it would have to be self-accessed.

MC4: Yes, more and more people do have their own personal computers and access to websites, and even, even the poorer end of the scale … they’ve usually got access to, to the web.

MC2: I think self-access would be fine. (Community midwives)

Is it something that would be feasible to just say ‘sit down, have a look at this, you know, before we see you.[In clinic] It’s gonna be an hour before we see you, have a flick through and see what you think’…they don’t need to see a nurse, they don’t need to see a midwife. There just needs to be a computer available. (Obstetrician, DA3)

Barriers to implementation of the decision aids

Several potential barriers to implementation of the decision aids were identified by the participants, and these fell into three categories: service issues, computer issues and people issues.

Service issues that were viewed as potentially preventing implementation of the decision aids included: professionals’ time and workload, workforce shortages, money and space. All groups raised the issue of the aids impacting on clinicians’ time and workload, in a context, where there is a perceived workforce shortage. This was more of a concern if clinicians were required to guide women through the aids or if access to the aids was in a community clinic or hospital setting. However, individual midwives and doctors felt that using the aids could ultimately be time-saving and enhance women’s care, as the aids could provide women with information that professionals did not have the time to provide.

It’s a bit of a time factor as well. I don’t think midwives, the main community midwives, have the luxury of actually women going and spending an hour doing that and then having the chance to actually spend a lot of time talking to them about it. I think that could be quite a pressure to be honest. (Hospital midwife, MA1)

It gives the chance to deliver information that we don’t necessarily have time for, we, you know, omit to mention at the time, I think. (Obstetrician, DA3)

The four hospital focus groups also raised the cost of setting up, maintaining and updating the decision aids and space limitations within their departments for women to view the aids on a computer, as possible service issues that would acts as barriers to implementation.

We haven’t got enough seats and people are standing and we’re running out of rooms now, aren’t we, as we’re getting bigger and doing more. (Hospital midwife, MB3)

Computer-related issues raised by the groups often focused on the computer literacy of potential decision aid users, especially if the aids were to be implemented for self-use. Together with language barriers (e.g. for ethnic minority groups) and potential lack of access to a computer, this was seen as a barrier to implementation that might result in inequitable access to the decision aids. There was consensus about these issues, both within and across the groups.

DB5: I think the internet way would suit some women but some groups who maybe don’t have a computer or aren’t as confident then they’d be disadvantaged.

DB3: And you’d have to be computer literate too. Many women may not be. (Obstetricians)

What about people that can’t read? I mean, you’d have to have it in different languages as well and stuff like that. We’ve got lots and lots of different ethnicities haven’t we? (Hospital midwife, MB2)

The final category of barriers to implementation was people issues. Professionals within all focus groups felt that women’s prior preference for a delivery method may act as a barrier to decision aid use, as those with a strong preference may show less interest in learning more about different delivery options and their associated risks and benefits.

There’s a lot of women who are quite pro vaginal birth, aren’t there. But I think there’s some women that, if they’ve had a really traumatic first delivery that’s ended up in a caesarean, then they just don’t want to go through that again. (Hospital midwife, MA5)

Individual clinicians’ delivery preference was also raised as a barrier by eight individuals in three of the focus groups, notably by midwives. It was felt that clinicians who had a preferred recommended mode of delivery for women with a previous section may not advocate the use of a decision aid that does not promote one mode of delivery over the other.

It really depends on how you can get the consultants behind something like this, because some of them don’t seem to care about the section rate...It’s just that they, they’ve just all got different opinions and you’re never going to get a programme that suits them all. (Hospital midwife, MB1)

But don’t you find that a lot of consultants do push them towards having an elective section or that it depends on the consultant? (Community midwife, MC3)

An additional people factor raised by some was the potential for decision aid use to result in longer more complex consultations with more challenging patient-professional dynamics, because of the additional information that women might want to discuss. However, there were a variety of views as to whether this would act as a barrier to implementation: some felt that it might, while others viewed the aids as potentially leading to improved consultation dynamics and more informed women asking fewer questions.

A lot of this isn’t something that you could discuss in, in a couple of minutes. I think, you know, if someone has spent a lot of time going over this website, then it’s not really going to be a ten minute consultation, I think it would be longer. (GP, DC1)

I think it would lessen them [consultations] a little bit. Like, if they can go away and read information, they might have less questions when they come in. (Hospital midwife, MA3)

There were some concerns that the aids would only help women who were already informed and interested and wanted to be involved in their care, rather than those who were less interested in information and more reliant on professionals to decide for them.

It would probably make no difference to the people that wouldn’t necessarily want to access the information in the first place and would rely on the doctor to give them the information and want them to make the decision for them. (Obstetrician, DB5)

Facilitators to implementation of the decision aids

A range of factors that might support or enable implementation of the decision aids were also raised by the group participants. The decision aids were viewed by some participants as potentially time-saving (see above). In addition, many participants described the decision aids as facilitating women’s access to information from reliable and standardised sources. This was particularly raised by some of the midwives and was viewed as beneficial as they felt that women often received conflicting advice from doctors within their usual care or sought information from unreliable sources. The decision aids would give the professionals confidence that the women were receiving trustworthy information.

MB1: I think it might sort of standardise the advice they’re given, that’s I think the main advantage. Because at the moment they get very conflicting advice depending on which doctors they see, so I think it would be a way of standardising them.

MB3: At least with a site like that [the decision aid] you know what’s on it, whereas some of the things they look at …Yeah, some of the questions, I say ‘Where on earth did you get that from?’(Hospital midwives)

Another valuable facilitator to implementation, raised within five of the six focus groups, was the potential for the decision aids to empower women, by helping them to think more carefully about the decision regarding mode of next delivery, to process and organise their thoughts, to stimulate questions and enable them to reach a decision that is best for them. In turn, this might result in greater satisfaction among women regarding the decision that is made. Participants from four of the groups also thought the aids would be a good educational aid and resource for professionals, similarly stimulating them to be more systematic in their use of information regarding birth options after caesarean section.

DA2: It’s systematically going through everything, isn’t it? Whereas we may not do that, we may be more biased

DA1: Partly it’s useful, because it just makes women think a lot about the decision, isn’t it? Cos I think sometimes women don’t think enough about the decision, so some of its usefulness is actually just in time spent thinking. (Obstetricians)

I agree with MA1 that the more information you can give a woman the more actually they’ll feel happier I think with the decision they make and they won’t feel they’re being forced into a decision kind of either way. (Hospital midwife, MA5)

Finally, there was consensus across most groups that uptake of the decision aids (or at least the opportunity to signpost them to women) would be most likely if they were incorporated within existing antenatal care, rather than requiring additional appointments or funds to implement provision. Increasing healthcare professionals’ awareness of the decision aids and obtaining their support for the use of the aids as an adjunct to clinical care were ways that were discussed to improve implementation success. In terms of actually using the aids, as noted previously, self-accessing the aids (for example via a website on one’s home computer), was viewed by professionals as a way of reducing many ‘service’ barriers to implementation.

I think if you make it an extra appointment ...you’re almost suggesting that the time they have now isn’t enough. (Hospital midwife, MB2)

If a lot of healthcare professionals know that those options are out there then hopefully somebody somewhere will be able to point it out to the right people (Hospital midwife, MA3)

It’s not just education of patients it’s education of healthcare providers as well. If you get them on your side and be motivated enough to, you know, roll it out to the community isn’t it? (Obstetrician, DB3)

Discussion

This study gives insight into healthcare professionals’ views on the role of decision aids for women choosing a mode of delivery after a prior caesarean section. It highlights practical issues, such as timing of the intervention and how the aids could be accessed. Potential obstacles to implementation include service, computer and people factors. Facilitators that may overcome some of these include increasing healthcare professionals’ awareness of the aids and women’s self-access of the intervention. The findings suggest that decision aids could be a useful adjunct to current antenatal care, with appropriate support from healthcare professionals.

Strengths and limitations

This study is one of the first of its kind to use qualitative methods at the end of a randomised controlled trial to evaluate how interventions (in this case decision aids) might be implemented into usual practice. It expands our knowledge of healthcare professionals’ views about decision aid implementation, which is important as professionals are key players in the decision-making process surrounding repeat caesarean section.5,6,24 While decision aids have been shown to be effective in this area,8,25 there has been little previous research on whether the aids have professionals’ backing and support.

We used focus group methods to evaluate professionals’ views on the decision aids. Homogenous (in this case, uni-professional), groups were chosen to minimise the potential for hierarchies to be created within groups and allow uninhibited discussion. However, many of the professionals within the groups knew each other within a clinical context and so may have been more likely to express socially acceptable views.26 This may result in pressure towards consensus and unanimity.27 While we found considerable consensus between and across groups, when disagreement did occur, participants seemed at ease to voice their differing opinions.

Our study was based in a single region in England, so the findings may not be transferable to clinical settings in other parts of the country. Those professionals who were invited but did not take part may have held different views from participants, but as the predominant reason for not participating was availability rather than lack of interest in the study, this is unlikely. Health professionals with a variety of professional and personal experience were included. Many of the issues raised were common across groups. It was felt that the major issues concerning aid implementation were raised during the groups, although it is possible that further data collection may have revealed additional information.

This research has not considered maternal evaluation of the decision aids and how they might be implemented. Women may differ from health professionals’ in their opinions about optimal timing of an intervention and how it is delivered.28 Qualitative research on women’s perceptions of the decision aids has been undertaken by members of the DiAMOND study research team (submitted for publication) and it may be useful to compare this data with the professionals’ perspectives.

Relationship with other literature

This study augments findings from literature on decision aid implementation in other countries.9,11,15 A USA study examined decision aids for breast cancer treatments9 and found that health professionals interest in distributing the aids and how aids were ‘lent out’ were important factors. Barriers to use included a lack of physician support and nurses’ perceptions about patients’ attitude towards participation in decision making. In a review article, Canadian authors concluded that the success of implementing decision aids into practice hinged on many factors including: the attributes of the decision aid itself, the practitioners and patients who used them and the practice environment in which they were used.11 Canadian healthcare professionals’ views were sought in a cross-sectional survey14 and qualitative interview study15 on three decision aids (looking at choices in HRT, COPD management and naso-gastric feeding). Despite an intention to use the aids, most professionals had not used them 3 months after the survey for logistical reasons.15 Within a UK maternity setting, one previous study has looked at the use of evidence based leaflets in pregnancy from a user’s perspective during a randomised controlled trial.10 Although healthcare professionals were positive about the leaflets, competing demands such as time pressure, perceived lack of choices available locally and lack of discussion of the leaflet content prevented their effective use.

Implications for practice, policy and future research

Our study shows the importance of investigating the views on different health professional groups. It provides insight into how decision aids may be more successfully implemented with regard to timing of the intervention, who delivers it, which populations could be targeted and how it may be implemented without adding extra strain to current resources and practice. Other studies5,6,24 support the idea that decision making about mode of delivery after caesarean section can be complex, can change throughout pregnancy and women’s requirement for information and support in this process will vary. Thus, being able to implement a decision aid in a flexible manner with regards to timing and amount of accessible tailored information, will be likely to benefit both women and their healthcare professionals.

In this study, self-accessing the aids was thought to be a pragmatic way of reducing costs, time pressure and burden on the health service. Holmes-Rolvner et al. found that physician referral was not a reliable mechanism for patient access to decision aids and suggested direct patient access from the internet as an alternative strategy.29 Access to the DiAMOND study aids via the Internet could be facilitated by linking them to existing online resources, such as NHS Choices (http://www.nhs.uk) or NHS Direct Online (http://www.nhsdirect.nhs.uk).

However, discussion of delivery options and support from health professionals are important in the decision making process5,6,24 and these should not be replaced by use of a decision aid. The aids should merely complement the existing care package. By increasing health professionals’ awareness of the aids and support of their use, women may be able to have a more meaningful discussion about the risks and benefits of vaginal versus caesarean delivery after a previous caesarean section.

A next step would be to pilot different strategies of integrating the decision aids into routine care to see whether they are effective and acceptable outside of the context of a randomised controlled trial.

Conclusion

This study adds support to trial evidence that the use of decision aids for mode of delivery after a previous caesarean section may be beneficial. It gives insight into health professionals’ views on the implementation of the aids. In general, the aids were seen as a useful adjunct to antenatal care. Barriers to their implementation included service, computer and people issues, but many ways to minimise these were highlighted. Further studies are needed to see if computer-based decision aids are useful in clinical care.

Disclosure of interests

None to declare.

Contribution to authorship

KR undertook recruitment, data collection, led on analysis and produced the first draft of the paper. AM was the principal investigator and contributed to study design and drafting the paper. KB contributed to the data collection, the analysis and drafting the paper. A.S was a co-applicant and contributed to the study design, analysis and drafting the paper. C.E contributed to the study design and drafting the paper.

Details of ethics approval

Ethics approval was granted by the South West Multicentre Research Ethics Committee on 2 November 2006. Reference: MREC/03/6/46.

Funding

The study was part funded by the BUPA Foundation as part of the DiAMOND study budget (reference 657/G10) and part funded by the School of General Practice, Severn deanery, South West Strategic Health Authority.

Acknowledgements

We thank the study participants in our project for their time, Roshni Patel for her help in recruitment and reading drafts of the paper and Beverley Lovering, Rachel Fielding and Anne Philip for their assistance in recruitment. We also acknowledge the other members of the DiAMOND study team, namely Tom Fahey, Claire Jones, Ian Ricketts, Tim Peters, Deirdre Murphy, Julia Frost, Peter Gregor, Sandra Hollinghurst, Anne Schlegelmilch, Kate Warren, Maureen Macleod, Irene Munro, Helen Watson and Kav Vedhara.

Ancillary