< Previous | Contents | Next >

A Framework for maternity services in Scotland

Appendix 2

Public and pofessional consultation

Research conducted by Scottish Health Feedback for the Scottish Association of Health Councils and the Scottish Executive Health Department.

Women's opinions about maternity services

There have, over the years, been many studies which have considered women's views of maternity services. Much of this work has been based on large-scale surveys, at the local and national level, and has used structured interviews and/or self-completion questionnaires to gather information from women about their views and experiences (see, for example, research carried out over 20 years ago by Cartwright, 1979).

Maternity services in Scotland have been studied in a variety of ways, including several local surveys in specific Health Board areas. In Lothian, for example, over a 12 year period (1987-1999) the same research team (now located within Scottish Health Feedback) has conducted 3 large surveys of recent maternity service users. The studies have charted changes in the configuration of services in Lothian, but suggest that the factors which affect women's satisfaction with maternity services are remarkably stable and consistent.

The main areas of dissatisfaction with services relate to: the provision of information and, more widely, communication between women and health professionals; issues of choice; and the extent to which there is felt to be continuity of care. Satisfaction is patterned by social status, with socially disadvantaged women - those living in areas of relative deprivation, those with lower levels of education, and those who are unsupported - expressing the greatest dissatisfaction with the care they have received throughout their maternity. It should be emphasised that although there are clear areas of dissatisfaction, most women in the surveys conducted within Lothian were reasonably happy with most aspects of their maternity care.

Survey methods, however, have their limitations. In particular, while there is good evidence that women can give reliable and valid accounts of their experiences (Martin, 1989; Martin, 1990) self-completion surveys cannot be used to explore the meaning or significance that respondents might attach to particular issues; issues which are simply not amenable to a "tick box" approach. As the success or failure of health interventions will largely depend on the extent to which they are informed by lay perspectives (Milburn, 1996) it is important to understand why, and in what ways, certain aspects of care may be important to women. For such investigations, there is a need for in-depth, qualitative approaches.

The role of qualitative approaches

In 1995 the BMJ published a series of articles on qualitative research. The aim of the series was to demonstrate to a rather sceptical biomedical audience why and how qualitative approaches complement the quantitative methods with which BMJ readers were more familiar (Pope and Mays, 1995). Qualitative methods do not seek to provide quantitative answers to research questions. Indeed, the formulation of the research question is critical in determining what methods should be used in different circumstances. In very general terms, quantitative methods would be most appropriate for answering more "factual" type questions (what proportion of women see the same midwife for antenatal care? Do women want to see the same midwife for their antenatal care?), while qualitative methods are more appropriate for answering questions about opinions or perceptions - for example: in what ways does continuity of care influence women's experiences of antenatal services?

Qualitative research uses a range of methods, including in-depth interviews, case studies, consensus methods, observation methods and focus groups. The consultation for the Maternity Services Framework largely used this latter approach and it is perhaps helpful to describe this form of data gathering. Focus groups are a form of group interview that "capitalises on communication between research participants to generate data" (Kitzinger, 1995). Each participant is not asked to respond to a question in turn, but instead people are encouraged to talk to one another, to exchange experiences and opinions. The method is particularly useful, as Kitzinger points out, for exploring people's knowledge and experiences, and for examining not simply what they think but how and why they hold particular views. Focus groups are widely used to explore people's experiences of health services and are particularly effective for understanding attitudes and needs.

Qualitative research is often criticised by those more used to the conventional methods of biomedical research as being "unscientific" and "merely anecdotal". Such criticisms ignore the reality underpinning almost all research: namely, that research is always selective and the researcher cannot capture the literal truth whatever method is used. The strategy to ensure rigour in qualitative research - as for quantitative research - is "systematic and self-conscious research design, data collection, and communication" (Mays and Pope, 1995).

The relationship between quantitative and qualitative approaches does not have to be antithetical, but should instead be viewed as complementary:

The rigid demarcation of qualitative and quantitative research as opposing traditions... does not encourage movement or interaction between the two camps. In effect, researchers on either side become entrenched and are often ignorant of each others' work (Pope and Mays, 1995: p43)

Although there has been qualitative research in relation to maternity services, this has been relatively small scale and localised or concerned with specific aspects of maternity care or women's experiences of pregnancy (see, for example, Oakley, 1979). There is, therefore, a need for qualitative research which has a current and Scotland-wide perspective, which explicitly focuses on what women want from maternity services.

The consultation with service users

The consultation with current and recent service users was carried out in 5 Health Board areas, and included 9 NHS Trusts. In each area, there was a systematic approach to recruitment to the study in which brief questionnaires were sent to all the women who met particular sampling criteria (see Tables 1 and 2 for details). The questionnaires and invitation to attend a focus group were mailed on behalf of Scottish Health Feedback by the NHS Trusts, but the completed forms were returned to Scottish Health Feedback. The questionnaire asked for information relating to each woman's expected or actual date of delivery, her age, and whether she had had an instrumental or operative delivery. We only asked women who were willing and able to attend a focus group on a specified date to return the questionnaire to us. In addition to recruitment via Health Boards, local community groups were asked to help with the recruitment of participants from ethnic minority groups. Full details of the response rates are provided in Table 3.

The focus group topic guides and the one-to-one interview schedules asked women to reflect on what aspects of their care mattered to them: what had been good about the maternity care they had received, what had not been good, and how the services could be improved. Because the subject matter for the focus group discussion was determined by the women participating in the group, the topic guides were constructed in such a way as to allow the facilitator to move easily between topics and issues as they arose within the discussion. The topic guides covered a range of issues including: access to services and service options, choice, risk assessment/risk management, continuity of care, facilities, resources, services, relationships and contacts with health professionals, information, education, record-keeping, advice and support.

A total of 15 focus groups with women were conducted. In addition, 20 in-depth interviews were carried out with women who may have been reluctant to attend a focus group (younger women and women from ethnic minority groups), or who might have had practical difficulties in attending a group discussion (those living in remote and rural areas). In all, almost 100 women were included in this consultation exercise.

The consultation with health professionals

Eight focus groups were convened with health professionals in the 5 Health Board areas. The group participants were recruited from each of the Trusts and separate groups were convened with midwives in hospital, midwives in the community, general practitioners, health visitors, neonatal nurses and neonatologists. In addition, 2 mixed groups were conducted: one with primary care professionals and one with obstetricians, radiologists and anaesthetists. Participation was voluntary and groups lasted for one hour. Fifty-four professionals participated in the consultation (Table 4). The focus groups with health professionals asked the participants to consider key words or phrases that described the current delivery of maternity services. Next, they were asked to repeat this exercise considering how services should be or might look if they reflected a (more) woman-centred perspective. Comparison of the 2 descriptions formed the centre of a discussion concerning how services could be delivered to meet the needs of women.

Data analysis

All of the focus group discussions and the interviews were fully transcribed and entered onto a software programme which supports the analysis of qualitative data. Narratives can be coded in terms of any number of categories and the software has powerful cross-referencing operations which facilitate detailed and systematic questioning of the data.

The analysis proceeded in 4 principal ways: first, each focus group and interview transcript was coded according to a number of basic characteristics, including whether they are lay or professional (and professional discipline), the area in which the group or interview was conducted, and other defining characteristics such as age category (less than 20 years, 20+ years), and the type of delivery (normal or assisted/Caesarean Section). In this way, it was possible to cross-reference any piece of narrative with these basic attributes.

Second, we identified key concepts which were felt to be at the core of the research - for example: continuity of care, access to services, relationships with health professionals, relationships between health professionals, choice, information, risk and safety, individual care and so on - and these were posited as primary coding categories.

Third, new coding categories were developed as the analysis proceeded. Any piece of narrative can be coded in multiple ways, and could be placed in any number of categories. Issues, themes, ideas were identified and explored from within the data.

Finally, if an idea or concept appeared to be important across respondents this was explored systematically using cross-referencing facilities within the software.

In these ways, data were systematically analysed to identify the key issues that women and health professionals feel would lead to user-centred maternity services.

The composition of lay and professional samples

Table 1 Composition and location of the lay focus groups

Table 2 Composition and location of the interview sample

Table 3 Response rate and final number of participants in each group

*This group included 3 women who had a home birth, and one person who had a normal hospital birth and had been assigned to Group F1 but was moved to Group F3 to increase the group. The response rate is based on responses from the home birth invitees.
Total lay focus group participants n = 79 (mean group size = 5: range 2-8)
Total interview participants n = 20
Mean response rate - lay focus groups 19% (min 10%, max 30%)
Mean response rate - lay interviews 22% (min 20%, max 25%)
Mean overall lay response rate 19% (min 10%, max 30%)

Table 4 Composition and location of the health professionals' focus groups

Total number of participants: n = 54; mean group size: n = 7 (min = 3, max = 10)

Research team

< Previous | Contents | Next >