Understanding delayed access to antenatal care: a qualitative study Rosalind Haddrill

4.3.6 Ethical issues

Ethical principles underpin all medical and social research, and the ethical aspects of research governance for the study have previously been discussed in section 4.3.2. Such principles have their roots in different models, with the focus on risks, benefits and outcomes in the medical model contrasting with the focus on respect, autonomy and process in social models. At the core of ethical research lies a commitment to integrity, quality and transparency (ESRC 2012: 2), and the principle that “The dignity, rights, safety and wellbeing of participants must be the primary consideration in any research study” (DoH 2005: 7). Fundamentally it has been argued that ethical practice is simply ‘doing the right thing’: conscience should be the researcher’s primary guide, with a responsibility to participants, researchers and potential users of the research (Green and Thorogood 2014). Consideration of ethical issues in qualitative studies involves a balance of responsibilities to these different ‘stakeholders’ and inevitable tensions, balancing conflicts of interest, the need to respect participant boundaries and privacy, benefits and risks, with a desire to tell their story. Homan (1991) argues that commitment to ethical practice is built on a recognition and understanding of such dilemmas. Ethical practice in qualitative research is achieved through honesty and negotiation with the participants, taking a holistic view of participants, their social and cultural context. Such ethical considerations are at the heart of reflexive practice (Goodwin 2006, Holloway and Wheeler 2010).
The abstract ethical principles of Beauchamp and Childress’ Principles of Biomedical Ethics (2008) are reflected in many research ethics frameworks (for example ESRC 2012), and the ethical considerations for this study. I recognised that the application of these in practice was likely be complicated. Some ethical issues were predicted, such as the need to protect confidentiality and obtain informed consent, however these (and others) developed during research, for example with the interviewing of teenagers and women with learning difficulties. In order to interview successfully and establish a rapport with the participants I needed both empathy and distance. I wanted to put myself in their place, but needed to be honest about my background as a midwife and could not completely detach myself from my professional role. This dual role was difficult to reconcile: I was not auditing good practice, or providing a caring role, but needed to acknowledge that safeguarding young women in particular was more important than confidentiality (Holloway and Wheeler 2010).

4.3.6.1 Informed consent and respect for autonomy

The principle of informed consent underpins all professional guidelines for conducting research (Green and Thorogood 2014). As Department of Health guidance (2005) states: “Informed consent is at the heart of ethical research” (p7). Informed, voluntary consent means that research participants are recruited appropriately, are fully informed about the research and the implications of participation, and give their voluntary agreement to participate in it. Participants are subject to neither implicit nor explicit pressure to do so (ESRC 2012, Holloway and Wheeler 2010). The principle of informed consent is linked to respect for autonomy and an individual’s right to self-determination, and is part of an ongoing process of honesty and collaboration throughout the research, so that participants understand they are able to withdraw at any time (Beauchamp and Childress 2008, Bowling 2009, Green and Thorogood 2014).

An opt-in process was used, whereby potential participants gave their initial consent to be contacted further to participate in the study. For the women, a participant information sheet (appendix 4.7) was given to potential participants, providing clear details about the research, the proposed interview, confidentiality issues, the opportunity to withdraw at any time and who to contact for further information or if they had any concerns. Accompanying this was a short consent form, giving me permission to contact the women by telephone to arrange a meeting, at their convenience. The information sheet, letters of invitation and consent forms formed part of the formal ethical approval process for the study.

Some authors have argued that making participants sign a form can deter and undermine the relationship between interviewer and participant, and this may be why some women chose not to participate. However, not receiving written consent is unrealistic in health research (Bryman 2008, Green and Thorogood 2014). For participants with a single involvement such as this study, Goodwin (2006) suggests that this sort of information sheet, outlining the research process, overall objectives and providing a clear statement about voluntary participation, is sufficient to obtain informed consent. She argues that the process of obtaining consent is complex however: “obtaining informed consent in qualitative research cannot be accomplished by the mechanistic production of a consent form signed at the outset of the research” (Goodwin 2006: 59). This was reflected in repeated opportunities confirming women’s willingness to participate prior to the interview.
All women were given background information verbally about the study, in addition to the sheet which they could take away. The information sheet aimed to provide significant information about the study, using clear, unambiguous language, without influencing participants behaviour by explaining too much about what I was looking for. Homan (1991) argues that there is a danger of contaminating people’s answers if too much information is given at the outset. All women were given the opportunity to read the information sheet and ask questions, prior to signing the consent form for initial contact by myself (appendix 4.9). All women were then given a minimum of 48 hours after signing the form before attempts were made to contact them, to give them time to consider. The subsequent telephone conversations gave a further informal opportunity for clarification and confirmation of consent to participate further. When meeting face-to-face, women were given a further opportunity to ask questions about the study and another consent form to sign prior to interview (appendix 4.10), including consent to be recorded, which was done overtly. Women were informed that they could refuse to answer questions and could also ask for their data to be removed from the study after the interview.
The process of obtaining consent adapted as the study evolved, reflecting the flexible and emergent nature of the research and the changing sampling methods used (Goodwin 2006). Particular care was taken when seeking consent from more vulnerable women, in particular the four teenagers aged less than 18 years and the three women with learning difficulties, to ensure that the study and participation was clearly explained and understood (DoH 2005). Green and Thorogood (2014) argue that researchers have a duty to represent such participants, whose voices are often not heard and who may be less able to be informed or to give true consent. Such women are over-represented among late booking women. As with medical treatment, the young women aged below 16 were not presumed to have competence to consent to participate in the research, however consent was obtained when it was felt that they had sufficient understanding and intelligence to understand fully what was proposed (sometimes known as 'Gillick competence' or the 'Fraser guidelines') (NSPCC 2014). This is in line with General Medical Council (GMC) guidance for research:

“You must decide whether a young person is able to understand the nature, purpose and possible consequences… only if they are able to understand, retain, use and weigh this information and communicate their decision to others can they consent… the capacity to consent depends more on young people’s ability to understand and weigh up options than on age.” (GMC 2007: 11-12, GMC 2010)

Extra time was spent with the women discussing the study and their part in the research, before asking for consent; particularly their understanding that participation was voluntary and that they had a right to refuse or withdraw from the research at any point. In some cases this discussion was in consultation with supporting adults. Particular attention was also made to the interview format and location for these more vulnerable women to try to put them at ease, for example giving them the option to be interviewed with a friend or family member present.
National guidance concerning research with children and young people recommends obtaining consent from a parent or carer with parental responsibility in addition to obtaining consent from the child or young person (Shaw et al 2011). Consequently an information sheet for parents was produced (appendix 4.8). However, three of the four women aged less than 18 who gave consent did not wish to involve their parents, which was respected, as it was that felt to do otherwise was to undermine their autonomy. The fourth woman, aged 17, who also had learning difficulties, requested to be interviewed with her mother present, though the mother did not participate in the discussion. The young woman signed the consent form and her mother received a copy of the parental form. Also in line with guidance, parental consent was obtained to conduct an interview with a 16 year old young women in her family home (Shaw et al 2011).

4.3.6.2 Right to privacy and confidentiality

Confidentiality and anonymity, particularly in relation to the appropriate use and protection of participant data, form a fundamental part of all stages of any research project, as recognised by the ESRC Framework for Research Ethics (2012) and DoH Research Governance Framework (2005). The aim was to uphold confidentiality and anonymity as far as possible, making clear the limits of confidentiality at the outset of the study (Green and Thorogood 2014). This was reflected in the wording of the participant information sheet which noted that “All reasonable steps will be taken to ensure confidentiality… we would like to emphasise that it will not be possible to identify you from any report of this study” (appendix 4.7). Access to the data was limited to myself, the principal Investigator for the study and the supervisory panel, for independent verification and interpretive challenge.
The process of confidentiality started with me disclosing my background as a midwife but reaffirming my role as a researcher, independent of their antenatal care. I recognised the importance of honesty with participants but also of not researching women in my care, because of the potential perceived power imbalance and the possible disclosure of sensitive information which might affect our relationship (Butler 2003). Achieving privacy for the interviews was another essential part of this confidentiality, in terms of location and the presence of others, but was balanced with the participant’s choice. Two women chose to be interviewed with other people present: as previously mentioned, one young woman with learning difficulties was interviewed with her mother, although she did not participate in the conversation; another woman whose English was limited was interviewed with her husband, who did. This was a pragmatic way of gaining the women’s perspectives, but one which subsequently included the views of the family member.
The anonymity of participants throughout the study, during data collection, analysis and writing up, was maintained by removing all identifiable content and allocating participant numbers to the women. Other steps were taken to ensure the confidentiality of information, such as not using local transcription services. Tapes, transcriptions and notes were stored without identifying details and separately to demographic details; all were stored securely in accordance with the research governance guidance for the study. Collection of data only occurred during planned interviews, and there was no informal collection, quoting or referencing of confidential documents such as medical notes or personal details. Information about participants was discussed anonymously during supervision and analysis but was not disclosed in other settings. Participants were made aware that others had limited access to the interview tapes for peer review, as part of the analysis process, and consent was obtained to quote them directly but anonymously.

4.3.6.3 Non-maleficence and beneficence

Bryman (2008) argues that doing no harm to participants is the litmus test of ethical research, but also that “virtually all social research is intrusive and exploitative to some degree” (p129). There is an argument that qualitative research has the potential to be more intrusive than quantitative research, and as such requires particular sensitivity and good communication skills to minimise the risks, the emotional consequences, to any participants (Holloway and Wheeler 2010). Interviews were conducted in as quiet and supportive environment as possible, with the participant’s choice of location and timing, recognising the sensitivity of the topic for the women particularly and the possibility of distress or anger. My approach towards the interviews intended to build a supportive relationship and to convey respect for participants’ (emic) perspectives and concerns, rather than my own, allowing their stories to be told honestly but also in a positive way.

4.3.6.4 Fairness and justice

A fair and just approach to the research was reflected in the research governance of the study and all the study methods discussed above, and particularly the honest, collaborative approach adopted towards recruitment and interviewing. The exclusion of non-English speaking women from the study reflected this commitment, as women could not be represented fairly and truthfully without the use of interpreters. It was also reflected in the consideration of my own safety when interviewing in people’s homes. This responsibility to the participants has continued during the writing up of the research, in terms of attempting to represent them fairly. Fairness was also reflected in the presentation of the findings, avoiding imposing interpretations that were not shared by the participants themselves, which would have reflected an unequal and hierarchical relationship (Bryman 2008). Green and Thorogood (2014) identify the risks of any analysis and interpretation; they argue that a balance must be struck between moving thinking forward and representing participants faithfully. For me, the exact reproduction of the words of the women particularly, through verbatim transcription and quoting, was part of this fair representation.