The challenges of securing ethical approval to undertake qualitative research in the English National Health Service have been previously detailed. Acknowledging these general challenges, this case study describes some further issues specific to gaining approval for ethnographic research in National Health Service settings. Ethnographers attempt to understand the world from the vantage point of those that inhabit it. As it entails a concern for the perspectives and experiences of others, ethnography is an invaluable approach to understanding the social and cultural dimensions of health and illness. Ethnographers spend time with people in their "natural habitat" or social setting, seeking to engage with people on their own terms. Ethnography can offer critical insight into the practices of healthcare, as well as the contexts within which these practices take place. Gaining access to National Health Service settings for the purpose of research that includes patients requires prior approval from a National Research Ethics Committee. The bureaucratic structure within which these committees operate is designed to weigh up the potential harms and benefits of biomedical research. It does not, therefore, fit neatly with ethnographic research. This case study describes three important questions which arose in the course of seeking approval for my doctoral research: first, how to provide an account of a research proposal that involves an inherently unpredictable process in a format designed for carefully controlled biomedical research; second, how personal perspectives affect research; and third, the difficulty of working across the complicated organizational structures found in and around the National Health Service. I provide some reflections on these issues, and my experiences, with the aim of helping other students as they prepare to undertake ethnographic research in the National Health Service.
By the end of this case, students should be able to
- Understand the tensions between undertaking ethnographic research and the biomedical framing of National Health Service (NHS) research ethics
- Reflect on their position as researchers and how this might affect their ethnographic research and ethics application
- Map out the organizational complexities that will affect their research.
This case study describes how I managed the process of securing ethics committee approval, and the preparations I undertook to start conducting fieldwork for my doctoral research. My project involved an ethnography of health and social care for people with multiple health and social care needs. I focus on three areas that will be of interest to other researchers who are seeking to undertake ethnography in health settings particularly, but not exclusively, in the National Health Service (NHS). These are managing issues around uncertainty, being reflexive about one’s role as a researcher, and navigating organizational complexity. This case study also discusses the tensions generated by the need to secure ethical approval for a research process which is inherently unpredictable and fluid, while operating within a framework that seeks, what might be called, anticipated certainty. I offer some pragmatic strategies for managing this tension, while avoiding dissembling. Subsequently, I reflect upon a particular ethical issue raised by the committee reviewing my application. They sought to consider the effect that my previous health service position might have on recruitment of research participants who were employed in the site. Before concluding with some remarks that might assist other students preparing to undertake ethnographic research in health settings, including in the NHS, I provide an overview of the complex organizational structures that researchers must navigate if they wish to undertake ethnography with research participants who use multiple health and social care services.
Project Overview and Context
My doctoral research examines the patient experience and the practice of integrated health and social care within its organizational, social, and policy context. Conceived of as a way to provide and manage health and social care resources more effectively while meeting increasingly complex needs, particularly those of the aging population, integrated care is a central tenet of U.K. health policy. It seeks to address fragmentation and improve quality by operating in a manner that is person-centered and coordinated. In my research, I adopted an interdisciplinary approach. Drawing on a diverse range of scholarship—including social policy, sociology, anthropology, and social theory—I sought to examine the question of how integrated care is shaped and practiced. Ethnography was employed as a way of understanding efforts to integrate care from the vantage point of those receiving, providing, and organizing care. A methodology employed by anthropologists, sociologists, and other social scientists, an ethnographic approach provides a route to developing an understanding of complex social settings, by, for example, participating in the everyday routines that they feature. Ethnographers aim to spend a significant amount of time with those who are involved in the particular field being studied. Exposure to their social setting facilitates a detailed understanding of others’ perspectives and the ways in which their social and cultural life is experienced. Ethnography can provide new insights into the contexts which influence individuals’ experiences of health and social care (Savage, 2000).
My study of integrated care focused on an urban sub-regional planning unit in England, where I undertook participant-observation, described by Emerson, Fretz, and Shaw (2001) as “establishing a place … on a relatively long-term basis in order to investigate, experience and represent the social life and social processes that occur.” During an 18-month period of fieldwork spanning 2014-2016, I followed three lines of inquiry: planning and commissioning of health and social care services, the experiences of 20 research participants (or “informants”) experiencing multiple health and social care needs, and the practice of health and social care professionals providing integrated care. Participant-observation of planning and commissioning was enabled by my concurrent role as a senior manager in a clinical commissioning group (CCG) in the study site, whereas participant-observation with people with multiple health and social care needs and practitioners required the specific recruitment of research participants. On occasions, I employed non-participant observation (Gobo, 2008, p.15), deliberately stepping back to observe interactions between patients and practitioners in clinical settings (with express patient and clinician consent).
Gaining access to NHS settings for any research involving patients requires prior approval from a Health Research Authority (HRA) ethics committee. These committees operate within a framework that has been designed to weigh up the potential harms and benefits of biomedical research, and which includes additional safeguards for specific patient groups who are regarded as vulnerable. Research involving vulnerable patients requires approval from a specifically designated committee with relevant expertise. It was probable that my research into the experiences of people with multiple health and social care needs would include individuals who could be considered as vulnerable. This vulnerability could be a function of their general frailty and possible social isolation, but also because of the increased potential for cognitive impairments or dementia. Recruiting research participants who lack the capacity to consent for themselves, as could be the case with someone with dementia, is an area that requires careful consideration to safeguard people’s rights to refuse to consent, and is one that can also be fraught with practical difficulties.
The possibility that research, such as clinical trials of medicines or medical products, might harm research participants is initially judged through the NHS research ethics committee application form, found on the Integrated Research Application System (IRAS). This form requires details of the planned research interventions or procedures, asking how many interventions/procedures will be carried out, whether these are part of routine care, how long the intervention/procedure will take, who will carry it out, and where will it happen. The framing of research as something that comprises “interventions” or “procedures,” events that take place at designated times and specific sites, poses problems for researchers who plan to undertake ethnographic research, which, by its nature, is inherently unpredictable and uncertain. Ethnographers will typically designate a period of time for their fieldwork, spanning months or even years, during which they will seek to learn about a group of people or a way of life by spending time in the field. They will follow the opportunities to meet new people and visit new places as they arise and as they learn about those they are studying. Thus, it is impossible to quantify the number of interviews they might conduct, or interactions they might have, in advance. This approach to research is very different from that of biomedical research. For example, a project that is concerned with testing the efficacy of a new treatment will be able to specify in advance the number of times the treatment will be given, the dosage, and other parameters. The practical difficulties created by the need to complete a form requiring pre-planned, quantified, and specifically located interventions when intending to pursue a more fluid and uncertain research method gives rise to ethical dilemmas.
As an ethnographer with a dual role, both researcher and local commissioner, I initially considered myself to be well-placed to access aspects of my research site. My status as an “insider” (Dixon-Woods, 2003), while requiring careful positioning and significant reflection on my role, at least meant that I already knew my way around the local services, and had identified key informants; people who I knew from my commissioner role had considerable experience and knowledge of the way in which health and social care services had developed locally. I had assumed that this aspect of my ethnography would be relatively straightforward in terms of securing interviews and being able to conduct fieldwork, more so than working with patients, as-yet unknown research participants who were experiencing multiple health and social care problems. However, I had not foreseen the complexity of securing ethical and managerial approval from all of the different organizations involved, or the ethical queries raised by the committee in relation to my commissioner role and my previous experience of, and relationship to, the site of my research.
The practicalities of planning ethnography, with vulnerable people, and of undertaking ethnography across a number of organizational structures, while retaining an “insider” role, gave rise to the ethical dilemmas described below.
A researcher setting out to undertake participant-observation with the aspiration of creating an ethnography, or an account of people and their social world, will not know what they might encounter once they are in the field. Just as the lab scientist will not know what the results of her experiment might be before she concludes it, neither does the social scientist know what she might find in advance. Scientists will be testing hypotheses or searching for answers to research questions; the nature of empirical work means that we are deliberately setting out to discover new knowledge. If we knew what we were going to find, we wouldn’t need to undertake the research. The problem for the ethnographer is not that she is unable to predict what she will find, but that she will be unable to predict exactly how she will go about exploring her object of study. A participant-observer who wants to understand the world from the vantage point of another person, or from within a particular social setting, must follow that person, or inhabit that setting, and not direct them. For example, an ethnographer will not know in advance how someone’s daily routines might play out, and yet they will seek to participate in these to understand how they experience their lives. Furthermore, unstructured, unplanned periods of time are required to surface findings that are only apparent from within that social world. This is where ethnography differs from other qualitative methods of inquiry such as interviews or focus groups that require detailed advanced planning regarding, for example, the specific questions one should ask particular individuals. This intrinsic uncertainty about not just what will be found during the research process but how the process will unfold is particularly problematic for ethnographers when they come to complete the research ethics application form that requires the research interventions or procedures to be specified in advance. The tension between the anticipatory regulatory regime and the emergent nature of ethnography in healthcare is discussed in detail by Murphy and Dingwall (2007). They focus on the difficulties of obtaining written consent without knowing in advance exactly what the research will involve and, therefore, when there is lack of clarity as to what the research participant might be consenting. They caution against an overly bureaucratic process of prior approval as this can over-shadow the need for researchers to respond ethically to dilemmas as they arise.
In addition to the tensions involved in seeking consent for a research process that, by its nature, is not wholly specifiable in advance, a further problem arises for ethnographers. The form they are asked to complete requires them to specify the number of interactions they will have with their research participants. It is unclear how those engaging in ethnographic fieldwork are expected to represent a deliberately uncertain future within a framework that presumes research entails planned activities, about which it requires advance details. This creates an ethical dilemma for researchers who must provide such information but, nevertheless, are aware that whatever they put will likely change. Indeed, ethnographers are aware that they cannot accurately predict how long interviews will last, where they will take place, or even who they will speak to. Allbutt and Masters (2010) highlight the possibility that ethnographers manage this tension by dissembling. The format of the ethics application form invites a response that is potentially misleading in its certainty about what will happen in the future. This is not just the uncertainty of any planned research, or indeed any planned activities. Certainly, not all biomedical research goes according to plan, the number of planned procedures might increase, or interventions be carried out during the study by a different member of staff than originally envisaged. The ethics approval process allows for this by reviewing amendments to be made to the initial protocol. However, what is problematic for the ethnographer is not that things don’t go according to plan but that they are being asked to write down something that fundamentally conflicts with the logic of their methodology. Thus, to comply with the requirements of the ethical approval process, an ethnographer might be tempted to dissemble, to provide the kind of answers expected of them in the knowledge that they create a misleading impression, they do not reflect the reality of their method, and they implicitly assume a greater degree of certainty than is possible. If they did not do so, committee members may start to question whether the proposed research can be conducted within the framework of research ethics.
The second ethical concern that arose during the process of seeking ethical approval for my doctoral research was a question about my pre-existing professional connection to the field. Although I had previously reflected on my position and how this would affect my research, I had not anticipated the specific questions raised by the ethics committee during their review of my ethics application about recruitment of staff research participants. My research design included interviewing and observing staff, clinicians, administrators, and other practitioners as they delivered integrated care. As I was employed in the case study site as a manager in a CCG, I intended to conduct my fieldwork as an “embedded” researcher, continuing to work in that role, albeit part-time, while I carried out my fieldwork. As researcher, I would, therefore, have the status of an “insider,” meaning I would be studying practices as I participated in them. In accordance with the ethics committee framing of research as a series of interventions or procedures, I designated my intended “informants” in this process as research participants. Although these research participants were not patients but members of staff, my planned interactions with them (interviews, shadowing, observations) were intended to generate data. Therefore, I planned to treat them in a manner similar to any other research participant including in relation to seeking consent and maintaining confidentiality. The ethics committee identified a potential concern about the effect my position within the field might have on the recruitment of the research participants who were members of staff (similar concerns were not raised about recruitment of patients). One committee member was concerned that my role might mean staff in organizations that had been, or were seeking to be, commissioned by the group I worked for would feel unable to refuse a request to participate in my research.
Resolution of the Dilemma
As I prepared my application for ethical approval, I faced an ethical dilemma: how to present my plans for an inherently uncertain research project in a format that required certainty—without dissembling. During the review of my proposed research, I then encountered a second dilemma: how my position might affect a research participant who did not wish to consent to the research. My strategies for resolving these dilemmas involved attempting to understand the vantage point of the ethics committee.
Ethics committees are charged with weighing up the potential harm and benefit of the research planned. To do this, they need to know what the planned research entails and the kinds of ethical issues the researcher thinks might arise during the conduct of that research. The form I completed included the following question:
What are the potential risks and burdens for research participants and how will you minimize them?
For all studies describe any potential adverse effects, pain, discomfort, distress, intrusion, inconvenience or changes to lifestyle. Only describe risks or burdens that could occur as a result of participation in the research. Say what steps would be taken to minimize risks and burdens as far as possible. (IRAS application form)
The potential harm of observational studies, interviews, and ethnographic studies is of a different nature to that posed by trial of medicines and medical products, and is most likely to be social or psychological harm (Murphy and Dingwall, 2007). I was planning to spend time with people in their homes and talking to them about their health. I also intended to spend significant amounts of time with people over an extended duration and to observe them during health and social care appointments and procedures. I identified that this research had the potential to be intrusive, that my presence might be seen as burdensome, and that my presence during healthcare appointments could compromise their privacy.
As some general principles could be applied, managing issues around privacy, dignity, and confidentiality was relatively straightforward. I was able to translate the language of interventions and procedures into interviews and observations, and prepared topic guides for initial interviews that showed the lines of inquiry I intended to follow. The difficulty arose when trying to specify the quantity and duration of interviews and other periods of observation at the outset of the study and, therefore, the potential burden or harm this could cause. At that point, I didn’t know how the ethnographic work would unfold. For example, I was uncertain about how many times it would be feasible or necessary to visit the research participants, or for how long people would welcome me in their homes, and whether they would allow me to repeatedly accompany them to their medical appointments. Without that kind of detail, the ethics committee would be unable to judge whether the study was well planned or whether it would cause significant harm. Rather than dissembling by devising an arbitrary schedule that I knew would not be followed, I focused on the aspect of the research that I had identified as being most burdensome and intrusive: the frequency and duration of time I intended to spend with research participants. If this was the greatest risk that my research posed, then I decided to establish the maximum burden my research might impose on each research participant.
To establish this maximum, I considered the duration of time I had allocated for my fieldwork and the amount of time I would have available for each research participant given the total number of research participants I anticipated recruiting. This enabled me to estimate the maximum amount of time I would have available for each research participant, representing the “worst case” scenario in terms of the potential burden from their perspective. Instead of feeling compelled to overstate my certainty about how the study would unfold, I was able to set limits on the duration of the ethnographic work. I felt confident that this gave the ethics committee a clear picture of the greatest burden that might be imposed on a research participant, and therefore allow them to make an informed judgment about whether the proposed study was justified. I also described how I would mitigate against the risks of this potential burden, by ensuring that I continually checked that research participants were comfortable to continue. This was made very clear in my initial consent forms, and became an important part of my ongoing practice.
In this way, I was able to resolve my first ethical dilemma before I encountered the ethics committee. However, I had not fully anticipated the second ethical dilemma, which was raised by the committee as they reviewed my application. Their concern was that my role as commissioner could unduly affect recruitment of staff as research participants, in that they would feel obliged to consent because of the organizational dynamics between commissioning bodies and organizations that were seeking commissions. The methodology of my research led me to treat staff informants in a similar manner to patient research participants, this included issuing them with participant information sheets and written consent forms. I intended to bring an ethnographic approach to their practice, as well as to the patient experience of integrated care. I was aware that research involving NHS or social care staff as participants does not normally require research ethics committee approval1 and so had not anticipated that the committee would raise any specific questions about staff research participants, assuming that this was in fact outwith their remit. However, the committee reviewed the entire project, not just those aspects affecting patient participants. This meant that the general principles of consent, including the principle that consent must be “voluntarily given with no undue influence,”2 was applied to all research participants. The committee wanted assurance that my position as a member of staff in a commissioning body that had a contractual relationship with an organization employing prospective research participants would not unduly influence their decision to consent to being involved in the study. Social research generally requires a degree of reflexivity, and ethnography in particular demands an examination of positionality and power dynamics. The question that the ethics committee posed provoked further reflection on my position within the case study site and how this could affect relationships with research participants. To mitigate against any undue influence, I made my dual role explicit in the relevant documentation (participant information sheets and consent forms) and added a statement to the staff research participant information sheets to the effect that their decision to participate or not in the study would not affect any commissioning decisions. This was similar to the statement included in the patient information sheets that their decision to participate would not affect their care in any way.
Practical Lessons Learned
As well as the ethical dilemmas I encountered in seeking ethics approval for my study, I had a final set of practical challenges to overcome before I could complete my application process and start my research. I needed to identify the research sites and secure research and development (R&D) approval from the organizations where my research would take place. The challenges of doing this within a single NHS organization, and the need to plan for sufficient time, are detailed by Paton (2017). The object of my research study is integrated care, how different services from health and social care work together in the interests of people with complex health and social care needs. This means that I specifically wanted to look at the range of multiple services that patients used, from general practitioners (GPs) to community services, to social services and hospitals. I also wanted to understand how clinicians and practitioners in these different settings and organizations went about their practice.
I started the task of securing consent from multiple organizations by mapping out the services used by patients in the local area, including the acute hospital trust, the community and mental health services trust, and the local councils including social services departments. There had been a major reorganization of the NHS shortly before my study started which had led to the abolition of Primary Care Trusts (PCTs) and the establishment of CCGs. This caused difficulties as the new CCGs did not retain the research governance functions previously held by PCTs, and it took some time to identify a route to secure R&D approval for primary care in the absence of PCTs. I then had to identify the organizations responsible for these research sites and a named individual within each organization. This was complicated by the fact that each organization had different structures and processes for handling research. The NHS Trusts each had a research department and an infrastructure that was able to accommodate requests for R&D approval, the CCGs were in a different state as mentioned above, and the councils each had their own approach. One council dealt with this through social care, another through their corporate governance team, and the third had a research and analysis officer who processed my request. For each organization, it took some time to identify a named person who could consider my request. However, once this had happened, I was then able to discuss with that person the process that I needed to follow. Having ethics approval and one R&D approval gave me more confidence in approaching the next organization, and that organization was also assured to learn that my study had been favorably reviewed by others. Each organization then had their own set of forms to complete, and I ensured that I used a consistent approach to completing these by closely following the information previously submitted on the IRAS application, despite the slightly different queries each raised. The additional requirements imposed by these organizations also differed. For example, one council required that I complete their in-house safeguarding course before starting research, and one NHS trust issued me with an ID badge to be worn at all times on their sites.
In this case, the process of securing R&D approval was complicated because of the multiple research sites and organizations. Sufficient time, and knowledge of the (changing) organizational structures, was essential in working through this process. Researchers should carefully scrutinize the latest guidance on securing ethics approval, as this is continually updated.
When I embarked on my doctoral research, I was a novice ethnographer. I had to simultaneously get to grips with ethnography as a contested phrase (method, methodology, approach) from an interdisciplinary perspective (anthropology, sociology) and to defend and present my proposed study to the ethics committee, the operation of which proceeds in accordance with the biomedical framework of research. The defense of a qualitative and non-positivist approach to research was an important aspect of my research training. Rather than trying to dissemble, I tried to think about the vantage point of the ethics committee. From a position of recognizing that the ethics committee is concerned with weighing up the potential harms and benefits of the proposed study, I was able to present my research accordingly.
The ethical dilemmas and practical difficulties I faced are likely to arise for other students who are planning to take an ethnographic approach to researching health and illness. Furthermore, similar issues will arise for researchers seeking to study their own social worlds or when researching a multi-site or complex organizational setting. Working through these issues with sufficient time, reflexivity, and knowledge is part of the learning process of becoming a researcher; some of the strategies referred to in this case study can help with this process.
Exercises and Discussion Questions
- How will you manage (inevitable) change in your planned research? How will you communicate this to your ethics committee?
- Are there any ethical issues relating to your current position or previous experience that might affect your research? Do you hold any relevant positions (paid or voluntary) in your study site/field?
- What other attributes might affect how you recruit participants?
- Will you be working with “vulnerable people”? How are they designated as vulnerable and what does that mean for your ethics application?
- Map out the different organizations that might be relevant to your ethnography, think about healthcare organizations but also social care and voluntary sector services.