• ethnography

Originally developed within anthropology, ethnography is one of the most longstanding social science research approaches. Its emphasis is on the description and analysis of “the everyday”—routine behaviours in their natural settings. Many would characterise ethnography as the process of querying understandings and practices that are taken for granted: it renders the everyday world problematic by making the “ordinary” into the “extraordinary”. It is best understood as a holistic approach that does not rely on any single method of data collection. Observation, which may be unstructured “hanging out” or more structured and purposeful scrutiny of situations to look for particular things, is perhaps the defining feature of ethnography. These observations are often supplemented by interviews (sometimes very informal and part of the “hanging out” process) or documentary materials collected from the setting (e.g. posters, internal memos, reports of meetings), photographs, artefacts, and so on.

The interpretation of these data is very much a function of the researcher’s own skills and judgement,¹ and will usually involve searching for themes and patterns in the data and generating explanations and theories grounded in them. Like literary criticism, there is no prescribed set of procedures or techniques but there are expectations of good practice. For example, researchers are required to be reflexive—that is, to reflect on and be able to give an account of how they produced their interpretations—and to be able to show that their interpretation is warranted by the data.

Ethnography is not for the faint hearted. It is a time consuming and demanding research process and can be a profoundly uncomfortable experience for the researcher. Ethnographers have often identified a natural affinity for their methods with relatively disadvantaged groups, seeing themselves as offering these groups a voice and a means of making explicit systems of oppression and coercion. When the ethnographer’s role changes to one where s/he is attempting to explain professional practices, important problems can arise. Access to the field may be difficult to negotiate. When that has been gained, further challenges lie ahead. It can sometimes be difficult to access the people who may be most important to understanding a particular phenomenon: they may be impossible to get hold of, or unwilling to speak or be seen. Covert observation may be ethically unacceptable, but clearly obtaining consent to observations is not always easy and introduces risks of the ethnographer influencing the behaviours under observation. Participants in the process can begin to feel they are being inspected and judged, and to feel disempowered. They may “act up” in the presence of the researcher. The participants may also become hostile or uncooperative, and much depends on the skill of the researcher to overcome these obstacles. Sometimes this is done by creating an “insider” status, identifying the researcher with the group being studied—for example, a nurse studying nurses. However, such groups, having accepted the researcher as “one of us”, may feel betrayed by the research account that is subsequently produced. Some researchers begin to feel voyeuristic and exploitative in some settings, or to experience conflicting loyalties. Ethical dilemmas about when and how to intervene are not uncommon. In addition, ethnographic research will also be subject to criticism of its apparent subjectivity, and researchers can find themselves accused of producing partisan, partial, or misleading accounts.

Many of these difficulties should and can be overcome, particularly in making a contribution to quality and safety in health care. Much can be learned relevant to quality and safety from already published ethnographies. Strong’s account of the “etiquette rules” governing face-to-face interactions between parents and doctors² demonstrates how difficult it is for patients to raise concerns about possible errors: to do so disrupts the “ceremonial order” of consultations and puts their status as “good patients” at risk. Patients may therefore be silenced when it comes to pointing out things they are concerned about. Pope’s work on waiting lists demonstrated the importance of gatekeepers, both clinical and administrative, in controlling access to health care.³ An ethnographic study of interactions between consumers and pharmacy staff and patient interviews⁴ challenges traditional professional assumptions about how the public understands the role of pharmacists, with important implications for medication safety. Findings from ethnography can also be integrated with findings from other study types including quantitative or other qualitative research.⁵

Generally, however, the obvious potential for ethnographic approaches to make a contribution to the study of safety and quality in health care has been underexploited. Ethnographic research is well suited to identifying conditions of risk, particularly where these involve human performance, organisational and cultural dynamics, and interactions between people and technology. Ethnography is especially good at probing into areas where measurement is not easy, where the issues are sensitive and multifaceted, and where it is important to get at the tacit, not the already evident. It can capture the winks, sighs, head shaking, and gossip that may be exceptionally powerful in explaining why mistakes happen, but which more formal methods will miss.

The paper by Taxis and Barber⁶ in this issue of QSHC makes a welcome addition to the canon. It shows how, used carefully, ethnography can identify types of errors and provide explanations for why these occur, through interrogating the everyday understandings that staff have of their own practices and identifying the cultural contexts of practice. It is difficult to see how the latent conditions for error described in this paper, which are likely to be hugely important, could have been characterised appropriately using any approach other than ethnography. Much more research of this kind is needed, particularly in complex areas where long chains of causation exist in terms of health and safety outcomes.