Article Text

Download PDFPDF

Using ethnography to study improving healthcare: reflections on the ‘ethnographic’ label
  1. Caroline Cupit,
  2. Nicola Mackintosh,
  3. Natalie Armstrong
  1. Department of Health Sciences, University of Leicester, Leicester, UK
  1. Correspondence to Professor Natalie Armstrong, Department of Health Sciences, University of Leicester, LE1 7RH, UK; na144{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

While methods broadly described as ‘ethnographic’ have been increasingly employed to research the organisation and delivery of healthcare,1–4 a single or widely accepted definition of ethnography has proved elusive and perhaps unnecessary.1 5 Nonetheless, even as authors publishing in this journal have adapted ethnographic approaches for the purpose of studying improving quality and safety in healthcare, they have often attempted to retain some of its anthropological ‘essence’.6 For instance, Dixon-Woods7 characterises ethnography in terms of its focus on observational methods, questioning of the taken for granted, description and analysis of routine behaviours in their natural settings, and use of the researcher’s own skill and judgement to both gather data and to interpret them drawing on social theory.

In a recent debate over use of the ethnographic label in this journal, Jowsey8 argued that ethnography was not simply a method of collecting data but also included theoretical analysis and interpretation of those, and that it requires a researcher’s recognition of their own positionality (i.e., where the researcher ‘sits’ in relation to those he or she is studying, e.g., in terms of gender, culture or power). Waring and Jones9 also drew attention to ethnography as an account of the ‘social and cultural organisation of ‘everyday life’’, and to the researcher’s insider perspective. Although these authors have not been ‘purist' about the ethnographic label, they have strongly advocated that researchers using the term ‘ethnography’ should retain a commitment to a ‘unified understanding of ethnography’9 as a package of methods, methodology and, importantly, the production of an account shaped by the researcher’s skilled interpretation. Bosk describes this account as a ‘resonant description’ or a ‘verisimilar account’—one in which its readers can trust.1

In BMJ Quality & Safety, Vindrola-Padros and Vindrola-Padros10 report a systematic review of the use of ‘rapid ethnographies’ in healthcare organisation and delivery. They argue that rapid approaches to ethnographic data collection are important in health services research for ‘generating findings within timeframes when they can still be actionable and used to inform improvements in care’. Based on the studies included in their review, they create a typology and working definition of ‘rapid ethnography’ as a new subcategory of ethnography.

We welcome their contribution for its scoping of a field of studies that self-identify as both ‘ethnographic’ and ‘rapid’, and for its seeking to address a perceived need for speedy findings. The authors’ pragmatic decision not to try to define either ‘rapid’ or ‘ethnography’ is understandable for the study they undertook, but we see value in opening up a space for discussion and debate around the need for a new category of ‘rapid ethnography’ and believe that further work considering these labels and what they mean would be helpful. For example, time constraints, which Vindrola-Padros and Vindrola-Padros argue is the defining characteristic of the included studies, are arguably characteristic of many ethnographic studies of healthcare, and it would be useful to consider what distinguishes the studies included in the review from the broader body of work calling itself ‘ethnographic’ but not necessarily ‘rapid’. One question is whether this further subcategorisation reflects an unhelpful elasticity in the label of ethnography. Another is whether the term ‘ethnography’ has become shorthand for describing just the main methods by which researchers gather data (e.g., a combination of observation and interviews)5 rather than meaningfully representing a methodological approach and body of work.

Underpinning the categorisation of ethnographies as rapid or otherwise is, of course, the dimension of time. There is certainly no reason why time alone should determine the usefulness of any ethnographic work; relevant here is Pink and Morgan’s analytical distinction between intense, anthropologically rooted and theoretically engaged short-term ethnographies, and alternative approaches that use the ethnographic label but fail to produce ‘distinctly anthropological ways of understanding’.11 Interestingly, some of these alternative approaches, such as Rapid Ethnographic Assessment and Rapid Assessment Procedures (which Pink and Morgan imply do not carry the essence of ethnography), were included in Vindrola-Padros and Vindrola-Padros’ review as examples of ‘rapid ethnography’.

For ethnography, what may be important in relation to time spent ‘in the field’ may include the ability to achieve immersion in the setting and engaging with less powerful voices in order to better understand context. Pink and Morgan argue that short-term ethnographies are still capable of producing valuable ways of knowing about people and the environments of which they are a part.11 What needs further consideration is the nature of this knowledge and how it is gained. Insights linked to temporal features such as hospital or ward rhythms, and serendipitous events, might risk being compromised by time-limited approaches. On the other hand, the benefits of employing a more focused approach in terms of avoiding burdening sites and supporting access to a greater number and diversity of sites may not be trivial. Reflections on these kinds of questions and trade-offs in a principled way will help build our methodological insights into the value and role of ethnography (and also its limitations) for understanding efforts to improve healthcare quality and safety.

Lastly, this review of rapid ethnographies perhaps gives reason to reflect on the tensions imposed by time constraints that may be driven by demands of funding or quality improvement agendas. The review’s focus on studies generating rapid findings that are immediately ‘actionable’ focuses attention on the issues associated with producing very practical findings and doing so at speed. For us, an important part of the value of ethnography for studying the improvement of quality and safety in healthcare is to explicitly seek to expose the nuances of culture and what actually happens in the setting (work as done rather than work as imagined). This requires an openness to question how well quality improvement endeavours are aligned with the cultural context and the interests of those working or receiving care in the setting (see, e.g., Mackintosh et al 12). Ethnographers recognise that the idea of a ‘neutral evidence base’ which can simply be ‘implemented’ or ‘translated’ into practice is problematic, and that ‘ethnography can make an important contribution to the debate about evidence itself’.13 However, this may well require time in terms of establishing new relationships and trust, in order to bear witness to and understand aspects of ‘backstage’ as well as ‘frontstage’ practice, recognising that how people behave and interact with others is shaped by the time and place in which this occurs, as well as by the ‘audience’ present to witness it.14

We encourage further debate about the use and value of the ethnographic label. For those of us who believe the term ‘ethnography’ properly applies to our approach in practice, we join with other previous commentators in urging attention to the ‘essence’ of ethnography and to furthering understanding of difference within this ‘ethnographic’ label and the implications of this for research, policy and practice.



  • Funding Caroline Cupit is supported by a Health Foundation Improvement Science PhD Studentship; Natalie Armstrong is supported by a Health Foundation Improvement Science Fellowship.

  • Competing interests None declared.

  • Patient consent Not required.

  • Provenance and peer review Commissioned; internally peer reviewed.

Linked Articles