Education and debate

Anthropology in health research: from qualitative methods to multidisciplinarity

Helen Lambert, senior lecturer in medical anthropology ^a, Christopher McKevitt, research fellow in social anthropology ^b. 

^a Department of Social Medicine, Bristol University, Bristol BS8 2PY, ^b Department of Public Health Sciences, King's College London, London SE1 3QD

Correspondence to: H Lambert H.Lambert{at}bristol.ac.uk

As a response to concerns about the standard of qualitative research, attention has focused on the methods used. However, this may constrain the direction and content of qualitative studies and legitimise substandard research. Helen Lambert and Christopher McKevitt explain why anthropology may be able to contribute useful insights to health research

Qualitative methods are now common in research into the social and cultural dimensions of ill health and health care. These methods derive from several social sciences, but the concepts and knowledge from some disciplinary traditions are underused. Here we describe the potential contribution of anthropology, which is based on the empirical comparison of particular societies. Anthropology has biological, social, and cultural branches, but when applied to health issues it most commonly relates to the social and cultural dimensions of health, ill health, and medicine.¹

	What is wrong with qualitative research?

Explaining qualitative research to health professionals has been an essential step in gaining acceptance of these techniques.² However, findings from such research have been deemed "thin," "trite," and "banal."³ Concerns about standards and the need for particular types of evidence have led to quality control measures being recommended for qualitative health research (procedures such as multiple coding, purposive sampling, and software packages for text analysis). Imposing these measures, however, may constrain the direction and content of qualitative studies⁴ and legitimise substandard research, as the procedures recommended can be incorporated without enhancing the quality of the empirical work or the analysis.⁵

The main problem with the quality of qualitative research in health lies not in the methods but in the misguided separation of method from theory, of technique from the conceptual underpinnings.⁶ Qualitative research is in danger of being reduced to a limited set of methods that requires little theoretical expertise, no discipline based qualifications, and little training. Such an exclusive focus on method should be resisted, an argument that parallels an ongoing debate in epidemiology. ^{7 8} Multidisciplinary research is necessary for investigating, understanding, and improving health, but simply using qualitative methods does not constitute multidisciplinarity. What is needed is not narrower specification of technical operations or better quality control procedures. Instead, we need research methods that are less generic, less atheoretical, and less narrowly focused, together with a more widespread application of concepts and knowledge originating in source disciplines.

Specifically, we advocate more anthropology. In the United Kingdom, the growing appreciation of anthropology as a contributory discipline to health research and health care has not been matched by efforts to incorporate its theoretical basis (sociology has a better established history of application to health issues). Anthropology has a distinctive approach to gathering and interpreting data that can yield productive insights. These insights derive from underlying assumptions about the nature of social reality and human action, as well as using participant observation (anthropology's most characteristic research strategy, which involves direct observation while participating in the study community and includes other methods, such as interviewing). ^{9 10} The following sections outline some basic characteristics of an anthropological approach with particular value for health research.

	"Our" knowledge and "their" beliefs

A core conceptual feature of anthropology is that what is "rational" is seen to be socially and culturally specific and valid in its local context. The salience of this view for understanding participants (other than patients) and issues in health care is not generally appreciated. Using a biomedical approach to problems in qualitative health research results in a narrow investigation of "lay" beliefs (and occasionally, practices), often with the intention of translating these to professionals, to inform ways of improving adherence to their interventions. An anthropological approach does not assume that biomedical concepts and practices are both normative and universal. Rather, it regards the knowledge and practice of "experts" as locally variableas are the knowledge and practice of lay peopleand it includes both within the boundaries of empirical inquiry. Some of the most relevant anthropological research for evidence based health care has considered differences between epidemiological, clinical, and popular concepts of health and disease in particular contexts and has thereby shed light on the implications of such distinctions for appropriate practice in these settings. ^{11 12}

A more general point is that qualitative research need not and should not be restricted to discerning and describing the ideas or practices of lay participants but should encompass those of professionals too. The study of health professionals' discourses and ideologies draws on a rich tradition in the social sciences of the social and cultural construction of biomedical knowledge. However, such study also links with a trend in medical anthropology that argues for the need to focus beyond clinical encounters between individuals to the power relations that produce and shape sickness (box 1). ^{13 14}

	Actions speak as loud as words

As box 1 shows, what people (including health professionals) say can be different from what they think and do. This goes unrecognised in most health research that is designated "qualitative" but which in fact relies mainly or solely on interview based methods.¹⁶ The ambiguous relation between language and action fundamentally informs anthropological research using participant observation. Ideas about treating illness and lay explanatory models, for example, are shaped by contingent circumstances and forms of practical "reasoning in action" that are not always expressed orally, especially in one-off interviews, which tend to produce orthodox responses. Qualitative health research often fails to distinguish between normative statements (what people say should be the case), narrative reconstructions (biographically specific reinterpretation of what has happened in the past), and actual practices (what really happens). Anthropological practice ensures awareness of these distinctions even when interpreting interview data, by "situating" an interviewee's statements and the circumstances of the interview as far as possible in the broader context of that person's life. Participant observation may not always be feasible or appropriate given constraints on time, funding, and expertise, but the methodological lessons from anthropology are transferable. These lessons are that words cannot be taken at face value and that naturally arising informal situations involving talk and action are more useful than formal interviews in highlighting this.¹⁷

	Context specificity and comparative evidence

A key anthropological contribution to health research lies in its empirically based grasp of the context specific nature of social processes. This focus on the particular, which anthropology insists on through documenting the complex details of everyday life, provides an important corrective to misleading generalisations and abstractions that can, according to Singer, "grotesquely flatten" the diversity of different settings.¹⁸ However, analysis of specific situations or cases can also provide more general insights into the type of phenomenon under study, through anthropology's comparative approach. Comparing primary data with secondary evidence about similar issues (such as a particular health problem) in different settings can produce stronger analytical insights with greater potential generalisability. This is achieved through logical (rather than statistical) inferences that make use of relevant empirical knowledge and theoretical principles.¹⁹

Just as most health professionals specialise in particular diseases or body systems, so most medical anthropologists specialise in particular regions of the world or topics. This specialist knowledge is a major source of comparative evidence and, like clinically specific knowledge, it is informed by core disciplinary concepts (such as classification, ritual, and symbolism) and theoretical approaches (such as those of political economy or cultural interpretation) (box 2).

(Credit: TOPHAM/FOTOMAS) Anthropology has its roots in a Western fascination with the "exotic," in trying to make the strange comprehensible; anthropologists working in health today question the familiar

	Questioning categories

Qualitative methods of data collection have become popular in health research mainly because they are seen to "reach the part other methods cannot"that is, the views of ordinary people in the real world.²³ Implicitly, the methods are a valuable but purely functional means of gathering data to answer an initial research question. Hence the bulk of qualitative work in, say, health services research, seeks to discover (through semistructured interviews and/or focus group discussions) people's views of a biomedically defined phenomenonfor example, a disease or a health service. Although such research can undoubtedly be useful in operational terms, genuinely new insights are rarely obtained because this approach fails to incorporate a central feature of social science researchthat of reconfiguring the boundaries of the problem.

A particular way that anthropology achieves this is by its focus on classification and meaning. This interest probably derives from anthropology's development as a discipline associated with the ethnographic study of "other" cultures, in which the nature and boundaries of apparently basic categoriessuch as family, religion, and medicinecould not be presumed but required empirical investigation. Thus an anthropological approach, rather than taking phenomenon x or y as a given and investigating views of or beliefs about it, also investigates the form and contents of the thing (x or y) itself. Insights derive both from examining the nature and meanings of apparently familiar categoriesfor example, clinical terminologies, or health service constructs, such as "patient satisfaction"and from investigating how and why such categories are constructed and maintained (box 3).

	Conclusion

Anthropology has its roots in a Western fascination with the "exotic" and the associated attempts to make the strange comprehensible. Anthropologists working in health settings today question the apparently familiar so that health issues may be better understood and health outcomes improved. This is a key promise of qualitative research generally for health professionals. Anthropology can offer relevant conceptual frameworks, substantive knowledge, and methodological insights. These are essential for truly multidisciplinary research, which extends beyond selective incorporation of specific methods to encompass research conceptualisation and theoretical synthesis. Funding sources, institutional support, and publication requirements should reflect this.

	Footnotes

Funding: None.

Competing interests: HL is the chair and CMcK is a member of the Royal Anthropological Institute's medical committee, which advises the institute on medical anthropological matters and presents and promotes anthropological perspectives and understanding among non-anthropologists working in health related fields.

	References

1.	Lambert H. Encyclopaedia of social and cultural anthropology. Medical anthropology. London: Routledge, 1996:358-361.
2.	Mays N, Pope C, eds. Qualitative research in health care. London: BMJ Publishing, 1996.
3.	Caan W. Call to action. BMJ 2001 bmj.com/cgi/eletters/322/7294/1115#14398
4.	Barbour R. Checklists for improving rigour in qualitative research: a case of the tail wagging the dog? BMJ 2001; 322: 1115-1117[Free Full Text].
5.	Williams B. Longer checklists or reflexivity? BMJ 2001 bmj.com/cgi/eletters/322/7294/1115#14196
6.	Popay J, Rogers A, Williams G. Rationale and standards for the systematic review of qualitative literature in health services research. Qual Health Res 1998; 8: 341-351[Abstract/Free Full Text].
7.	Krieger N. Epidemiology and the web of causation: has anyone seen the spider? Soc Sci Med 1994; 39: 887-903.
8.	Davey Smith G, Ebrahim S. Epidemiologyis it time to call it a day? International J Epidemiol 2001; 30: 1-11[Free Full Text].
9.	Savage J. Ethnography and health care. BMJ 2000; 31: 1400-1402.
10.	Ellen RF, ed. Ethnographic research: a guide to general conduct. London: Academic Press, 1984.
11.	Kaufert P, O'Neill J. Analysis of a dialogue on risks in childbirth: clinicians, epidemiologists, and Inuit women. In: Lindenbaum S, Lock M, eds. Knowledge, power and practice: the anthropology of medicine in everyday life. Berkeley, CA: University of California Press, 1993:32-54.
12.	Davison C, Frankel S, Davey Smith G. "To hell with tomorrow": coronary heart disease risk and the ethnography of fatalism. In: Scott S, Williams G, Platt S, Thomas H, eds. Public risks and private dangers. Aldershot: Avebury, 1992:95-111.
13.	Baer H. How critical can clinical anthropology be? Med Anthropol 1993; 15: 299-317[Medline].
14.	Morsy S. Political economy in medical anthropology. In: Johnson T, Sargent C, eds. Medical anthropology: contemporary theory and method. New York: Praeger, 1990:26-46.
15.	Rapp R. Chromosomes and communication: the discourse of genetic counselling. Med Anthropol Q 1988; 2: 143-157.
16.	Power R. Never mind the tail, check out the dog. BMJ 2001 bmj.com/cgi/eletters/322/7294/1115#14358
17.	Lambert H. Methods and meanings in anthropological, epidemiological and clinical encounters: the case of sexually transmitted disease and human immunodeficiency virus control and prevention in India. Trop Med Int Health 1998; 3: 1002-1010[CrossRef][ISI][Medline].
18.	Singer M. The application of theory in medical anthropology: an introduction. Med Anthropol Q 1992; 14: 1-8.
19.	Clyde Mitchell J. Case and situation analysis. Sociol Rev 1983; 31: 187-211[ISI].
20.	Del Vecchio Good M, Munakata T, Kobayashi Y, Mattingly C, Good B. Oncology and narrative time. Soc Sci Med 1994; 38: 855-862.
21.	Gordon D. Embodying illness, embodying cancer. Cult Med Psychiatry 1990; 14: 275-297[ISI][Medline].
22.	Di Giacomo SM. Can there be a "cultural epidemiology"? Med Anthropol Q 1999; 13: 436-457[CrossRef][ISI][Medline].
23.	Pope C, Mays N. Reaching the parts other methods cannot reach: an introduction to qualitative methods in health and health services research. BMJ 1995; 311: 42-45[Free Full Text].
24.	McKevitt C, Wolfe C. Quality of life: what, how, why? The views of health care professionals. Qual Ageing 2002; 3: 12-19.

(Accepted 14 February 2002)