Spotlight on research methods

Understanding our patients an anthropological approach

Alexander L A Reid, MBBS, FRACGP, is Emeritus Professor of General Practice, Faculty of Medicine and Health Sciences, the University of Newcastle, New South Wales.

Objective To provide an accurate description of the views of a suburban population on health, disease and its causes and the provision of medical and other forms of care. A specific purpose of this paper is to draw attention to the relevance of this research methodology and its similarity to general practice.

Method An ethnographic study. A discrete Australian coastal suburb with a broad socio–demographic mix.

Results The health beliefs of the population and their views on the delivery of care are briefly summarised. A fuller report is available from the General Practice Evaluation Program - Grant No. 275.

Conclusions The application of ethnographic methods of Australian suburbia provides new, valid and reliable information which is needed if the health messages and care provided to people are to be relevant.

Received 5 June 1997; accepted 27 August 1997

General practitioners (GPs) and other health workers build up an anecdotal picture of their working environment and of the people they encounter: while recognising that patients may see the world differently. This, however, has never been clearly described.

Medicine as a discipline is strongly conditioned to think of science in terms of the randomised double blind trial and to place a low value on supposedly soft qualitative data; but there is a much older tradition of science which aims to provide an accurate description of the world. It is inductive and not hypothesis driven. This study outlines the use of such a method to construct a coherent picture of the way in which people construct understandings and explanations in connection with the causes of illness and the basis of health: this is the concept of 'lay epidemiology'.1 It also provides data on the provision of medical and other health services.

Methods

The study setting

Oceanpoint is a predominantly residential suburb with a population of 2732 people. There are a variety of light industries nearby, but many residents travel to Newcastle to work. It is geographically defined and isolated from nearby suburbs by bushland reserves, the ocean and a large area used for landfill waste disposal. A former mining community, its small size and sense of community identity made it ideal for intensive ethnographic research.2,3 Geographically and economically it subdivides into:

The bluff - newer houses, professional and tradespeople (n=258); median income $40000–50000.
The flat - old Oceanpoint, lower paid workers and retired people (n=326); median income $35000–40000.
The caravan parks - substantial numbers of unemployed 'permanent' residents (n=333); median income $20000–25000.

Other respondents came from a large retirement village and from a prosperous new estate. This internal differentiation provided a range of household types, ages, socio-economic levels and phases of the domestic cycle broadly representative of an Australian population. The entire area is serviced by: three GPs with further easy access to GPs in nearby suburbs; a pharmacist; medical ancillary services; and a number of alternative health services.

Research design

An ethnographic research design incorporated the following formal methods of data collection:

one hundred and twelve semi-structured interviews with residents of differing socio–economic backgrounds and health needs, using snowball sampling (residents interviewed refer the researcher on to another resident and so on). Informed consent was obtained.
seven focus groups with contrasting membership;
health diaries4 over 6 month periods were examined and discussed with 23 families involving 64 people.

Recruitment, sampling and interpretation were made possible by participant observation. This required the researcher to live in and become involved with the community in order to study processes, relationships, continuities over time and patterns of behaviour as well as the broader socio–cultural context from an insider's viewpoint. One of the team had lived in the area for 8 years, and the research associate moved there for 18 months accompanied by her husband who undertook a number of the interviews, particularly with male informants.

Details of the methodology have been described by Whittaker,2 who draws attention to the strong parallels between the method of ethnography and that of general practice, although the objectives are very different. Each discipline involves interaction and building relationships with a variety of people over long periods of time and the establishment of understanding people's perspectives, experiences and values. Information from each source was transcribed verbatim.

Analysis

All transcribed material was organised using 'Ethnograph 3.0', a computer program that facilitates the management, coding and analysis of qualitative data.5 The use of the program allows the researcher, when reading the text, to note and code the themes that emerge. At a later stage it provides rapid retrieval of segments of text from any of the sources that have been coded.

The process of analysis of this data involved its reduction into useable forms allowing close examination and comparison of themes and the development of concepts from which it is possible to build up theories about the relationships between the emerging concepts. Particular cases were also examined in detail, especially in relation to the narratives or illness trajectories across time. This allowed us to develop ideas about how a person organises the description of their illness and what this might say about the person's experience, and the framework within which they view it, which in turn may influence the service(s) used. From individual cases it was possible to build models to be tested against other cases. Comparison and matching of findings from a number of different research methods and contexts, as well as input and feedback from the various members of the research team, ensured a high level of validity and reliability in the findings. This process of analysis proceeded throughout the study.

The uniqueness of this approach

Unlike quantitative research where the researcher begins with a specific hypothesis which is then proved or disproved (the deductive approach), qualitative studies such as this one are inductive. As Patton comments:

Because each qualitative study is unique, the analytical approach used will be unique. Because qualitative inquiry depends at every stage on the skills, training, insights and capabilities of the researcher, qualitative analysis ultimately depends on the analytical intellect and style of the analyst. The human factor is the great strength and the fundamental weakness of qualitative inquiry and analysis.

In this study, although the analysis was primarily directed by the experienced research associate, there was close interaction and agreement with other members of the team, thus reducing the potential for idiosyncratic interpretations.

One of the informants commented on the similarities between the qualitative research process and general practice when he said: So you're actually like a doctor to us; you'll sit and listen to what we're saying and you're going to work out your own thing, your own mind.

Results

Understanding health and illness

A number of distinct concepts about health emerged with some respondents defining an overlap and others adopting a narrow definition:

Absence of illness was frequently taken to correspond to a state of health, although less so for older respondents.
Bad health involves serious ongoing problems that affect a person's ability to function and fulfil social obligations.
Health is associated with a positive attitude and is quite compatible with the presence of chronic or serious conditions, for example, those with a history of heart attacks or cancer did not see themselves as unhealthy. A negative attitude was seen as making one more vulnerable to disease.
Health is an ascribed status: luck, one's genes and fate emerged strongly as factors determining health. Some people are seen as lucky in having a 'strong constitution'. This view does not fit well with the ideologies of health promotion.
Health is an achieved status: we should work at being fit, and if we are unhealthy then this may be our own fault. This strongly moral view of personal responsibility emerged as more in tune with the messages of health promotion.
Heath was frequently seen as being linked with a healthy environment, especially colouring the community's views about cancer.
Health as physical fitness was associated with images of the body as a machine which has to be well maintained. It links to ideas of health as lifestyle and ability to function.

Lay epidemiology

This term refers to the process by which people construct understandings of health, disease and death. It arises from their personal experience, social beliefs and interpretation of information from the media and a variety of health information sources. This epidemiology overlaps but differs significantly from scientific epidemiology. The population generally is aware of official health messages but chooses to interpret these for themselves. For example, Davison draws attention to 'Uncle Norman' who ate bacon and eggs every day (and probably smoked as well) and lived till he was 93.

The role of stress, fate and heredity

In Oceanpoint, the picture emerged clearly of the 'coronary candidate' who had the wrong parents, was 'unlucky' and so a potential victim of fate, aggravated by stress, which was seen to arise from a number of different sources, including children (especially teenagers), work, and high expectations associated with modern lifestyle. It was strongly believed that stress is a major causative factor in coronary disease. A natural corollary of this is that the best thing to do to avoid a heart attack, since you cannot influence luck or your genes, is to do the things that relieve stress, namely have a few drinks, smoke and eat the foods you really like! Those who hold this belief, although well aware of the current health messages, are likely to be unresponsive to them. Older men who had experienced heart disease did acknowledge accepted risk factors and some had modified their behaviour accordingly.

Cancer, the environment and health

While acknowledging the role of heredity, smoking and diet in the causation of cancer, the people of Oceanpoint expressed firm views that their environment, despite appearing healthy (seaside, sunshine and fresh air), was a significant factor in causing cancer. They identified two principle sources:

The extensive sandmining which had occurred in the past and was known to expose sources of radiation previously buried.
Leachate from the local tip is also regarded as a source of potential cancer. Investigations by the Public Health Unit have not shown a surplus of cancer cases, but this information has not affected community perceptions.

Community perceptions of GPs and specialists

Our informants described themselves as active participants in their relationships with their GPs, negotiating decisions, resisting advice through non-adherence and choosing and changing their practitioners when dissatisfied. Most people found, or hoped to find, a doctor they 'got on with'. This relationship was generally used for more serious problems, while quick and easy problems could be dealt with by any GP. The former was more valued by mothers with young children and older people than by the young adults. GPs were selected for their caring characteristics: women stressed their need for someone to accept their emotional problems, whereas men valued directness and honesty. Competence was largely taken for granted.

There was a strong distinction made between GPs and specialists who were commonly described in terms of their technical competence but criticised sharply for their general lack of empathy and communication skills. This lack of skill was partly explained by not having the same choice to select a specialist in the way they do a GP, and was partly forgiven because technical expertise is valued.

This value was reflected in people's views about payment for medical services: they had few problems with paying specialists with whom they had no personal relationship, but they felt that payment to GPs was in conflict with their caring, friendly relationship, and was often handled insensitively. They acknowledged that doctors run a business, but there was no doubt that this interfered with images of a 'dedicated profession'. Some informants talked nostalgically about the good old style doctor, particularly for their willingness to do home visits and after hours calls, but others felt that the new doctors were better at listening. The practice at Oceanpoint does home visits but uses a locum service out of hours and this impersonal service, often further marred by requests for money up front, was felt to be professionally inferior because the doctor could not 'know my case' and involved long waits and short consultations. While acknowledging doctors' needs for time with their families and for relaxation, use of a locum service was perceived as a violation of the caring dedicated relationship.

Community use of health resources

Apart from biomedical health services, Oceanpoint residents used a wide variety of home therapies, medicines and vitamins from the pharmacist, and alternative therapies, often combining these into complex therapy regimens coordinated by the sufferers themselves or significant others. There was no distinct boundary between those using medical and complementary therapies, which were often taken with the approval of medical practitioners, especially when medicine had not produced relief and the alternatives were perceived to be harmless. The only conflict arose over the advice from some advocates of natural therapies that children should not be immunised and practitioners' concerns that alternative therapies may divert patients with serious illnesses from seeking appropriate treatment. Enquiries showed a variety of views about causation and management of illness. The causal roles of cold weather and climatic change and stress were prominent. Management models included 'natural' remedies perceived as less likely to cause dangerous side effects than prescription drugs or over-the-counter pharmaceuticals; mechanical corrections by manipulative therapists with the laying on of hands, seen as being valuable both physically and in reducing stress. A number of informants emphasised the mind-body distinction but many rejected this and appreciated a 'holistic approach' which they felt was provided by some alternative practitioners. Blanket approval of alternative therapies was rare: people might approve one modality and reject another, or use alternative therapies for certain conditions and biomedical care for others.

Conclusions

The research method outlined briefly has yielded a unique data set whose richness can only be hinted at in this brief summary. If they are to deliver effective health care, doctors need to be aware of their patients' perceptions of health and illness as well as their evaluation of general practitioners within the spectrum of heath care services available.

References

1. Davison C, Smith G D, Frankel S. Lay epidemiology and the prevention paradox: the implications of coronary candidacy for health education. Sociol Health Illness 1991; 13:1–19.
2. Whittaker A. Qualitative methods in general practice research: experience from the Oceanpoint study. Fam Pract 1996; 13:310–316.
3. Whittaker A, Freeman S, Reid A, Connor L, Robinson K, Sprogis A. The Oceanpoint Study: General Practice in its Community Context. Report of the General Practice Evaluation Program (GPEP) Grant No. 275. GPEP, Dept of Health & Family Services, Canberra, 1996. (Copy available National Information Service, Dept of General Practice, Flinders University, Bedford Park, SA 5042.)
4. Bentzen N, Christiansen T, Pederson K M. Self-care within a model for demand for medical care. Soc Science Med 1989;29:185–193.
5. Seidel J V, Kjolsetn R, Seymour E. The ethnograph: a user's guide (Version 3.0). Amherst M A: Qualis Research Associates, 1988.
6. Patton M Q. Qualitative evaluation and research methods. Newbury Park: Sage publications, 1990.

The Royal Australian College of General Practitioners

Consulting skills