Participant observation was conducted to explore the understanding of diabetes and examine the implications of these understandings for providing effective patient centered care in Cameroon. Ethnographic techniques—content and thematic analysis—were used to analyze the data collected from diverse techniques. Most participants distinguished “natural,” “supernatural,” and “man-made” causes of diabetes. Such aetiologies guided the behaviour and approaches adopted for treatment and helped explain why biomedical and traditional healing frameworks could so readily be used in tandem. Clinical encounters are often only one small part of the diabetes care process, alongside recourse to traditional medicine. With rituals, agents causing diabetes are apparently more convincingly explained as powerful reinforcement and a cure promised in traditional medicine. Though it seems “irrational” and dangerous to clinics when patients alternate between therapeutic regimes or pursue both simultaneously, it seems perfectly rational and beneficial to patients and beyond. So long as biomedical practitioners fail to recognize that their patients will probably also have recourse to traditional medicine, they and their services may compound the problems they face for patients to discuss openly how they have been managing their condition.
Although sub-Saharan Africa remains afflicted by very high rates of communicable diseases, noncommunicable diseases occur at age specific rates above those found in rich countries, such as those of Western Europe and North America [
However, without a multisectoral, all-of-society approach, the disturbing projections will be realized [
However, there is an ongoing debate about the contribution of other health care partners, like traditional healers (TH), in chronic disease care in Africa. This has been very topical in South Africa where they have been pulled in and trained to educate people about HIV/AIDS [
WHO’s Innovative Care for Chronic Conditions Framework [
The study setting compared two health districts: one urban and one rural, located in Yaounde, Cameroon’s capital, and Bafut, in Northwest Cameroon, respectively. In these districts, eight primary health care units were selected from which diabetes patients sought care. Tertiary care diabetes centres in Yaounde also formed part of the patients’ treatment networks. The study took place over a 25-month period from June 2001 to June 2003, with twelve months of ethnographic fieldwork in each place.
Ethnography entails the intensive study of people in their cultural contexts; it aims to build detailed descriptive accounts of social life and culture integrating several qualitative methods [
Matrix of methods used in participant observation during 25 months of fieldwork.
Method of data collection | Brief description of method | Number/episodes and participants |
---|---|---|
Daily observations | A research method, in which the investigator systematically watches, listens to, and records the phenomenon of interest. A research method in which the investigator takes part in the social phenomenon of interest by participating with a group and observing the interactions between them and between the researcher and subjects to achieve a greater understanding. Used in ethnographic approaches. This entails observing clinical encounters and activities of groups and individuals and at times participating actively. | Numerous observations of 20 patients within their families and community |
|
||
Clinic observations | Observing and participating in clinical encounters and other activities going on in clinics. | Over clinical encounters in four clinics with 200 patients in more than 300 encounters. |
|
||
Case studies | A research method which focuses on the characteristics, circumstances, and complexity of a single case, or a small number of cases, often using multiple methods. The case is viewed as being valued in its own right and whilst findings can raise awareness of general issues, the aim is not to generalise the findings to other cases. Heritage & Atkinson (1984) | 18 |
|
||
Biographies | Information provided by an individual or groups of people, tracing their life events and how they coped with diabetes | 18 |
|
||
Fieldwork conversations | The central goal of was the description and explication of the competence that ordinary speakers use and rely on in participating in intelligible, social interaction. At its most basic, the objective was one describing the procedures by which fieldwork conversations amongst participants produce knowledge of processes that build around diabetes care. | Daily as situations permitted |
|
||
In-depth interviews | The research method involved asking questions from an interview guide and tape-recording participants’ replies. This was recorded by the researcher writing down answers verbatim and summarizing the findings allowing data analysis to be conducted later | 34 |
|
||
Focus group discussions (FGD) | A qualitative technique developed by social and market researchers in which 6–12 individuals are brought together and interactively give their views and impressions upon a specified topic. (Morgan 1992). These were used to evaluate attitudes towards the topic. Focus groups conducted in this study were usually homogeneous with members being generally of the same age, gender, and status to encourage participation | 27 |
|
||
Documentary evidence | Using documents to get information to complement what other methods cannot provide. Clinical and handheld records of patients were used in this fieldwork. | Observation of clinical records and their use at the same time as observation of clinical encounters (>200) |
Twenty patients and their families, ten each in Yaounde and Bafut, were selected for close follow-up. Eight families were visited more or less daily over extended periods; the rest were visited weekly, sometimes more than once. This was more intensive in the rural than in the urban district. The mean number of family members was seven. The duration of visits ranged from 30 minutes to two hours. Meanwhile, more than two hundred patients were part of the network of observations in the clinics and in the communities where they lived. That is, all patients at the sampled clinics formed part of the observation network within the clinic during consultation sessions and out of the clinical settings. Ten traditional healers were the main frame of traditional healers observed regularly, at least daily for some and biweekly for others. Twenty-four diabetes consultants were the main biomedical health care providers (BHCP) observed in clinical encounters, though the activities of others within these care units were closely monitored. The wider communities of these health districts were part of the daily fieldwork observation. Table
An ethical clearance was obtained from the National Ethical Committee of Cameroon and authorisation to conduct the study from the Ministry of Scientific Research. All health care providers formed part of the units of observation but those caring for diabetes patients were the main focus. Patients were identified and selected in diabetes clinics. The consent of fieldwork participation was sought as to whether participants would accept being observed in clinical encounters and in their homes. Once they accepted, patients were observed in the clinics and later in their families and communities. While in the families the consent of family members was equally sought. Informed consent was documented on print for patient and health care providers and in audiotapes and orally for other participants. Observations in the families gave insights into treatment networks beyond the clinic, leading us to traditional medicine and traditional healers.
Aspects of the daily lives of patients, family members of patients, traditional healers, health care providers, and the wider community were the focus of the ethnographic fieldwork. Participation was moreover sustained simply by “being there,” on hand to take part in family or community events, or to engage in conversations with whoever was present in a variety of contexts, extending well beyond what might be considered as strictly health related. This was particularly the case in the rural context of Bafut.
All tape-recorded materials of conversations, interviews, and focus group discussions were transcribed and fieldnotes completed for all observations, conversations, biographies, and case studies. Documentary evidence was summarised from clinic and patient held records.
Time was spent listening and transcribing recorded interviews and focus group discussions. Fieldnotes taken in the course of observations were completed. The transcripts and documentary evidence were typed into word processing software. The transcripts captured features of conversations such as emphasis, speed, tone of voice, timing, and pauses. These elements made up the crucial aspect of interpreting data, given the multiple events that produced them. The data from transcripts were read transcript by transcript. Reading and completing the transcripts and making notes from them permitted us to identify ideas, make observations, and get insights and inferences. I organized the data and sorted out all the materials needed to enable getting on with the initial analysis. All transcripts were labeled with file names, typed in a word processing package, and stored using these file names. Once each transcript was read and classified, it was dissected, pulled apart, and scrutinized transcript by transcript to enable the full understanding of the nature of the data we were collecting.
The coding process was a mechanical way of assigning names to portions of text. The coding involved reading through the data, identifying themes and patterns, and placing them under headings or brief descriptions summarizing what they mean. Sections of text that related to a code were cut and pasted into the document created to represent that code as we worked through each transcript. It enabled us to build up a logbook of relevant text related to each created code. The source of each text that is collated to the codes was labelled so that I could easily refer back to it and understand the context in which the text was produced. Some text fitted within more than one code. In the first reading of transcripts, the codes were general. Subsequent reading enabled the splitting of the general codes into more specific ones. Through the coding process, similar portions of data were pulled together. Codes provided a fast snapshot of similarities, differences, patterns, themes, categories, concepts, and relationships from huge amounts of data sets across transcripts. The codes permitted the linking of initial transcripts and fieldnotes to subsequent ones. These codes were the main emerging ideas from the data.
Computer software was not used to code data. Rather manual coding using both pen and paper to code manual transcripts was conducted and the computer to code electronic transcripts. A logbook of codes was developed and stored electronically and manually. We wanted to try an experience that could be documented and shared with people who do research in settings with limited access to technologies that can facilitate data management.
A multistage analysis was done by sorting and pulling the themes together and verifying them to get emerging details. The analysis draws a bit on Garfinkel [
The analyses comprised a primary analysis stage and profound analysis stages, with supplemental data collected at subsequent stages, then further analysis. The primary analysis for this study included preliminary recognition of broader themes and the subsequent detection of themes and subthemes that required more precise analysis. This widened the meanings inherent in texts to overcome any distortions. At the primary analysis stage, the analysis of themes and the development of concepts were established with more accuracy. The profound analysis stage in this ethnography included repeating recognition of themes considering the dynamics that followed subsequent data collection and subsequent analysis. The contents of transcripts were reviewed and analyzed as data collection continued. This process permitted the amendment of the questions that were asked to enable more data gathering as fieldwork evolved. This stage took into account the supplemental data obtained at the previous and subsequent stages of fieldwork. It enabled the detecting of themes that required more precise analysis and the identification of dynamics in themes and concepts leading to the collection of supplementary data. These analyses gradually built up into a theory grounded in the data from which the themes and concepts emerged.
The themes and concepts revealed relationships and differences in themes according to the categories of research participants. A logbook about the themes was kept alongside the transcripts. With the logbook we started having an emerging story that revealed ideas, more insights, and explanations. The themes in the logbook were followed up in subsequent fieldwork activities to look for alternative explanations and evidence to support the ideas and interpretations. Further fieldwork permitted the verification of the themes and the identification of new themes and subthemes as well as evidence that might not support them. The reading of transcripts and comparing of themes and subthemes were done several times until a point where it made sense was reached and arguments were produced that addressed the research objective.
The analysis led to obtaining lay definition of the concept of diabetes and how the lay definitions influence the way people seek diabetes care. The definitions are grounded in people’s thoughts and behaviour as revealed by the ethnography. The shifts in meanings illustrate a multistage profile that provides the possibility of organising a pattern-matching mechanism that permits the display of the structure of quoted expressions and their actions into minute component subparts of actors involved in diabetes experiences and diabetes care in Cameroon.
This paper is a guide to practical and theoretical considerations for researchers new to qualitative data analysis. Data examples are given to illustrate decisions that were made when transcribing or assigning the task to others. Verbal and nonverbal interactions together shape communicative meaning [
The journey of studying the treatment of diabetes started with patients at the clinics and followed through to their homes and the wider community where family members and traditional healers emerged as key partners in providing support for diabetes care. The understanding of diabetes by each of these key partners and their role in seeking care is critically reviewed within the patient centred care model. Patient centered care is defined as an approach to “providing care that is respectful of and responsive to individual patient preferences, needs, and values and ensuring that patient values guide all clinical decisions” [
In seeking to understand how people with diabetes and their immediate families make sense of this condition, an ethnographic approach starts from the categories and distinctions made by those studied. A fundamental distinction here concerned aetiology as illustrated by the following quote below.
Natural causes are explained in terms of lifestyle. Sugar and sweet things in general seem to be one of the causes of diabetes that people highlight and understand, reflecting how the language of “high sugar level” (hyperglycaemia) in the clinic found its way into popular idioms. Most people with diabetes, irrespective of social status, often declared, “I know that eating a lot of sweet things causes diabetes. This one is natural.” This kind of causation places the “blame” or responsibility on the individual for excesses in eating and drinking, and some people with diabetes by and large realise that this is the implication. Alcohol, high fat food, physical inactivity, obesity, and overweight are seldom considered as lifestyle risk factors of developing diabetes but rather valued as indicators of affluence. Advice on diet and alcohol consumption is not seen as straightforward and easy to follow. As a person with diabetes participating in a death celebration puts it,
A second causal framework is the supernatural ancestors. The ancestral framework, cuts across many areas in Cameroon as ancestral punishment, by way of illness, has pride of place in people’s understanding of why illness strikes one person rather than another at any time. Ancestral interventions to “cause” diabetes are not always thought of as being attributable to recent oversights or human failings. When a person within a family develops diabetes, a diviner may trace the precipitating problem back to several generations, perhaps even a century or more. Most patients may ironically also interpret a family history of diabetes as a kind of family inheritance (created in this idiom by an unrealised and unresolved problem, dating several generations back). It is believed that ancestors can be appeased through rituals and a cure for diabetes obtained both traditional medicine and biomedicine. Taking treatment action needs meditating around the cause as revealed by a patient,
Witchcraft is the third widespread explanation for diabetes (among a long list of chronic health problems potentially attributable to the action of witches). Witchcraft is known as man-made diabetes. This might be represented in several ways:
Diabetes is rarely simply an individual matter, to be handled by the patient alone. It is essentially a family one but also extends to encompass the wider community as illustrated by the following fieldnote:
But of course all this could work the other way around, for there are instances where people with diabetes are unhappy with family reluctance to provide the appropriate diet, typically on grounds that the extra cost is too great. The consequence can easily be a feeling of a lack of care and a resulting mutual resentment. Ensuring that the person with diabetes remains healthy is usually seen as a shared family responsibility. But the ideal is not necessarily realised in practice, and there is often considerable tension between spouses and parents and children, particularly where one is felt to be undermining the efforts of the other through blame. Dietary and alcohol issues are proven to be even harder to deal with outside the home in public socialising and festive communal occasions. If diet provokes argument, alcohol does so even more as a diabetic explained in this quote, “My friends do not more want to go out with me because I drink Soda Water. They tell me to stay at home and be doing that.”
Sex also is quite commonly an area of considerable difficulty between most husbands and wives. Diabetes can easily compound these general pressures of urban social mores, as possible complications of the disease arise and as libido declines, as this quote explains.
There are two types of health care providers that we consider here: biomedical and traditional health care providers.
In theory, biomedical health care providers (BHCP) have the potential to educate diabetes patients about their disease. But there is little or no education tailored to particular individual needs on the grounds that too many people attend consultations each day, as exemplified by the quote below:
The more senior biomedical health care providers are socialized into a biomedical approach to health care, the more they commonly neglect those facets of the experience of illness that might be considered psychosocial and cultural. Such BHCP rarely seem to understand the cluster of reasons that may pull patients away from clinics and tend to base their understanding of patients’ actions only around clinical indicators and measurements made in the clinic setting. However, understandable this is, it serves to make their job of communicating with their patients about the many complications of diabetes more difficult, for figures and measurements are hard to set in context for most patients. The consequence is that more senior BHCP and patients have only a very partial grasp of each other’s idioms and frames of reference.
The more the gulf in understanding between clinic staff and patients is manifested or maintained, the greater the tendency is to passivity among those patients who attend. Yet staff may not take into account that patient passivity hides various states of mind. But health care providers term this as noncompliance to biomedical regimes, “Patients make the job difficult for us because they do not comply with their treatment.” Moreover, the financial costs of adhering to clinic advice are also a factor, adding to this resistance. A nurse declared, “at times they do not have the money to buy the medicines. But even those who have the money do not still comply.”
The contribution of traditional healers (TH), their claims, and their particular perspective, on diabetes are taken very seriously. TH in Yaounde are flexible in their approaches and strategies in the process of caring for people with diabetes. They have the ability to represent themselves in many forms. They become diviners, native doctors, traditional healers, traditional doctors, soothsayers, herbalists, and spiritual healers following the circumstances that they find themselves and the calibre of people that they come across. Besides using herbs, incantations, prayers, and animate and inanimate products in their practice to cure diabetes, they apply divination, an aspect believed to be out of the reach of lay people. Some, like the spiritual healers, add prayers to their practice as illustrated by the quote below.
It is difficult to keep questions of treatment entirely separate from discussion of causes, for the two are closely interwoven. Some of the aetiologies have already been discussed with patients above. TH are, after all, members of the same community, heirs to the same cultural values or currents of change. The following extract from an interview with a traditional healer amplifies the different perspectives and actions:
Some traditional healers use health facilities to confirm diagnosis as illustrated by the following quote:
Diagnosis at the hospital reveals to the patient that the signs and symptoms he or she is suffering from is diabetes. But what it fails to explain are where the diabetes has come from and why it has affected that person and not another. In some cases, traditional healers’ knowledge of some of the signs and symptoms helps them to diagnose that a person has diabetes. A traditional healer who refers a patient to a health facility for the confirmation of diagnosis does this only when he is very sure that his diagnosis would not fail. But there are situations where some traditional healers would say:
Many people interpret illness attacks in Cameroon in terms of weak and strong blood. If your blood is considered as weak, you are considered to be highly vulnerable to illness attacks of any nature. If your blood is considered “
Adapted integrated triangle of care for diabetes in Africa.
Figure
The difference between the adapted triangle of care derived from this study and that of WHO is that this adapted integrated triangle of care for diabetes includes traditional healers, traditional medicine, and traditional treatment in the treatment network while that of WHO excludes them.
In all countries, the lives of patients with chronic diseases extend far beyond the barriers of formal clinics and the reach of health care teams. Moreover, communities can provide services that complement and support the care provided in health care organisations bridging the service gap between the organised clinical care and the real world of patients. Little is reported about the requirements of implementing an adapted integrated patient centred care model in the African context (see Figure
De-Graft Aikins [
This paper explicitly looks precisely at the relationships and dynamics of a network of actors through participating in their lives. De-Graft Aikins [
De-Graft Aikins [
Patients’ knowledge and interactions with alternative therapeutic options indicate inadequacies in biomedical approach to diabetes care to fill the knowledge gap and keep the patients within their confines, trust, and confidence [
Patients and families seem to have high expectations of traditional medicine [
Behaviour change is difficult to initiate and sustain, and the reasons why efforts to promote healthy behaviour fail have been explored in this paper as has been scrutinised by others [
Interventions based on the patient centred model may enable physicians to intervene in a patient centred manner [
Though this study has used the ethnographic approach to study diabetes in Cameroon, the study limits itself to two sociocultural ecologies of Cameroon—the Forest and Grassfield ecologies. It is likely that these findings may not strictly apply to the other two ecologies (Coastal and Sahel). However, the findings are useful in informing policy makers on issues about cultural interpretation of illness and disease and the alternative perspectives of care for chronic diseases. To add more evidence to these findings a quantitative study design may help quantify these findings and relate them to treatment outcomes with different patterns of care.
This paper presents ethnography of diabetes, illustrating the role of different actors involved in treating diabetes. It ascertains the validity of the integrated clinic care model developed from the patient centred care model. The Integrated Chronic Care Model (ICCM) provides a flexible but comprehensive base on which to reconstruct health systems in Africa in accordance with local resources and demands. The ICCM stipulates that modern biomedical system needs to take into account the patients and community views to avert inherent frustrations. The triangle of care that I have adapted from the ICCM adds the aspect of traditional medicine and traditional healers as important partners to biomedicine in the care for diabetes. The fact that TH have started to acknowledge the efficacy of biomedicine over some aspects of TM and can provide clearance to patients to attend diabetes care units is indicative that they are capable of accepting weaknesses about their curing claims. Therefore, their role in preventing diabetes risk factors, diabetes, and diabetes complications should be acknowledged.
Acquired immune deficiency syndrome
Biomedical health care providers
Human immunodeficiency virus
Innovative care for chronic conditions
Traditional healers
Traditional medicine
World Health Organisation.
The author declares that there is no conflict of interests regarding the publication of this paper.
The author gratefully acknowledges the contribution of the research participants in making this study possible. The author would like to thank Professor Peter Phillimore and Professor Nigel Unwin for their guidance of this research and the continued mentoring that they provide. The author is grateful to the North-East Diabetes Trust, the International Diabetes Federation, World Diabetes Foundation, and Newcastle University which provided funding for the Ph.D. on which this paper is based. The author particularly thanks Professors George Alberti and Jean-Claude Mbanya and Mrs Hazel Glass for their inputs in enabling this research to go to a successful end.