The presentation of the findings from the Tanzania study is organised into the three themes: (1) The structural context of biomedical and ethnomedical care; (2) health-seeking; and (3) ethnomedical health-seeking.
The structural context of biomedical and ethnomedical health care for people with diabetes in Tanzania
Recently, a network of diabetes clinics had been established throughout Tanzania which had provided approximately 100,000 people access to affordable diabetes treatment and health education. These diabetes clinics had been established by the Tanzania Diabetes Association in collaboration with the Tanzanian Ministry of Health (MoH) and other partners and were run by district, regional, and referral hospitals. Consultations were free of charge and clinical assessment procedures such as weight, blood pressure, and blood glucose level were heavily subsidised and cost 1000 Tsh. (1 USD at the time). The poorest patients paid nothing because of the national exemption waiver system, although in practice an evaluation showed that many of those who should be eligible for exemptions did not have access to it[
15], which was also our experience. In the past, before these diabetes clinics were established, diabetes services were provided at regional hospitals by staff with no specialist diabetes training. Specialist diabetes care was only available at five referral hospitals, run by a small number of consultant diabetes physicians and diabetes educators and hence access to diagnosis and treatment particularly for people in Tanzania's rural areas was extremely limited[
16]. At the time of the fieldwork, the Tanzanian Ministry of Health was formulating a national strategy on non-communicable and chronic diseases that once completed and implemented should further improve health interventions in this area (Interview MoH, Malene Krag Petersen, March 2008).
The biomedical treatment offered in Tanzania has improved, but there is still much to be done. The three public clinics at the district hospitals in Dar es Salaam that MK and MvD visited were full of patients waiting for their consultation, and across the diabetes clinics the increase in patients and a shortage of qualified personnel affected provision of timely and appropriate treatment[
16]. A nurse at the diabetes ambulatory at the National Hospital told us that the increase in patients meant that the frequency of appointments was reduced, lowering the quality of the treatment offered to patients (Interview February 2008). Furthermore, at the district hospitals insulin and other medicines had previously been available heavily subsidised, but at the time of inquiry Tanzania had had a nationwide shortage of insulin for more than two years (Interview MoH, Krag Petersen, March 2008). In consequence, many people had to buy the medicine at private pharmacies where it was still available, but at a high cost. This was the case for a female informant, Helen, 52. We spent a day accompanying Helen in need of buying insulin which was part of conducting participant observation in the local context of informants' daily lives.
Case 1. Helen's illness experiences
Helen left her home on the outskirts of Dar es Salaam in the morning and commuted on crowded dala-dala minibuses through the bumpy dirt roads to the city centre. During the bus rides she used her diabetes to gain a seat on the dala-dala or a better place in a queue. MvD and MK met Helen the first day of the fieldwork at one of the public diabetes clinics. We had arrived an hour before the clinic opened and the waiting area was already full of patients and the relatives accompanying them. We had been talking to a few patients with the assistance of our interpreter when Helen approached us and wanted to talk to us. One of the things she wished to express was that diabetes patients were overlooked compared to patients with HIV/AIDS who received free treatment and medication which she felt diabetics should receive as well. She pointed out the HIV/AIDS clinic which had recently been renovated and which was located across from the diabetes clinic. It was an impressive white building in comparison to the diabetes clinic which was in fact a container transformed into a clinic and located at the margins of the district hospital. Helen, we learned, was one of the patients at the district clinics who had managed to make use of the exemption waiver system mentioned above and did not pay for the medical check-ups. These illness actions; using her diabetes to gain a seat on the bus, a better place in the queue or managing to make use of the exemptions, made Helen appear resourceful. Yet, the day we accompanied her looking for insulin, she spent the whole day going from pharmacy to pharmacy to find insulin at an affordable price, at which she failed, and returned home empty handed. Going without insulin from time to time was nothing unusual to Helen which proved her difficult situation in spite of our impression of Helen as one of the more resourceful patients at the diabetes clinics.
It was our experience that many of the informants in the Tanzania study, like Helen, did not take their medicine regularly because they were unable to purchase the medicine. It was not only insulin that was unavailable at the hospital pharmacies. Also oral tablets were in shortage as a female informant explained (Rose, 54):
"Most of the time, the hospital pharmacy never has the tablets. Then I have to go to the private pharmacy and I mostly buy there since the store at the hospital doesn't have the medicine."
The shortage of subsidised insulin and oral medication was evident from the informants' accounts. Providing people access to insulin and oral medication as well as diabetes services is important in order for them to engage continuously in treatment and a female informant addressed the problem of gaining access to medication and treatment in a focus group discussion in this manner. She said (Zalika, 63):
"This problem is something that happens to poor people. For someone who's rich, it's not a problem."
In Tanzania the medical sector is pluralistic with biomedical health care systems and ethnomedical health clinics existing side by side and offering different
explanatory models, a term coined by Kleinman[
11], concerning the causes of diabetes and means of effective treatment. Ethnomedical health clinics of various ethnomedical approaches are dispersed all over the city of Dar es Salaam and many are placed next to or close by a biomedical clinic or hospital. MK and MvD interviewed a herbalist healer whose clinic was located further down the road from one of the district hospitals. He explained that he usually sent his patients to a diabetes clinic for diagnosis if he suspected the patient suffered from diabetes. Once the person had been diagnosed, the healer would initiate a herbal treatment which he claimed was able to cure any patient within one month. In our experience, for many people with diabetes such ethnomedical cure provided an attractive alternative to the prospect of lifelong biomedical treatment as will also be explored below.
Health-seeking in Dar es Salaam
The majority of the informants in the Tanzania study was not financially independent and people's self-care practices have to be seen as a collective praxis and not simply as an individual matter. The family members in the household, particularly the members of the nuclear family, provided care and treatment to the person afflicted by diabetes in terms of acquiring medicine, accompanying the person to health care services, knowledge sharing, and upholding a healthy diet since buying food was a collective matter especially in times of financial hardship. This mutual involvement and care performed by the family network supports existing anthropological knowledge on the function of the kinship system in which the kinship groupings are the most important social units for most people across the African region. The kinship system is characterised by social dependency and mutual obligations between kin and has a high degree of self-reliance in coping with disease and illnesses, as patterns of family treatment and care are deeply embedded within this wider kinship system[
17]. Although the family network was a fundamental support and an enabling factor for the actions that people took in relation to their illness, the interdependent character of the relations among relatives also had constraining consequences for these actions. Most of the informants' self-care practices were severely constrained because the needs of other relatives also had to be met. Many of the informants had to adjust their illness actions to the needs of other relatives which Haiba's illness experience illustrates.
Case 2. Haiba's illness experiences
Haiba, a woman aged 54, lived in a semi rural village on the very outskirts of Dar es Salaam in a household with her 12 family members. Five of these were her children. Her grown-up daughter was seriously ill from tuberculosis and required demanding daily care and medicine. Haiba was not alone in her care-giving role. Other females in the households assisted her, but being the mother of the ill daughter, Haiba had the primary care-giving responsibility for her daughter[
12] and since more than one person within the household was sick, it entailed choosing whom to help. Haiba chose her daughter's health to her own. This meant that she occasionally went without insulin in order to ensure that there was enough money within the household to buy her daughter's medicine.
Haiba's story illustrates how the needs of other relatives also had to be met which often compromised the needs of the person with diabetes. It further illustrates how the double burden of disease puts tremendous pressure on the remaining relatives[
17] and how prioritising was a continuous process that severely jeopardised the sustainability of the long-term treatment schemes necessary for survival.
In terms of financial support, our findings also show that there were limits to the obligations of kin. People were seeking out other ways of getting support as the mobilisation of resources had been extended to wider social networks. Many of the informants drew on support within their social network by approaching friends, colleagues or people within their local communities. A male informant explained to us in the following part of a home interview (Haamed, 35):
"Sometimes, when things get hard, I go to the business where I used to work and the colleagues they give me something. It is not fair every time to go to my brother."
Similar observations have been revealed in a recent study from western Kenya[
18]. This tells us of the need to revise the role of the family in general and include alternative support possibilities in our understanding of the conditions for illness actions in particular.
Furthermore, we argue that the diminished capacity due to hardship and deaths within the networks made the dynamics of illness action discontinuous as the intentions of the primary agent may be disrupted by other events. Such events may be the sudden death of the household provider or other relevant family members. The unpredictability and discontinuity of the life condition of the people with diabetes in the Tanzania study are illustrated through the story of one of the informants, Fatima, a woman of 32.
Case 3. Fatima's illness experiences
MK and MvD met Fatima during the first days of observation at one of the diabetes clinics. While most patients at the clinic were accompanied by a relative, Fatima was alone and sat to herself. She appeared introverted quite opposite to Helen who approached us at the clinic, wanting to talk. Fatima could not afford the medical check-up and only attended the doctor to obtain her prescriptions. Fatima lived with her younger sister in a small rented shack in an impoverished neighbourhood in Dar es Salaam. Fatima had moved to the city after her husband had committed suicide, her sister who once had employed Fatima in her restaurant had died from AIDS and her father had died from illness.
Our first home visit was characterised by a tense atmosphere as the hardship of life was obvious and constantly emphasised throughout the interview. When we contacted Fatima to arrange a follow-up interview, she had left Dar to visit the village where her mother lived with two of Fatima's children. We assumed that such a journey was an unaffordable expenditure for her since neither Fatima nor her sister had a job and they were both financially completely dependent on their older brother who lived in Dar. Unfortunately, we learned that Fatima's visit to her village was due to dire circumstances. A brother had passed away and she went to take part in the funeral. When we asked her how she managed to pay the bus fare, she explained that her relatives had rented a car to transport the body of the deceased. Realising that the brother was not living in the village but in Dar, we asked if the deceased brother was the one supporting her. Tragically, it was. While at work, robbers had attacked and shot him. With this information the interview took a whole different turn than planned since it was obvious that this sudden death in the family was bound to have a profound impact on Fatima's life condition. The brother had also helped support her children, and in fact the majority of Fatima's family members used to depend on the deceased brother. Apart from paying her food and housing, Fatima's brother had also been paying for one of the three medications she had been prescribed at the clinic and with his sudden death there did not seem to be any chances of her receiving more medical treatment.
We interviewed Fatima during a time of mourning which traditionally lasts 40 days after the funeral. When the mourning was over, her relatives were to decide what to do; not only about Fatima's situation, but also about the brother's wife and children and others. The outcome of their decisions was uncertain.
Fatima's situation was one of the most vulnerable among the informants and her story sums up the unpredictability and discontinuity of the life condition of the people in the Tanzania study. Fatima's vulnerable internal health conditions combined with the extreme vulnerability of external conditions disrupted all illness actions and left Fatima with few choices which may have had severe consequences for her future.
The lives of the informants were characterised by poverty, insecurity, uncertainty, and hence unpredictability. According to our observations, this suggests that the possibilities for health-seeking behaviour were often assessed on a day-to-day basis. This made illness action vulnerable to disruption and tactical in character. This furthermore opposed the long-term strategies promoted by health professionals for the treatment of diabetes and often favoured the short-term treatment found among herbalist healers, claiming to cure their illness. This will be explored below.
Ethnomedical health-seeking
It was also our experience that medical pluralism increased the range of therapeutic choice and complicated health-seeking behaviour which Hardon et al. also have argued[
19]. From interviews with health care professionals at the public diabetes clinics, MK and MvD were told that most or many of their patients had tried, or took, some sort of traditional therapy. Many of the informants in the Tanzania study had some contact, directly or indirectly, with herbalist treatment concomitant with attending treatment at biomedical diabetes clinics. Several also confirmed that they had interrupted their biomedical treatment in order to follow an ethnomedical treatment. From interviews, we found that the primary reason for using ethnomedicine was that it was less expensive and thus more affordable compared with the price of medicine at private (biomedical) pharmacies. Many informants had not attended a herbal healer, but their relatives had bought the herbal medicines for them. This was the case of Rose, 54:
"The first time I tried to use some traditional medicine it was my sister who brought it to me. Then, when I went to church there was a relative of our pastor from the church who knew of a medicine that cures diabetes, but again it didn't help me anything either. Now, I'm using another kind of medicine which I don't know where it comes from, but it is packed in a yellow plastic bottle and it's written that it contains aloe vera. When I take this last medicine, I don't feel so tired as before and so it reduces the tiresome that I normally have. It's my daughter who buys this one for me at her work."
Another reason for using ethnomedicine was that herbalist healers, or others selling ethnomedicine, claimed it would cure diabetes. This was also the motivation for Rose to continue experimenting with herbal medicines which she also expressed when we asked her what advice she had been given concerning the herbal medicine she was taking. She said:
"I was told that when I take this, it will help curing many diseases which are in the body. You have to take six plastic bottles of this size in order to be completely okay. It really helps!"
In some cases beliefs in herbal medicine had a profound impact on the treatment of the person with diabetes, but we also found that informants merely saw the herbal medicine as an opportunity which they had to try out. Rose's experience illustrates this as she kept experimenting with new ethnomedicines, hoping to be cured, while still taking the oral tablets prescribed by the doctor at the diabetes clinic at one of the district hospitals. However, these tablets were not taken uninterrupted:
"I usually take them as required by the dose and when they finish I beg for assistance of money from relatives and friends. But sometimes it might happen that I stay for two three days without the medicine until I collect enough money to buy the medicine."
Another informant, Hasani, 35, had also been advised by a relative to try herbal medicine. He had hoped it would improve his condition as he was dissatisfied with the oral tablets he had been prescribed by a doctor at a private diabetes clinic he attended. Hasani had migrated to Dar es Salaam when he was 16 and had lived for many years with an uncle who had recently passed away due to diabetes. It was this uncle who had advised Hasani to try herbal medicine. This passage about Hasani's experience with ethno-medicine is from our second interview with him:
"My uncle, who was the one who received me when I came to Dar and employed me in his grocery, he was also the one I told you last time that he advised me to use the aloe vera herbs when he discovered that I had the problem. He had the aloe vera plants in his farm and so he advised me to use it. I had started taking tablets, but the tablets gave me headaches. I was given tablets to use for one month and my uncle advised me that I finished using the one month tablets and switched and tried to use herbs like aloe vera and some other herbs whose names I don't remember. With the tablets my weight had continued to fall until the point I reached 54 kgs. So when I started using the herbs, my weight begun to re-gain back until it reached 70 kgs. I then stopped taking the medicine and stayed for almost three months without taking any medicine at all. Then I started feeling ill again with the same symptoms that I felt at the beginning of the problem. When I decided to re-use the Aloe Vera the doctor told me that this aloe vera may be poisonous and could result in kidney failure. That was when I decided to go to the district hospital to open my first clinic card."
Ethnomedical treatment often relieves people short-term; physically by easing the symptoms and psychologically by imposing hope of a cure, which Hasani's experiences also show. Unlike Rose, Hasani stopped taking the herbal medicine once he was told it could be poisonous. The tactical character of illness action made ethnomedical treatment seem even more favourable to biomedical treatment and when placed in a position of financial hardship, the long-term future was blurred to most of the informants and unpredictable which led them to make decisions concerning their health based on immediate opportunities and obstacles.