The aim of the study was to explore beliefs about health and illness that might affect self-care practices and health-seeking behaviour in Zimbabwean men and women who have diabetes mellitus.

The Zimbabwean Health care delivery system has four levels: the primary, secondary, tertiary and quaternary levels (Table 1). Non-complicated diabetic patients can be managed at both the secondary and tertiary levels, while complicated cases are referred to the quaternary level (central hospital-based diabetes clinics) for further investigations, management of diabetes-related complications and continued diabetes education. Diabetic patients from municipal clinics and doctors’ private surgeries at the primary level in the urban areas are also referred to the same clinics. Subsequent follow-ups are done monthly until glycaemic controls are achieved, and then the patients are referred back to the lower levels for continued care [20]. The studied diabetes clinic is under the Department of Internal Medicine at a central hospital and run by a physician specialised in internal medicine, a group of general physicians and non-diabetes specialist nurses. When visiting the clinic blood and urine tests are taken as a preparation for waiting to be seen by the physician. While waiting for the results some health education is sometimes given by the nurses at the clinic but due to work overload this is not regularly organised. Patients are not self-monitoring blood glucose due to lack of equipment or financial resources.

Most health care services are provided by the govern-ment (65%) complemented by not-for-profit missionary/ church hospitals and private-for-profit institutions [20]. Currently, there is no National Health Insurance system in Zimbabwe. Public health care services are free for pensioners and the under-fives, while others either pay in cash or use private medical aid schemes.

Traditional healers and herbalists are also an integral part of the health care system, represented by the Zimbabwe National Traditional Healers Association [20].

Essential drugs and supplies have been greatly diminished, with stock availability reduced to 22-36% from 2008. In most cases, drugs to control diabetes were unavailable in most public hospitals and private pharmacies due to the economic crisis and hyperinflation from 2006 to 2009 [9]. Hence some patients either defaulted or bought drugs from nearby countries and from illegal traders on the city streets, while others consulted traditional healers, herbalists etc. for treatment.

Lack of drugs for treatment of diabetes increase the risk for complications and by that also morbidity and mortality [1]. The economic circumstances and reduced quality of care at most health facilities [9] might have forced some diabetic patients to engage in negative illness actions such as healer shopping (going to different care-givers) as they could not afford the high cost of drugs.

A qualitative descriptive study design was used. Data were collected through semi-structured interviews in order to give the respondents freedom to present their own subjective beliefs and to reach a deeper understanding of the topic being studied [21].

A purposeful sampling procedure was used. Persons diagnosed with diabetes mellitus visiting the diabetes clinic were offered, by a nurse (the principal investigator), the opportunity to participate. The inclusion criteria were: diagnosis with diabetes for one year or more, aged ≥18 years, no known psychiatric disorders and able to speak either English or Shona. The respondents were managed in an outpatient diabetes clinic at a central hospital in an urban area also managing persons referred from rural areas.

The sample included 21 participants, 11 females and 10 males, aged 44 (range 19-61) vs 52 years (range 22-65) in median (Table 2).

Most males were on oral agents while insulin treatment was more common amongst females. The majority of the participants were married. Females had a lower educational background than males and most were unemployed while most males were gainfully employed. (Table 2). The same pattern of socioeconomic circumstances was found in a previous report from Zimbabwe [20].

Each interview started with standardised questions on socio-demographic and medical variables. A thematic interview guide with open-ended questions including common health problems related to diabetes was then used in order to give respondents freedom to express their own views and obtain a broad range of information in an area that is scarcely studied [21, 22]. The aim was to describe and explore individual beliefs about health and illness and the influence on self-care and health-seeking behaviour. Themes investigated were: content of health; factors of importance for health, explanations and perceived consequences of diabetes, activities to restore health and care-seeking behaviours. The interview guide was developed based on previous investigations in Sweden [16, 23] and in Uganda [10] and peer-reviewed by general practitioners and nurses working in diabetes care. It was translated from English to Shona by the first author and a senior nurse in diabetes care and adapted to the Zimbabwean situation. A pilot-test was made with six persons (not included in the study), and further, minor changes were made to wording and meaning.

Interviews were carried out by a bilingual nurse (first author), who was not involved in the management of the participants. Eighteen interviews were conducted in Shona (according to the participants’ choices), and three in English, both of which are two of the three official languages in Zimbabwe [20]. The interviews were held in a secluded room outside the clinic and lasted about 1.5 hours. They were tape-recorded and transcribed verbatim. Transcripts in Shona were then translated into English.

The study was approved by the hospital Ethics Committee. It was done in accordance with the Helsinki Declaration [24] and with written informed consent from the participants.

Qualitative content analysis was used in order to discover and describe the variations in beliefs about health and illness [25]. Collection and analysis of data proceeded concurrently until no new information was added in the analysis. The tapes were listened to after the interviews and notes on themes and contradictions were taken as they emerged from comparisons of the respondents’ statements.

The transcripts were read thoroughly several times to achieve a sense of the whole. The texts were broken into smaller textual units where after, each meaning unit was coded and grouped together. Finally, the codes were compared for differences and similarities and sorted into subcategories and main categories (example see Table 3). The focus on the analyses was to be as open to as much variation as possible, and regularities, patterns and contradictions were searched for as the main analytical categories. The Lay Model of Illness Causation [19] and the model for health-care seeking behaviour by [26] were used as previously described [7]. Illness can be perceived as caused by factors related to the individual, natural, social and supernatural worlds [19] and health care sought from the popular (family members), folk (traditional healers, herbalists) and the professional sectors [26]. In order to increase the trustworthiness of the results, data analysis was done independently by two researchers (first and last authors) [25] and comparisons showed a high degree of agreement. Content of categories was also checked by a third researcher (second author), to confirm their relevance [25].

Health was described in terms of individual factors, ‘freedom from diseases’ and ‘feeling of well-being’. In order to feel well and experience good health, males (M) focused on compliance with diet, drugs and exercises while females (F) emphasised the importance of diet, drugs and personal and environmental hygiene.

… eat good food; exercise every morning and get treatment on time (M3).

… body-building foods, maintain body and environmental hygiene, follow doctors’ instructions about drugs (F1).

All participants acknowledged tangible instrumental support from family members. Provision of money (social factors) for food and drugs was necessary to most participants. Males focused on food provision and preparation and psychological support, in contrast to females who spoke of acceptance and understanding a sick person.

… my children should provide money for medications and transport to go for reviews. Wife to ensure I’m well fed and keep my clothes clean (M3).

… help me to live positively… help with different things … provision with money for drugs and food (M10).

… accept my condition … assist when I become sick (F11).

Females considered health professionals to be important for improved health as they assisted with information about diabetes and its management, while men mostly talked about assistance in health maintenance by monitoring of the disease, picking up potential problems and regulation of drugs. In both groups, some mentioned the need for emotional support of diabetics.

… give information … teach me the does and don’ts … give advice on food and giving of injections (F1).

… give advice on looking after myself, food preparation, taking of drug … regular blood sugar monitoring (M3).

As factors harmful to health, both males and females cited the strained economic situation, often leading to overworking (social factors):

… present harsh economic situation … leads to overworking, no time to rest (M4).

All respondents talked about the negative effects on health caused by their economy, which resulted in being unable to buy medications and/or food, or prescribed drugs which were perceived as expensive. Some said they had less frequent check-up visits while others discussed increased stress and associated diseases:

… situation now too hard … medication is scarce, everything is expensive…have reduced hospital visits (M10).

… life has become hard … take only one meal a day … food shortage… drugs scarce and very expensive (F5).

Females also added social problems like being ‘ill treated’ and ‘lack of respect’ by relatives. One man talked about harmful influence of smoking/excessive drinking, while others mentioned diseases like hypertension. Irrespective of gender, respondents reported the use of bodily symptoms as a way of checking whether health was improved or worsened; ‘passing lots of urine, feeling of being tired … headaches’ (M10).

Most males and females talked about compliance with drugs and recommended diet as the main measures to achieve good glycaemic control (individual/nature factors). A few females talked about regular monitoring of blood sugar levels in contrast to only one man. Avoidance of diabetes-related complications was the main reason for achieving good glycaemic control.

… attend monthly check-ups at the doctors’ rooms (M16).

… avoid too refined foods, take plenty of vegetables, white meat … stick to correct drugs (F6).

Measures to improve health when diagnosed with diabetes were mainly prayers (supernatural factors), different household remedies, e.g. Panadol, Aspirin, and herbal teas (natural factors).

… prayers work … herbal teas (e.g. zumbani), are good for colds, flu and stress (M19).

… Panadol for headaches and bodily pains (M7).

With reference to traditional ceremonies, some participants claimed they never used them while others claimed they ‘didn’t believe in’ them being important for health. Some were ‘prohibited by their churches’.

Most females reported contact with self-help groups like the Diabetes Association where they got ‘information about diabetes from experts’. Others had ‘heard about it but not joined’ while some males didn’t know about its existence and/or were not interested.

To maintain health and prevent complications, participants reported ‘individual’ and ‘social’ factors in terms of compliance with prescribed drugs and check-ups, avoiding and coping with stress. A female talked about adaptation in terms of working hard to earn a living:

… avoid foods that contain too much sugar … live a life that conforms to diabetes … attend regular check-ups, test blood sugar levels, take diabetes pills and injections on time (M10).

… find ways to deal with stress (F20).

… follow doctors/nurses’ advice about diabetic diet and drugs … work hard to earn a living (F2).

Most respondents, irrespective of gender, had experienced several symptoms when they fell ill with diabetes, such as passing lots of urine, feeling thirsty, losing weight, sweating and tiredness. With a few exceptions in the female group, none had suspected diabetes. Reactions in both groups included feelings of dying and helplessness. Males suspected sexually transmitted infections, HIV and AIDS, in contrast to a few females who suspected diabetes, malaria and witchcraft. A few participants thought it was natural. Most were seen by doctors/nurses at a clinic (professional sector) after consulting family members (popular sector). Some males consulted their mothers or wives, one female consulted her sister in-law while (popular sector) another visited a witch-doctor (folk sector).

… passing lots of urine, feeling thirsty, drinking lots of water … got worried … thought it was an STI … visited a doctor... later referred to a hospital (M4).

… loss of weight, excessive sweating and passing lots of urine at night…..ignored it initially … thought it was natural … later taken to the doctor (M10).

… excessive weight loss, drinking too much water … thought I had been bewitched. Firstly, I was taken to a ‘n’anga’ then to the doctor who later referred me to the hospital (F5).

… feeling tired, weight loss, passing lots of urine, dry mouth … got confused … feeling of dying … Sought help from sister in-law … was later taken to a doctor (F9).

Different causes of diabetes were stated, e.g. excessive starch intake, pancreatic problems, and antihypertensive drugs (Table 3). Males focused on ‘too much sugar intake’ through beer/alcohol, while females mainly discussed ‘heredity’ (individual factor), two claimed witchcraft (supernatural factor), and others were ‘not sure’. In relation to discussion of a list of potential causes of diabetes, most talked about heredity, obesity, inactivity, incorrect food, pancreatic diseases (individual factors), drugs such as diuretics (natural factors) and stress (social factor). Supernatural thoughts, e.g. punishment by God and witchcraft, were more often mentioned by females than by males.

Generally, participants knew about what happens in the body when having diabetes and about insulin function. More females than males knew the functions of the pancreas, although knowledge was limited. Females related the pancreas to insulin production, while males either related it to digestion of sugars or were not sure.

Most participants did not use any other name for diabetes besides ‘sugar disease; ‘diabetes’, while a few referred to it as a disease of the evil spirits (chivanhu, chirwere chokumusha kwangu), animal (mhuka) and cancer (nhuta).

The majority of the participants perceived diabetes to be lifelong while some stated ‘they didn’t know’ how long it would last. Most males were treated with oral agents while more females were on insulin. Participants saw provision of diabetes drugs as of paramount importance in the regulation of blood sugar levels.

Both females and males reported diabetes-related physical problems, e.g. backache and reduced/misty vision, however, more frequent in females. A few males talked about hypertension, coma (individual factors) and financial constraints (social factors), while some females discussed inability to perform daily activities and poor support from family members.

I developed high BP, backache and muscle pains … inability to work for my family (M19).

… passing too much urine, weight gain, reduced vision … inability to carry out daily chores at home (F17).

… my family is not supporting me and my children (F2).

Most participants expressed fears of diabetes. Males focused on death and getting into coma, while some did not express any fears. All except one female feared visual, kidney, and heart complications while others feared pregnancy-related death and infertility.

I am afraid of dying (M3).

… diabetes coma (M15).

… kidney damage, coma, painful cramps, misted vision (F21).

I fear infertility and dying from pregnancy-related complications (F5).

The majority of the participants except one female reported being informed about regular reviews of disease progress:

Monthly reviews to check on progress (M15).

Never received any advice (F5).

All participants except two men reported being given advice concerning the importance of monitoring blood glucose using bodily symptoms, conforming to correct diet and avoiding sudden changes of blood sugar levels so as to minimise complications:

I monitor at home … using bodily symptoms like tiredness, loss of appetite, funny feelings (M3).

When discussing dietary recommendations, all participants, except a few, cited the importance of reducing carbohydrates, increased intake of green vegetables, fruit and regular small meals and snacks. Two females got advice from a dietician. All females had been informed of the importance of regular exercise, but only a few males. Irrespective of gender, all participants had been advised about drug compliance: ‘I should take daily drugs on time … avoid taking unprescribed drugs’ (M7).

All participants claimed they followed the advice and one female participant stated: 'It helps me to maintain good health' (F8).

Common diabetes-related health problems, such as hyperglycaemia, repeated episodes of hypoglycaemia, gastrointestinal infection, common colds, pharyngitis, urinary infection, problems with the feet, spasm in the calf and hypertension, were more frequently reported in females than males. They also indicated difficulties in identifying the causes of the problems and stated they were unsure, while males often included an association with diabetes. Most participants had used self-care measures related to the individual sphere (food intake, rest, expectancy), often combined with nature (medications or herbs) and/or the supernatural sphere (prayers and/or holy water) to retain their health. Women exemplified a greater variation in self-care measures used and mentioned to a higher extent the use of herbs and supernatural measures (prayers and holy water) in comparison to males. When necessary, help was sought from the professional sector (doctors /or nurses) either at the local clinics or at hospitals.

When discussing the occurrence of wounds on the feet, a few had experienced wounds from minor injuries. Help had been sought from the professional sector at clinics/hospitals after self-care cleaning with salt and water.

I was cut by a sharp object … cleaned with salt and water at home (F5).

I got hurt while cutting firewood … cleaned the wound with salty water, bandaged it then sought treatment at the clinic (M19).