Study setting

Located in the southern region of Malawi, the Zomba district has a substantial HIV/AIDS prevalence, affecting nearly one in ten men and one in five women [50]. In addition, the prevalence of hypertension and/or diabetes in HIV clinics in Zomba is 26.6% [15]. The provision of hypertension and diabetes care follows the clinical guidelines set by the Malawi Ministry of Health [51]. Drugs for HIV are generally free and available, whereas medications for diabetes and hypertension are free only when in-stock at government-supported clinics.

Study sample and recruitment

Fifty individuals comprising 25 couples were recruited from three facilities that serve approximately 19,000 persons living with HIV, and represent a diverse range of geographic and social contexts. To be eligible, participants had to be 18 years or older and married or cohabiting for at least 6 months. Each couple had to include one partner (referred to as the “index patient” for clarity) who was diagnosed with HIV and either hypertension or diabetes and who had disclosed their HIV and CMD status to their partner. Recruitment of index patients took place during their HIV clinic appointments, with research assistants announcing the study during the health information sessions in the waiting rooms. Research assistants, working in male-female pairs, approached potential participants, screened for eligibility, and if eligible, provided information cards to give to the primary partner. Research assistants would later contact and speak privately to the primary partner on the phone to assess their eligibility and willingness to take part in the study. Some diabetes and hypertension patients were screened and referred by the nurses in accordance with the clinical guidelines for CMD [51]. Partners who collected the index patients’ HIV medications were also approached and given information cards about the study, and then research assistants later contacted the index patient on the phone to assess eligibility. Once both partners were deemed eligible and willing to participate, research assistants scheduled the first appointment for the couple at the HIV clinic. During the appointment, the couple’s eligibility was confirmed, both partners gave written informed consent in separate locations, and an interview and short survey were conducted. Participants were reachable by phone. The research team adhered to established recruitment and follow-up procedures, making three contact attempts at different times and on different days to account for situations where phones might be out of network or uncharged.

Study procedures and data collection

Semi-structured interviews were conducted in Chichewa by trained interviewers who were matched to the gender of the participants. Partners were interviewed separately in private areas to ensure confidentiality and independent responses, and to minimize bias. The interviews lasted between 60–90 minutes and were tailored to the CMD of the index patient, either hypertension or diabetes. If the patient had both conditions, questions were focused on diabetes due to its lower prevalence. The partner version of the guide focused questions on the index patient’s conditions. Both partners were asked questions according to domains of Kleinman’s theory of explanatory models [18] including symptomology, etiology, prevention, and disease management, as well as beliefs about the intersection of HIV and CMD. Examples of questions asked were “How do people get hypertension or diabetes?”, “What are the symptoms of hypertension or diabetes?”, “What can people do to prevent getting hypertension or diabetes?”, “What are some medicines that are used to treat hypertension or diabetes?”, and “What other ways can people manage their hypertension or diabetes besides medications?” To assess how participants compared and contrasted CMD with HIV, we asked: “How would you describe the relationship between HIV and hypertension (or diabetes)? Does one cause the other? Does one disease affect the other?”

Data analysis

The interviews were audio-recorded, transcribed, and translated from Chichewa to English by research assistants who are fluent in both languages. We conducted an individual-level analysis and dyadic-level analysis using a framework analysis approach, which allows for comparative analysis of themes and cases by utilizing data matrices [52]. We read and reviewed the interview transcript pairs, created summary tables, and organized the data by couples and by the larger categories from Kleinman [18] that were derived from the interview guide (e.g., “beliefs about symptoms”, “beliefs about treatment”). We compared the responses within and between couples by including raw quotes from each partner and then summarizing all the participants’ responses by category and noting any agreements or disagreements between partners (e.g., wife believes diet causes hypertension; husband believes the causes are diet and stress). Within these categories, the main themes were selected based on the frequency in which they came up and by their intensity (i.e., not necessarily common, but important or unexpected) [53]. We also coded for the broad types of explanatory models that emerged inductively from the data as other researchers have done when applying Kleinman’s model [54].

The research team, which included representation from Malawi and other African settings, met regularly to discuss the findings using the summary tables and the themes that emerged.

Ethics approval

This study received ethical approval from the National Health Science Research Committee in Malawi and the Human Research Protection Program at the University of California San Francisco. All participants provided written consent.

Sample characteristics

Participants were married or cohabitating couples with a mean age of 51 years (range: 37–66) and mean relationship duration of 20.8 years (range: 4–40). Most participants were Christian (88%), and a minority were Muslim (12%). Participants were recruited from a district hospital clinic (56%), a rural private community hospital (36%), and a peri-urban public facility (8%). The majority (80%) of participants were living with HIV, and of those, all were on ART. Fifty-seven percent had hypertension and 85% of those were on antihypertensive medications. Fourteen percent had diabetes and of those, 86% were on oral medications for diabetes (none were on insulin). The characteristics of the study participants are displayed in Table 1.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Demographics and disease status of participants.

https://doi.org/10.1371/journal.pone.0296473.t001

Overarching explanatory models

In their narratives, participants described the etiology, symptoms, disease course, prevention, treatment, and disease management of CMD. Table 2 outlines the common themes that emerged under each category with representative quotes from couples. Based on the themes that emerged in the data, we grouped explanatory models into two main categories: biomedical and traditional/cultural models. Biomedical explanatory models consisted of beliefs around physical health (e.g., lifestyle factors, prescribed medications) and mental health (e.g., stress, depression). Although other studies have considered stress or mental health as alternatives to biomedicine, we consider that beliefs about stress as a cause of CMD are consistent with biomedical research linking stress to CMD and cardiovascular disease [55–60]. We categorized beliefs as falling under a traditional/cultural explanatory model when grounded in religious beliefs and beliefs in natural medicine (e.g., herbs, traditional medicine) as a form of healing. Participants sometimes subscribed to a combination of explanatory models, both biomedical and traditional, sometimes giving more weight to one model over the other.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Emerging themes and supportive quotes from couples living with cardiometabolic disorders and HIV based on Kleinman’s theory of explanatory models.

https://doi.org/10.1371/journal.pone.0296473.t002

Etiology beliefs

More than half of the couples with hypertension adopted a biomedical explanatory model with a focus on mental health. They attributed their illness to “thinking too much” due to stress, which was described vaguely as “depression”, “the heart is not calm”, or feeling “restless” (see Couple #1 in Table 2). Some participants described stressful situations related to interpersonal relationships as causes for hypertension such as “the marriage is not going well”, “quarreling with a friend”, “someone recently died”, “disappointing each other”, and “living in an environment that is not peaceful”. Most of the participants who believed that stress was a cause of hypertension were index patients, but 23% of partners also mentioned stress. Participants seemed to come to this causation belief through personal experiences with people they knew to be stressed, or what they heard in their community rather than from healthcare providers. Of note, stress was not mentioned as a cause of diabetes.

In addition to a mental health model, index patients and their partners also adopted models focused on physical health and attributed hypertension to various dietary factors, such as high salt intake, excess fat, and sugar, certain meats, certain vegetables, cooking oil, pesticides on food, sour foods, and tea. In some cases, couples had more nuanced views on the type of diet that caused hypertension (such as Couple #2 in Table 2). A few participants attributed hypertension to aging. Regarding diabetes, commonly mentioned causes were the consumption of sugary drinks or food, including tea, followed by genetics and aging. A few participants said that they did not know what caused diabetes and one participant thought that viruses could cause diabetes. Physical inactivity and obesity were not identified as causes of hypertension, and only one participant mentioned that physical inactivity can cause diabetes.

HIV was not discussed as a potential cause of hypertension or diabetes, although a few participants suggested a possible link through the immune impact of HIV or how the stress of living with HIV over many years could lead to the development of hypertension. One partner believed that ART directly caused her husband’s hypertension because, in her view, he was diagnosed with hypertension once he started taking HIV medication; however, her husband did not mention the same belief.

Symptoms

Couples reported a varied range of physical and mental symptoms of hypertension and diabetes, with only one person stating that hypertension can be asymptomatic. Hypertension symptoms were based on experiences as patients, observations as partners, or what participants had heard from others. One-third reported a fast heart rate (“heart racing”) as a symptom and a few others described “heart pain” or “heart pricking” as symptoms. Other reported signs and symptoms varied widely among participants and included fever, sweating, headaches, body pains, difficulty breathing or “panting,” poor vision, “seeing darkness”, dizziness, weight gain, numbness, and loss of appetite (see Couple #4 in Table 2). Stroke was also mentioned alongside signs and symptoms as a sequela of hypertension. Secondary to the emphasis on physical signs and symptoms, some participants and partners also believed in mental health-related symptoms, stating that being easily stressed, short-tempered (see Couple #4 in Table 2), or “thinking a lot” was indicative of hypertension. Hypertension symptoms were sometimes compared with those of HIV, for example, one female patient asserted that both illnesses share similar symptoms. She seemed to equate the stress response of an HIV diagnosis to a similar set of symptoms for hypertension, stating: “The symptoms are the same because when you are diagnosed with HIV…you feel your heart racing, the body gets hot, the head hurts. Similarly, when you are suffering from hypertension your head hurts, your body feels hot, and the heart also races.”

For diabetes, most patients and their partners mentioned frequent urination, weakness, and vision changes as the main symptoms (see Couple #5 in Table 2). Other signs and symptoms included weight loss, drinking a lot of water, wrinkles on the skin, fainting, and other skin-related symptoms such as “blisters in the vagina” and poor wound healing.

Disease course and severity: Comparing HIV and CMD

Around two-thirds of couples agreed that the index patient’s hypertension or diabetes was more serious than their HIV (see Couple #3 in Table 2); the remainder of couples had discrepant views, including a few couples with a partner who was unconcerned about each condition or worried equally about both. For most participants, the threat of hypertension was viewed as more concerning than that of HIV, with the general belief that HIV was more manageable, but hypertension could cause sudden death. There were beliefs that people with HIV generally outlive those with hypertension or diabetes, who could suddenly collapse without any signs.

Other explanations for the severity of hypertension had to do with the belief that missing pills for hypertension was deadly: “With AIDS, even if they can skip medication for some days, they can still live. While hypertension, if you skip medication for some days… you are likely to be in danger.” Similar sentiments were noted about diabetes in reference to HIV, with participants suggesting that when sugar levels are too high, a person can die unexpectedly (“this disease comes with force”) or become very weak and then fall unconscious. Most people noted that HIV medications were reliable, and people rarely died if taking ART; however, medications for hypertension and diabetes were believed to be insufficient for controlling the disease or preventing death, and thus, could not always be relied on: “Hypertension is the one that’s really scary. …what’s happening in most cases is that they are taking medication but it’s not helping them. Even those taking medication are still dying.”

Hypertension and diabetes were also understood as being harder to manage due to symptoms waxing and waning over time, and because disease control depended on more than just taking medication, as with HIV. People expressed frustration that hypertension and diabetes were difficult to manage because of factors beyond their control, such as exposure to stress and being unable to afford the recommended foods, medications, and devices necessary to monitor blood pressure and sugar levels. Unstable access to CMD medications is a major challenge in settings like Malawi, with structural barriers and clinical-level barriers such as medication and staff shortages, clinic wait time, as well as transportation costs to and from the clinics [34]. The uncertainty of hypertension and diabetes was also overwhelming for partners, some of whom expressed worries about taking care of the children if a spouse were to die or caring for a disabled spouse after a stroke.

Prevention beliefs

Most couples believed that changes in diet and physical activity could prevent hypertension and diabetes. Salt reduction was noted as a preventive measure, and some participants suggested sugar reduction, even for hypertension. Participants described how sugar or salt can increase one’s heart rate, which was believed to cause “BP” (meaning hypertension). Exercise, like jogging or running, smoking cessation, and alcohol reduction were mentioned as ways to prevent hypertension and diabetes (see Couple #6 in Table 2). For diabetes prevention, the majority of participants stated that dietary changes, exercise, and medications were strategies for prevention. Some participants suggested avoiding certain vegetables, certain meats, cooking oils, rice, other starchy foods, tea, and nuts, and reducing sugar intake (see Couple #7 in Table 2). Similar to causation beliefs, stress reduction was frequently mentioned as a method to prevent hypertension, with suggestions like "avoiding thinking too much", “speaking calmly to a partner”, and "living peacefully" (see Couple #6 in Table 2). In a few cases, traditional explanatory models were invoked by a few participants, with praying being noted as a prevention strategy and some believing that God could keep a person healthy and free of CMD. One female participant with HIV and hypertension believed that condoms would prevent hypertension, although she was the only participant to link sexual risk with CMD.

In some couples, partners had different views on what they believed was most important for prevention. For example, one female participant with hypertension mentioned that medications could prevent hypertension, whereas her husband without CMD said: “the individual has to avoid thinking too much…they should remove all the stress they have and if they do, things will go well.”

Treatment beliefs

All couples emphasized the importance of taking medications as prescribed for hypertension management, despite some concern around taking them “for their entire life”. While a minority of participants believed medications could cure CMD, most understood they only controlled the disease (see Couple #8 in Table 2). Lifestyle modifications, such as dietary changes and exercise, were also emphasized for CMD management. Some suggested reducing oils, meat, and salt for hypertension management. Diabetes management strategies mentioned by participants included dietary changes such as sugar reduction and medication (see Couple #9 in Table 2).

Although less commonly mentioned in comparison to the biomedical approaches around physical health, some participants mentioned the importance of stress reduction, self-acceptance of the diagnosis, and faith in God for hypertension management. Participants and their partners commonly shared the belief that herbs, such as hibiscus, moringa, lemongrass, ginger, garlic, and onions, could be used for CMD management. Some had personal experience using these herbs to manage their CMD, and while others had only heard of traditional medicines like mvunguti (derived from the tree species Kigelia africana) for CMD management. Most participants had not consulted traditional healers due to conflicts with religious beliefs or perceptions that traditional healers were unsuitable for managing HIV and CMD, with some having done so, but then stopped after receiving their CMD diagnoses at healthcare facilities.

In most couples, partners were often congruent on general beliefs about medical and lifestyle management (with some nuanced differences such as agreement on diet type). However, there were some couples who seemed to diverge in the type of explanatory model invoked. For example, the husband of a wife living with hypertension believed that spirituality was more important than medications: “It’s not difficult to get cured and does not depend on medications because hypertension does not have medications (a cure). A person can take medications all his life, but if you would take God and put him close”, he also noted how he preferred that she take herbs that would cure her condition and “put an end to it”. His wife, the index patient, emphasized the importance of following the advice of the health providers, admitted that she had not tried herbs yet, and did not mention spirituality. In another case of a male participant living with hypertension, the wife stated: “I feel like the cure for this disease is that first, you should accept it…when you accept it and believe that this thing (hypertension) is not going to be with me, it is possible…the heavenly one (God) is the one who has all the possibility of helping you…I tell my husband that you should ask God to take this away from you”. Her husband, the index patient, believed in taking prescribed medication and following medical advice on lifestyle changes, indicating a discrepancy between partners’ explanatory models for treatment.