Adults' Experiences with Type 2 Diabetes in Rural Guatemala: A Qualitative Study

Abstract

Background. Given that the diabetes burden is rising rapidly in low- and middle-income countries, it is critical to understand perspectives of people living with diabetes in these settings. This qualitative study examines perceptions of causality and treatment among adults with type 2 diabetes in rural Indigenous communities of Guatemala. Methods. We conducted semi-structured interviews with 29 people living with type 2 diabetes from a population survey in two majority Indigenous Guatemalan towns. Data were coded inductively and themes were elucidated by consensus. Results. Participants emphasized strong emotional experiences and stress as perceived causes of diabetes, as well as diet and heritability. All participants took biopharmaceutical treatments, but many also endorsed diet, exercise, herbal remedies, and naturopathic products as important remedies. Conclusion. Perspectives from people living with diabetes in two rural Indigenous towns in Guatemala differ from the biomedical model of disease and have implications for clinical practice and program development.

More than 80% of the global burden of diabetes occurs in low- and middle-income countries, where limited infrastructure and capacity for managing chronic illness lead to disproportionate morbidity and mortality.^1,2 Guatemala is the largest country in Central America, and it is experiencing a rapidly increasing diabetes epidemic. It is estimated that the adult diabetes prevalence in Guatemala rose from 5% to 10% from 1980 to 2014.³

An important feature of the demographic structure of Guatemala is that Indigenous Maya constitute most of the population in rural and agricultural areas of the country. Due to the legacies of a protracted civil war, structural racism, and minimal investment in rural community infrastructure, this population suffers disproportionately from poverty, food insecurity, and preventable illnesses such as malnutrition.^4,5 Chronic non-communicable illnesses—such as diabetes—are now also an emerging problem in rural Indigenous Guatemalan communities; indeed, the estimated overall prevalence of diabetes of 12% in Indigenous Guatemalan communities is higher than the national average.^6,7

Given this growing diabetes burden, it is critical to understand perspectives and experiences of people living with diabetes in rural and Indigenous Guatemalan communities on disease causality, treatment modalities, and health-seeking behaviors. Prior qualitative research has shown that the high cost of biomedicines and foods are major elements of individuals' experiences living with diabetes.⁸ The use of traditional ethnomedical practices and plant-based remedies is common, and this is in part directly due to the prohibitive cost of many biomedical remedies.⁹ In addition, complementary alternative medicine has soared in popularity in Guatemala, where practitioners often make claims of being able to cure diabetes to attract customers.¹⁰ Finally, multiple researchers have documented how discrimination within health care as well as communication challenges arising from linguistic and cultural differences between Indigenous patients and non-Indigenous providers generate reluctance to seek biomedical care.^7,8,11

Importantly, prior qualitative research on the experiences of Indigenous people living with diabetes in Guatemala has relied on sampling frames drawn from clinic- or center-based panels of individuals with diabetes. Given the difficulty accessing diabetes care in Guatemala, these settings are not representative of the larger community-based population of individuals with diabetes. To address this issue, in this paper we take advantage of a recent population-representative, cross-sectional chronic disease survey in a rural Indigenous municipality. We conducted qualitative interviews with individuals identified during this survey as having diabetes, exploring their perspectives on diabetes causality, treatment, and curability.

Methods

Study context, sampling, and recruitment

This qualitative study was conducted with adult participants (≥18 years) recruited from a population-based survey assessing incidence and risk factors for chronic kidney disease in two majority Indigenous Guatemalan towns.¹² The first town is in the central highlands (Tecpán, Chimaltenango) and has strong agricultural roots and foundations in commerce and trade due to its location on a major international roadway. The central area of Tecpán is increasingly urbanizing, but it retains a large rural population base in surrounding villages.¹³ The second town is on the southwestern coast (San Antonio Suchitepéquez, Suchitepéquez) and is dedicated largely to sugar cane and coffee production.¹⁴ Despite differences in geography and economy, both towns are characterized by rapidly progressive nutrition transitions, decreasing physical activity, and increased consumption of highly-processed foods, all of which are associated with increasing diabetes prevalence.¹⁵ In both towns, Indigenous racial/ethnic identity, which has historically been defined by the speaking of Indigenous languages, wearing of traditional woven clothing, and subsistence agriculture, has been evolving over the last several decades in the face of structural inequalities, economic pressures, and globalization.^16–18 As a result, while both towns have historically been Indigenous communities, the way individuals within them self-identify varies considerably and includes an increasingly large category of individuals who identify as non-Indigenous (Ladino) but with Indigenous ancestry.^16,19 In our prior ethnographic work, some report that they identify as "poor" before any other category, reflecting the way that poverty and rurality intersect with Indigeneity, together giving rise to social marginalization, limited health care access, and poor health outcomes.¹⁹

The work was conducted in collaboration with Maya Health Alliance, a primary health care organization with a clinical presence in both communities. From June 2018 to February 2019, the population survey used a representative mapping technique to recruit 807 non-pregnant adults from 533 randomly sampled households as fully detailed elsewhere.¹² As part of assessment for risk factors for chronic kidney disease, survey participants were asked questions on prior medical conditions and had blood samples drawn for laboratory testing. We defined diabetes among people with a prior self-reported diabetes diagnosis or with hemoglobin A1c (HbA1c) ≥6.5%, which is in accordance with World Health Organization guidelines for classifying diabetes.²⁰ These procedures yielded a total population-representative sample of 117 individuals with diabetes.

When laboratory results became available, population study staff performed a follow-up visit to inform participants of their HbA1c result and refer individuals with diabetes to local health centers as needed. Population study staff referred participants with diabetes interested in participating in an interview to qualitative study team members. The qualitative study team approached all 42 participants who had enrolled in the population study from June 2018 to February 2019 who had an HbA1c ≥6.5% and reported a previous diagnosis of diabetes. The team used both criteria (elevated HbA1c and prior diagnosis) as a conservative measure, given the frequency of erroneous self-reported diabetes diagnoses in the study communities.⁶ Participants were approached from November 2018 to June 2019 in consecutive order. Participants were not offered an incentive or compensation for participation in the interview. Twenty-nine agreed to participate, 10 declined, and three were unable to find time to schedule an interview.

Data collection

Twenty-nine semi-structured interviews were conducted exploring participants' perceptions of causality, treatment, and curability/chronicity of diabetes. We developed a novel interview guide in these thematic areas, which were of particular interest based on our prior published work in Guatemala.^8,10 Our interview guide included prompts about participants' lived experiences relating to the chosen thematic areas. A full interview guide is provided at: https://doi.org/10.7910/DVN/H7ZOFB. Interviews occurred four to six months after participants enrolled in the population study, lasted 20–60 minutes, and occurred in participants' homes. Interviews were conducted in each participant's primary language, either Spanish or Kaqchikel Maya, with the support of an experienced Kaqchikel interpreter. Participants expressed a strong preference not to record interviews, and so detailed interview notes serve as the basis for this analysis. Where direct quotations are reported, a second researcher was present at the interview and transcribed responses and quotations at the time. Interviewers reviewed the accuracy of all notes and transcribed quotations immediately following the interview.

Qualitative data analysis

Researchers developed a codebook through review of the first 12 interviews using an inductive strategy. Two study team members coded each interview. Coding was reviewed by a third team member. Consensus discussions resolved discrepancies in coding and fostered theme elucidation. Saturation was achieved at 20 interviews, as no new codes or themes subsequently arose, suggesting an adequate sample size. Subsequently, themes were compared across communities and analyzed for similarities and differences across sites. We did not compare themes based on other demographic characteristics such as sex, age, or racial/ethnic identity for several reasons. First, available project resources were sufficient for elicitation of general understandings of diabetes causation, treatment, and chronicity/curability but not for the considerably larger sample of individuals that would have been required to compare across all demographic characteristics. Second, as described below, most participants were women, limiting ability to compare themes by participants' gender identity. Third, the majority of individuals were middle-aged, limiting ability to compare themes across age groups. Fourth, based on the overall larger project design, participants gave self-reports of their ethnic identity as Indigenous or of mixed descent (Ladino), but as described above, local understandings of Indigeneity are multidimensional and do not map neatly onto these two broad categories. The software NVivo Pro, Version 11.4.1 for Windows was used for qualitative analysis. The codebook and all coded data and frequencies for community comparisons are provided at https://doi.org/10.7910/DVN/H7ZOFB.

Quantitative data analysis

Comparisons of sociodemographic characteristics between the qualitative sample frame and the larger population sample were conducted using survey-weighted data elements which have been previously described.^12,21 Stata 17 (StataCorp. 2021. Stata Statistical Software: Release 17. College Station, TX: Stata-Corp LLC) was used for analyses, and p-values are generated from Student's t-tests for continuous variables and chi-squared tests for categorical variables.

Researcher characteristics

Our research team considered the impact of our social values and identities on the research. Team researchers were from both Guatemala and the United States, and all had research experience in lay conceptualization of chronic diseases. The team aimed to mitigate our influence on the qualitative data by ensuring that interviewers were not also involved in collecting data or samples for the larger population study. Furthermore, U.S.-identifying researchers conducted interviews collaboratively with an Indigenous colleague or an Indigenous language interpreter.

Institutional context and ethics

The Institutional Review Boards of Brigham and Women's Hospital, Maya Health Alliance, and the Institute of Nutrition of Central America and Panama approved this study. Participants provided written consent for the population study and spoken consent for qualitative interviews. This manuscript was developed using the Standards for Reporting Qualitative Research guidelines.²²

Results

Basic sociodemographic characteristics of the 29 interviewed individuals with previously diagnosed type 2 diabetes and an elevated hemoglobin A1C are given in Table 1. Sixteen participants were from Suchitepéquez and 13 were from Tecpán. Overall, the

Table 1.

SOCIODEMOGRAPHIC CHARACTERISTICS OF THE QUALITATIVE INTERVIEW SAMPLE OF INDIVIDUALS WITH TYPE 2 DIABETES COMPARED TO THE OVERALL POPULATION REPRESENTATIVE SURVEY IN TWO RURAL GUATEMALAN TOWNS

qualitative sample did not significantly differ from the overall population-based samples on most characteristics, except for higher hemoglobin A1C values (Table 1). The sex distribution in the overall survey sample was skewed by frequent out-migration and higher rates of survey non-response among men.¹² Although many participants self identified as Indigenous (58.6%), all but one preferred to conduct the interview in Spanish. Given significant under-treatment of individuals living with type 2 diabetes in the region, mean ± standard deviation hemoglobin A1C values were greatly elevated (10.6% ± 2.39).^{6,10,12,21–22}

Qualitative themes from participant interviews are organized below by the three major interview focus areas (causality, treatment, and curability/chronicity). In addition to the narrative description with incorporated select quotations, additional sample quotations are given in Box 1. As detailed below, similar themes were observed across both communities except for the theme of exercise as a treatment for diabetes.

Perceived causes of diabetes

Participants cited three major causes of diabetes: strong emotional experiences, diet, and inheritance.

Emotion

Most participants described strong emotional experiences as provoking diabetes (79%, n=23). In both sites, most participants endorsed emotional causation of diabetes (69% Suchitepéquez, 92% Tecpán). Interviewees perceived their own diabetes as precipitated by fear, anger, worry, and grief related to traumatic events, such as robberies, physical assaults, and experiences of the armed conflict in Guatemala. Some reported developing symptoms immediately after experiencing strong sudden emotions over intra-familial conflicts, dealing with alcoholism among family members, and unexpected deaths of family members. As one participant stated:

Well, my husband was a drunk. He came to the house, and he hit my daughter. I thought she died, oh, my daughter … I passed out. When I got up, people gave me water with sugar. I was scared. They gave me a cup of coconut water. That is when I myself felt that this disease started. I didn't even sleep anymore, because I was urinating so much.

(ID18)

Another participant attributed her diabetes to severe fright after being physically assaulted by thieves while she was trying to collect firewood: "I say from the fright … I ended up with [diabetes] and I ended up with anemia. The disease comes from a fright, from a sadness, from a joy." (ID21)

While participants viewed any type of strong emotion as able to precipitate diabetes, only one participant reported a positive emotion of sudden happiness as causing her diabetes. Notably, while participants emphasized the importance of an acute trigger of strong emotion, some had experienced repeated traumas leading to chronic emotional stress, which interviewees also perceived as contributing to diabetes.

Diet

A minority of participants attributed diabetes to diet (21%, n=6). Similar proportions did so in Suchitepéquez (19%) and Tecpán (23%). These participants cited soda, sugary drinks, sweets, foods with fat and grease as contributing to overweight and diabetes. Two participants cited hunger and irregularity in eating schedule as causing their diabetes. As one of them described it, "It was from poverty and suffering hunger at the time." (ID4)

Heredity

Few interviewees described heritability or genetics as a cause of diabetes (14%, n=4). Similar proportions invoked heredity across sites (13%, Suchitepéquez and 15%, Tecpán). Three of these participants reported that other family members had diabetes but expressed doubt over whether diabetes is a hereditary disease. As one interviewee responded: "I don't know, heritability? Because my siblings basically all [have diabetes]. I have eight siblings; six have diabetes." (ID17)

Treatment

When asked open-ended questions about treatment for diabetes, interviewees volunteered four major options.

Medication

All interviewees reported using medication for diabetes treatment (100%, n=29); this theme was observed in every interview in both sites (100%, Suchitepéquez and 100%, Tecpán). Most commonly, participants used metformin and/or sulfonylureas,

Box 1. SAMPLE QUOTATIONS ON THE PERCEIVED CAUSALITY, TREATMENT, AND CURABILITY/CHRONICITY OF DIABETES

Code	Theme	Sample Quote
Note:
ID = Identification
Causes of Diabetes	Emotion	"I blame it on anger. I was fine, normal, and then I got very angry. There were the words that hurt me [during a conflict at a meeting]. I left at once but I started to get dizzy. My face felt flushed … I imagined it was because I got angry. It provoked all my problems, I thought. But the doctor told me that it was very likely that I had already been diabetic without realizing it." (ID14) "I realized that one of my children was drinking a lot, like his dad. People had told me that before, but I did not believe it, so I went to see him, and he was drinking so much. I hit him, and the next day I became very swollen. I went to the hospital and they told me that I have diabetes." (ID16) "One day I was so happy looking at some of my animals, as I like animals. Then I got such an ugly thirst. That was 5 years ago." (ID15)
	Diet	"I have eaten [a lot of] cake. I drank Pepsi Cola for many years, ice cream, a lot of soda, a lot of carbonated drinks while at work." (ID25)
	Inheritance	"Well the truth is I do not know, but perhaps by inheritance I think… maybe my dad gave me this disease." (ID13)
Treatment	Medications	"I am taking my medicines every day, I only forget once in a while. I have various checks and sometimes my sugar goes down a tiny bit, 140, 195. That's where it has been. So, I send my sheet [of sugars] to the doctor. And he says … keep taking the medicines. I go to the pharmacy to buy them. I don't bring a prescription." (ID12)
Treatment	Herbal Remedies and Commercialized Naturopathic Products	"When I found out I was diabetic, I starting using tamarind with water, and something else from the woods called chalcupa … Other diabetics told me these things were good." (ID22) "I had kidney pain, and I went to a naturopath. They gave me a treatment to clean the kidneys, and another for diabetes, another to control the pressure, another for fat. I took four naturopathic vials, but yes, I took naturopathic medicine." (ID17)
	Diet	"My daughters help me and are trying to give me food that is better for diabetics." (ID27) "The best treatment is to eat healthy food." (ID4) "I got rid of all sodas, cakes, juices, everything that will raise my sugar. Sometimes it's hard. Just a tiny glass of a drink seems so little, but that's the most I ever have." (ID25)
	Exercise	"One day I went to the doctor, and when he did the exams, [he said], 'What happened?… Your sugars are normal!'… It was all the walking [I was doing]. I was running or walking blocks every day at that time."
Chronicity and Cure	Maintenance over Cure	"Yes, you can live well, but you have to take care of yourself." (IDI) "We can control ourselves with our own will. There are many ways to eat healthy. Everything is good, but only [in moderation]." (ID25)
	Role of God/Religion	"[There is no cure], but only God can heal you. I have seen this myself. In the hospital they told me I was going to die because I had dry kidneys, but I am still here bugging people. They were going to cut off my leg [below the knee] because I had a hole in my foot, it was purple and black. I asked God to heal me. When I went to the hospital, they asked me, 'What did you take?' I said, 'Nothing, I just prayed to God. But he hasn't finished curing me.'" (ID24) "Well, they say there is no cure, but I have faith in Christ that everything can be done, so God is the only one who can and I trust him." (ID26)

typically obtained from a government health facility free of cost or from a pharmacy through out-of-pocket costs. Perceptions of insulin were mixed; some feared that it could cause blindness or death, and others perceived its potential benefit:

Many people say that insulin is bad and that you take it right before you die. But the doctor says it is good. Insulin is really good and can help the body because my body isn't producing enough [insulin].

(ID1)

Most participants expressed the view that medications were effective in controlling diabetes. However, as described below, interviewees frequently used herbal remedies and commercialized naturopathic products in addition to allopathic medications.

Herbal remedies and commercialized naturopathic products

Most interviewees used herbal remedies and/or commercialized naturopathic products to treat diabetes (69%, n=20). Similar proportions reported using these treatments across sites (63%, Suchitepéquez and 77%, Tecpán). Reflecting a strong local tradition of ethnomedical healing with locally grown plants, participants reported harvesting plants from their land or the woods, or purchasing these plants at low cost in local markets, in order to drink herbal teas or broths, add herbs to their food, and consume blended fruits to treat diabetes. Patients referred to plants and herbs generally ("monte," "hierba" in Spanish, "q'ayïs" in Kaqchikel), rather than offering specific remedies; specific plants used in diabetes treatment have been detailed in prior ethnobotanical studies.⁹ They employed these remedies sometimes based on personal knowledge of herbal healing, and sometimes at the suggestion of family members, friends, or acquaintances. For example, one interviewee reported:

I started a month and eight days ago. I boil it [the herb] and put it in pure water. I will continue with this plant for three months. The lady [who told me about it] is also diabetic and says that it helps a lot.

(ID15)

Participants contrasted herbal remedies, perceived to address the root of diabetes, with medications, which were perceived to address only the symptoms of diabetes. For example, one participant reported taking plants to "clean the blood" (ID6). Participants combined plant use with use of commercialized naturopathic antidiabetic products sold in markets, on public transportation, or in consultation with a naturopath. Naturopathic medications included (a) vials of medications marketed specifically to improve kidney function, blood sugar, or blood pressure; (b) packaged plant-based creams or derivatives sold on buses; and (3) packaged vitamins and vitamin products such as trademarked Herbalife, Omnilife, and Neurobion products. Some interviewees perceived these products as mostly free of chemicals and therefore superior to allopathic medications. Notably, naturopathic products were of comparatively high cost, with price for one packaged product or treatment ranging from USD 5-175. Price of commercialized products was prohibitive for some participants:

I'm taking moringa [a local plant] … I was going to get a vitamin injection, but I can't afford it right now. I pray to God that I can get my vitamin injection. I was told it would detoxify me.

(ID27)

Diet

About half of interviewees described dietary modification as a treatment for diabetes (52%, n=15). Similar proportions reported this in Suchitepequez (50%) and Tecpán (54%). Most commonly, participants reported increasing vegetable intake and decreasing intake of sugar, soda, fats, and corn tortillas, a calorie-dense staple consumed at every meal. Some substituted rice, noodles, toast, or bread for tortillas. Notably, interviewees expressed difficulty adhering to the diabetic diet, particularly regarding limiting intake of tortillas. Tortillas were felt to be the filling and energy-giving part of a meal and important for individuals working in agricultural or manual labor:

I followed the diet completely. I wasn't even eating even two tortillas. I couldn't walk anymore. If I only eat two, I get very dizzy. My kids say maybe two isn't enough, because the tortillas provide strength. Today, I eat seven or eight tortillas at a meal. If I don't do this, I feel I don't have strength.

(ID18)

Exercise

Few participants described exercise as a treatment for diabetes (17%, n=5). All these participants were from Tecpán. One described running, two described walking, and two described exercise generally. Two considered housework, such as manually washing clothes in a large basin, a form of exercise. Overall, there was limited recognition of the role of exercise in diabetes control.

Chronicity and cure

When asked if diabetes had a cure, 34% felt unsure (n=10), 24% of participants responded yes (n=7), and 41% thought there was no cure (n=12). Slightly more participants in Tecpán expressed uncertainty (46%) or felt diabetes was curable (31%) compared with Suchitepéquez (31%, 19%, respectively). More participants in Tecpán felt diabetes had no cure (58%) compared with those in Suchitepéquez (31%). Those who were unsure struggled to reconcile information they had heard with experiences from family members and acquaintances. For example, one interviewee stated:

Well, people say yes, and people say no. My father was diabetic, and at [the hospital] they told him there is no cure. But several people say they had diabetes, and now they have nothing … but it doesn't go away. Well, we saw with my dad that there was no cure.

(ID7)

Others who were unsure indicated that they had faith in God despite hearing from medical professionals, family, or friends that there is no cure for diabetes. For example, one participant remarked: "I am in the hands of God. I do not know what to say, but I have heard that there have been people who have been cured." (ID20)

Indeed, most participants who viewed diabetes as curable did so due to belief in complete healing through religious faith and miracles. As one participant remarked: "Doctors say [there is no cure], but I have faith that God is going to cure me" (ID1). Some who perceived diabetes as curable described herbal remedies, commercial naturopathic products, and dietary changes as able to eliminate diabetes.

Interviewees who perceived diabetes as uncurable tended to make a distinction between cure and disease control or maintenance through self-care, dietary discretion, and medication adherence: "You can only control it, it doesn't have a cure" (ID17). Even participants who were unsure about or endorsed curability emphasized the importance of disease control. For example, one participant felt cure was contingent upon dietary modifications and supported by religious faith:

It has a cure if we take care of ourselves. But if we don't take care of ourselves, we eat a lot of fat, a lot of pork rinds, then we won't cure ourselves. But I have seen a lot of people get healthy. They have asked God to clean their blood … my mother-in-law was cured by God and support from the family. I have faith in God.

(ID6)

Discussion

In the face of a growing burden of type 2 diabetes in rural, Indigenous communities in Guatemala, it is important to understand perspectives and experiences of people living with diabetes in these areas. Prior qualitative work with individuals living with diabetes in rural Guatemala has primarily involved convenience samples of participants recruited from health clinics or tertiary care facilities and therefore may overrepresent perspectives of individuals with more severe disease or whose higher socioeconomic status helps them to successfully seek health care. In this paper we took advantage of a representative population survey to systematically sample the perspectives of community-dwelling individuals with diabetes. In general terms, our study findings are quite similar to those from other studies.^7-11 We also found similarities in understandings of causality, treatment, and disease management across both communities, with the exception of perceptions of exercise as a type of treatment for diabetes, which was expressed in Tecpán but not Suchitepéquez. This finding may relate to increasing urbanization and simultaneously greater departure from daily hard physical agricultural labor as a way of life in Tecpán but not Suchitepéquez. Based on our prior ethnographic work, we have observed differences in local conceptualizations of what types of activity constitute "exercise" (e.g., going for a walk) vs. "work" performed during the course of usual agricultural day labor (e.g., walking through fields while harvesting).⁸ Taken together with prior studies, our findings suggest shared cultural perspectives on diabetes causation, treatment, and management across community settings and disease severity categories.

Several findings merit further discussion. First, although participants in our study show appreciation of dietary and hereditary dimensions of diabetes pathology, the majority place great emphasis on the role of stress and psycho-emotional factors. The role of stress or "susto" in diabetes has been well documented in prior studies in Guatemala and other Latin American countries.^23–25 What is remarkable about the responses in our study is the predominance of this perspective (79%). This result has possible implications for clinical practice and program development. It is commonplace for biomedical providers in Guatemala to dispute their patients' etiological statements about emotion and its role in diabetes. A more productive and collaborative approach may be to affirm this insight by linking individual perspectives with the role of stress response in diabetes and the social determinants of poor health.^26,27 Clinicians may find it helpful to approach encounters with rural Guatemalan people with diabetes with cultural humility, i.e., attempt to openly understand and respect the complexity of their patients' sociocultural identities while practicing introspection about clinicians' own cultural values.

Second, a major finding here—also noted by prior publications—is that the majority of people with diabetes use both biomedical pharmaceuticals and natural plant-based remedies.^9,10 Plant-based remedies for numerous illness are an important part of ancestral healing practices in Maya cultures, which are actively practiced to the present day.²⁸ Despite a popular discourse in Guatemala pitting biomedicine against traditional Maya practices, our findings suggest that people with diabetes have a pragmatic openness to multiple healing modalities. In the area of cancer care in Guatemala, others have urged for bridge-building and collaboration between distinct healing systems.²⁹ Similar opportunities for diabetes care should also be explored by the Ministry of Health and other institutions delivering diabetes care.

However, the use of Maya healing practices should not be confused with the rise of parapharmaceutical complementary medical products such as Herbalife and Omnilife, which many of our participants also endorsed. As previously described, these products are aggressively marketed as "natural" cures societally through billboards and radio advertisements as well as interpersonally through traveling salespeople and naturopaths.^30,31 However, they do not derive from or resemble traditional ancestral plant-based practices and are typically quite expensive—often substantially more expensive than biopharmaceuticals.³² The rise of these products within the diabetes marketplace, both in Guatemala and elsewhere, requires close attention.^33,34 The poor regulatory environment in Guatemala makes it unlikely that there will be robust governance around the production, marketing, purchase, and use of these non-prescription products.³⁵ In the future, more systematic efforts to alert health care providers about the widespread use of these products are important to protect diabetes patients from these highly exploitative products.^10,35

A remarkable feature of our sample is the poor overall glycemic control (Table 1) despite the majority taking glucose-lowering medications. There is clearly a need to strengthen systems of care for the prevention, treatment, and control of diabetes. One key area is access to insulin. In other published work, we have documented very low levels of insulin use in rural communities, even among diabetes patients with very high levels of hyperglycemia who receive care at government health clinics. A key contributing factor is that current Ministry of Health primary care guidelines restrict insulin dispensing to regional hospitals and other tertiary care facilities, effectively denying access to most of the rural population.²¹ An area of active interest for our group is exploring strategies for improving the quality of diabetes care, like other ongoing Ministry-level efforts to improve the quality of care for cardiovascular risk factors associated with diabetes. These could include support for training of health post staff on evidence-based guidelines, formulary guidance and expansion to include insulin, and frontline worker-led initiatives to provide support for more effective diabetes self-care.^36,37

Finally, a large minority of individual participants felt that diabetes was curable through divine intervention (24%), and another significant proportion were uncertain (34%). Guatemala has experienced a rapid increase in belief in charismatic faith healing occasioned by a religious and demographic transition toward evangelical and Pentecostal Christian religious sects over the last several decades.³⁸ Our findings here suggest that outreach to religious leaders of these sects by health care providers and institutions might be an important step forward to improve the care for religious sect members who are living with diabetes.

Our study has important limitations. Although we systematically sampled people living with diabetes from two majority Indigenous municipalities in central Guatemala, findings from these communities may not be generalizable throughout Guatemala. Men were underrepresented in the underlying population survey, and their perspectives are thus underrepresented here. Our study was designed to assess local models of causation, treatment, and management of diabetes rather than differences in perceptions of diabetes based on gender, age, socioeconomic status, or other demographic characteristics. Our research team would like to explore these questions in the future. While our study elicited a local model of disease causation, treatment, and management, we did not include specific questions about participants' sources of information about these topics. Therefore, we cannot provide evidence about the individuals and institutions that help create the local model.

In conclusion, in a systematically sampled group of participants from rural Guatemala with type 2 diabetes, we found strong beliefs in the causality of psychoemotional factors and stress for diabetes pathology, concomitant use of biopharmaceutical and traditional plant-based remedies, and a belief in faith-based healing. These findings have implications for diabetes programming in this population. Exploring the social determinants of health with patients and communities and building collaborations with religious leaders and traditional Maya medical practitioners could be productive steps in improving care for rural Indigenous people with diabetes in Guatemala.