Diabetes Treatment in Rural Guatemala

• ClinicalTrials.gov Identifier: NCT03626909
• Recruitment Status: Completed
• First Posted: August 13, 2018
• Last Update Posted: December 24, 2020
• Sponsor: University of Wisconsin, Madison
• Collaborator: Stanford University
• Information provided by (Responsible Party): University of Wisconsin, Madison

Study Description

Brief Summary:

In this study, the investigators will be using a smartphone application that the investigators developed to guide community health workers through the clinical assessment of patients with diabetes including collection of demographic data and past medical history, assessment of medication history, adherence, and adverse effects, measurement of glycemic control, screening for complications, medication administration and titration, and patient counseling.

Condition or disease	Intervention/treatment
Diabetes	Other: Primary visit and assessment; Other: Follow up visits; Other: 3 month visits; Other: Unscheduled visits

Detailed Description:

The burden of chronic adult diseases is surging worldwide, particularly type 2 diabetes, the prevalence of which is expected to double by 2030. The diabetes epidemic will primarily impact developing countries, with 80% of adult cases occurring in low- and middle-income countries (LMICs). Because many LMICs currently face a shortage of health professionals, the increasing burden of noncommunicable diseases, like type 2 diabetes, will tax already strained health systems. Furthermore, because many LMIC health care systems were developed to target acute illnesses and communicable disease, they are ill-prepared to treat and manage chronic adult disease. The divergence between the growing burden of chronic disease and the development of the health systems necessary to treat these diseases indicate the potential for a grave health, economic, and human crisis in the following decades. The WHO has consequently demanded that physicians designs systems providing "Innovative Care for Chronic Conditions" to meet this challenge4.

However, existing tools may provide a foundation for solutions to this growing crisis. Community Health Workers (CHWs, as known as health promoters) have become central to global health strategies since the Alma Ata Declaration of 1978, particularly in regions with physician shortages. In recent years, CHWs have had notable success in targeting childhood disease, particularly malnutrition and diarrhea, and offer a growing variety of primary care services. The success of these programs in providing consistent, sustainable care at the local level implies that longitudinal treatment for chronic adult diseases could be provided through parallel structures. While the treatment of chronic disease has become increasingly complex, the proliferation of smartphone and tablets across the globe have raised hopes that mobile health technology (mHealth) platforms can provide CHWs with algorithmic guidance on assessing and treating a broader set of diseases. The potential use of mHealth is a burgeoning field of global health research. The combination of CHWs and mHealth guidance may provide a solution to the rise of chronic disease in regions with physician shortages and weak health systems.

While many mHealth applications have been developed for Diabetes (over 1,000 are commercially available), only a small percentage (7.6%) are targeted to providers - and even fewer to providers in LMICs. Instead, these tools most commonly serve as tools for patient self-management, patient education, and medication adherence. A handful of programs have utilized smartphone technology to connect remote patients to health care workers in LMICs as well as to provide clinical guidance to providers, but such programs have been minimal and publications have been process oriented. In addition to improving diabetes care in the target population, the project also seeks to add to the evidence for this approach by designing an application-based algorithm that can assist CHWs in providing long-term diabetes care, titrating first- and second-line oral diabetes medications, and identifying dangerous diabetes complications in a setting of a lower middle-income country with a low physician density.

To test this delivery approach,the investigators focused on developing a diabetes treatment program in San Lucas Tolimán, Guatemala. This program seeks to provide treatment to diabetics living in the group of 19 rural villages with a combined population of 17,000, which surround San Lucas. San Lucas is an ideal community for studying these topics because it is facing a heavy burden of untreated Type II Diabetes, has medical personnel with mHealth experience, and has a well-developed CHW program. This CHW program is sponsored by the San Lucas Mission (SLM), an NGO providing health services in the area and a University of Wisconsin and Stanford University partner organization. Local health workers describe the increase in Type II Diabetes as an epidemic and there are few systems in place to provide community members with diabetes screening or effective and consistent treatment. Startling regional data on Type II Diabetes supports this concern: in Guatemala, the prevalence of diabetes has been estimated at 9.1-9.4%, with over 40% of cases undiagnosed22-24. The prevalence of diabetes has doubled over the past 30 years25. Fortunately, San Lucas has already developed a strong CHW program, including a tablet-based mHealth application that targets early childhood malnutrition, through a collaboration between the San Lucas Mission and Stanford School of Medicine. This application has enhanced the successful malnutrition program, allowing CHWs to more easily identify and manage malnutrition and decreasing training requirements for CHWs26. Utilizing the existence of the CHW program infrastructure and the established mHealth platform, the project seeks to develop and implement a CHW-led diabetes treatment program in San Lucas that is assisted by a smartphone application.

In order to inform the development of the smartphone application and program protocols, the investigators conducted a community needs assessment during the summer of 2016. Clinical data was used to provide a baseline estimate of diabetes prevalence and distribution in the communities as well as demographic risk factors. Interviews were conducted with local physicians, CHWs, and managers of the CHW system to understand current methods of diabetes treatment and define the limitations of these systems. Out of the 119 patients currently diagnosed with diabetes in the rural communities, 31 were interviewed to illuminate how the disease is currently diagnosed and treated, the effect the disease has on patient lifestyles, and patients' desired attributes for a diabetes treatment program. Finally,the investigators visited local diabetes clinics to determine the current state of diabetes treatment, the availability of medications and resources, and the level of care provided to patients.

Key findings of the community needs assessment were as follows:

1. Patients with diabetes in the rural communities have poor access to quality diabetes care. Only 58% of patients are taking medication on a regular basis and only 13% have achieved good glycemic control
2. Outreach clinics run by CHWs are disorganized, undersupplied, sporadic, and ineffective
3. CHWs lack the experience and training to effectively titrate oral diabetes medications, assess for possible complications, and provide health education for patients
4. Patients lack basic diabetes knowledge, particularly regarding self-management

Utilizing the knowledge gained with this needs assessment, established treatment guidelines for diabetes, and the expertise of SLM medical director Dr. Rafael Tun and the coordinators the SLM CHW program, the investigators developed protocols for the diabetes program, including a smartphone application to allow for algorithmic management. This process was iterative and collaborative and involved local partners at every step.

The investigators then trained a group of 10 CHWs, including 5 CHW coordinators (who have more clinical experience and take on a supervisory and training role for less-experienced CHWs) in the basics of diabetes management, program protocols, and the use of the smartphone application the investigators had developed. With close physician supervision, the investigators have beta-tested the use of the application with a small group of patients. Based on this experience, the investigators have further refined the application and program protocols. The investigators now endeavor to implement this program on a wide scale in the San Lucas area to both improve access to care for patients with diabetes and to establish the efficacy, feasibility, and safety of CHW-led, smartphone application-guided diabetes treatment.

An overview of study activities is as follows:

• The investigators will train additional CHWs in basic diabetes care, use of point-of-care (POC) testing technology, and use of the smartphone application that will guide their management of patients with diabetes.
• CHWs will recruit patients with diabetes in the rural villages outside of San Lucas to participate in the program.
• At the enrollment visit, CHWs will use the smartphone application to screen patients for appropriate inclusion in the program, establish glycemic targets, assess current glycemic control with hemoglobin A1c and blood glucose, measure height, weight, blood pressure, and waist circumference, assess for the presence of diabetes complications (diabetic ulcers, angina, diabetic eye disease), administer oral medications (metformin and/or glyburide, known locally by its alternate name glibenclamide) based on a medication dosing algorithm, and provide diabetes self-management education.
• CHWs will meet with patients on a monthly basis to assess medication adherence and for adverse effects, glycemic control (with blood glucose), screen for diabetic complications, refill medications with titration as needed (if experiencing medication adverse effects or blood glucose is significantly above or below treatment goals), and provide further diabetes education. Again, these activities will be guided by the smartphone application. Every 3 months, the monthly visit will also include A1c measurement for a more definitive measurement of diabetes control and to allow for titration of medications. Patients who are identified as having complications or who are not meeting treatment goals despite maximal dosing of metformin and glibenclamide allowed by the algorithm will be referred to SLM medical director Dr. Rafael Tun for definitive management.
• After all visits, including enrollment and monthly visits, Dr. Tun, in addition to the study investigators, will review data for all patients seen, including treatment recommendations made by the application and carried out by the CHWs, and make any changes to the treatment plan as needed based on his clinical judgement.
• Mean hemoglobin A1c and proportion of patients meeting treatment goals (primary endpoints) will be assessed at 6 months and compared to baseline, in addition to a number of secondary endpoints and safety measures as described in the relevant sections of this protocol. If possible, patients will also be followed out to 12 months with reassessment of primary and secondary endpoints.
• SLM hopes to continue this rural diabetes treatment program indefinitely, with the results of this study informing a quality improvement process to ensure the provision of high quality care.

The investigators believe that the novel aspect of this intervention, the use of a smartphone application to guide treatment decisions, improves on previous protocol-driven approaches in several ways. The use of a mobile computer-based algorithm as opposed to a paper algorithm allows for greater complexity and the incorporation of additional factors relevant to patient safety, such as the patient's current dose of medication, medication adherence, and medication side effects, in order to provide more specific recommendations. In this way, it decreases the cognitive burden placed on CHWs and the potential for human error. Rather than having to follow a complicated paper flowchart, CHWs will input information into the smartphone application, which will process the data and present the CHW with a concrete recommendation. Additionally, a computer-based system allows for easier review by the supervising physician and auditing and analysis of both program process measures and outcomes.

While CHWs will be acting on recommendations from the smartphone application without direct physician supervision at that moment, they will in essence be acting on "standing orders" from the physician because the treatment algorithms were designed by physicians and approved by the SLM medical director. CHWs will also be able to obtain point-of-care treatment recommendations from the medical director via telephone if there are questions about application recommendations or if a situation arises that falls outside the scope of the protocols.

Study Design

• Study Type: Observational
• Actual Enrollment: 89 participants
• Observational Model: Cohort
• Time Perspective: Prospective
• Official Title: Diabetes Treatment in Rural Guatemala
• Actual Study Start Date: January 14, 2018
• Actual Primary Completion Date: December 31, 2019
• Actual Study Completion Date: December 31, 2019

Groups and Cohorts

Intervention Details:

• Other: Primary visit and assessment

  At the initial visit on entry into the program CHW will collect information and the application will guide initial treatment and referral recommendations. At this visit:

  Collection of demographic information and relevant past medical history Medication history, adherence and side effects Glycemic testing, vital signs, and other anthropometric data. Screening for possible complications of diabetes Recommendations for referrals Medication recommendations and counseling

• Other: Follow up visits
  CHWs will meet with patients once per month to follow up on adherence and tolerance to medications, assess glycemic control, assess for complications of diabetes, refill medications (with dosing adjustments as needed for side effects, non-adherence or poor glycemic control), and provide diabetic education. These monthly visits will be facilitated by the smartphone application monthly protocol
• Other: 3 month visits
  At every 3rd monthly visit, starting with the visit 3 months after enrollment, A1c will be assessed. When A1c is checked, the month 3 medication titration algorithm, is used rather than the monthly titration algorithm, which uses blood glucose. Other than checking A1c and using the A1c-based algorithm as indicated, the procedures performed at the month 3 visit are identical to those of the monthly visit.
• Other: Unscheduled visits
  The CHWs live in the same communities as the patients they will be serving. As such, the investigators recognize that patients may come to them with concerns outside of the structure of monthly visits as described above. The investigators have designed an additional module for the smartphone application that guides the CHWs through an assessment for hypoglycemia or severe hyperglycemia and possible complications of diabetes. If a patient or CHW has a concern outside of the scope of these protocols, the CHW will contact a CHW coordinator and/or the medical director for guidance

Outcome Measures

Primary Outcome Measures :

1. Change in HgbA1c [ Time Frame: 3,6,9 and 12 months ]
   Change in mean percent of HgbA1c compared to value at baseline.
2. Patients meeting their HgbA1c treatment goal [ Time Frame: 3,6,9 and 12 months ]
   Percentage of patients meeting treatment goal for HgbA1c compared to this percentage at baseline. Treatment goal will be A1c ≤7 for most patients, A1c ≤8 for patients age 65 or older or who have 2 or more comorbidities, or other individualized goal for select patients as determined by the medical director

Secondary Outcome Measures :

1. Fasting blood glucose [ Time Frame: 3, 6, 9 and 12 months ]
   Mean fasting blood glucose compared to baseline value
2. BMI [ Time Frame: 3,6,9 and 12 months ]
   Body mass index compared to baseline value
3. Waist circumference [ Time Frame: 3,6,9 and 12 months ]
   Waist circumference compared to baseline value
4. Blood pressure [ Time Frame: 3,6,9 and 12 months ]
   Mean systolic and mean arterial pressure compared to baseline value
5. Diabetic complications [ Time Frame: 3,6,9 and 12 months ]
   Number and percentage of patients identified as potentially having the following complications: Angina,foot ulcers,eye complications,hypertension
6. Referrals for more advanced care [ Time Frame: 3,6,9 and 12 months ]
   Number of patients referred for potential complications, poor glycemic control or intolerance of diabetic medications and the number who complete referral
7. Medication adverse effects [ Time Frame: Through study completion, an average of 1 year ]
   Prevalence of adverse effects of metformin and hypoglycemia symptoms
8. Medication adherence [ Time Frame: Through study completion, an average of 1 year ]
   Patient adherence with medication regime
9. Community Health Worker adherence [ Time Frame: 3,6,9 and 12 months ]
   Community health worker adherence to recommendations provided by the application

Eligibility Criteria

• Ages Eligible for Study: 18 Years and older (Adult, Older Adult)
• Sexes Eligible for Study: All
• Accepts Healthy Volunteers: No
• Sampling Method: Non-Probability Sample

Study Population

Subjects will include adults 18 years of age and older and will be recruited from the 19 rural villages surrounding San Lucas Tolimán, Guatemala. Treatment of children with diabetes entails greater complexity and is beyond the scope of this program. Though there are no enrollment restrictions based on race or ethnicity, the majority of the people living in these villages are from the Kaqchikel ethnic group, an indigenous Mayan people of the midwestern highlands of Guatemala.

Criteria

Inclusion Criteria:

1. Willing to provide written informed consent
2. Willing to comply with all study procedures and be available for the duration of the study
3. Male or female, at least 18 years of age
4. Prior diagnosis of type 2 diabetes
5. Resident of one of the rural communities served by the CHW network of San Lucas Tolimán, Guatemala

Exclusion Criteria:

1. Type 1 diabetes
2. Women who are pregnant
3. Current use of insulin
4. Renal insufficiency (eGRF <30 mL/min/1.73 m2)
5. Unable to provide informed consent -

Contacts and Locations

Locations

Guatemala

Hospital Obras Sociales

San Lucas Tolimán, Solala, Guatemala

Sponsors and Collaborators

University of Wisconsin, Madison

Stanford University

Investigators

• Principal Investigator: James Svenson, MD, MS; University of Wisconsin, Madison

  Study Documents (Full-Text)

Documents provided by University of Wisconsin, Madison:

Study Protocol and Statistical Analysis Plan  [PDF] February 20, 2018

More Information

Publications of Results:

Duffy S, Norton D, Kelly M, Chavez A, Tun R, Ramírez MNG, Chen G, Wise P, Svenson J. Using Community Health Workers and a Smartphone Application to Improve Diabetes Control in Rural Guatemala. Glob Health Sci Pract. 2020 Dec 23;8(4):699-720. doi: 10.9745/GHSP-D-20-00076. Print 2020 Dec 23.

• Responsible Party: University of Wisconsin, Madison
• ClinicalTrials.gov Identifier: NCT03626909
• Other Study ID Numbers: 2017-0596; A534100 ( Other Identifier: UW Madison ); SMPH\EMERG MED\EMER MED ( Other Identifier: UW Madison )
• First Posted: August 13, 2018
• Last Update Posted: December 24, 2020
• Last Verified: December 2020

Individual Participant Data (IPD) Sharing Statement:

• Plan to Share IPD: No

• Studies a U.S. FDA-regulated Drug Product: No
• Studies a U.S. FDA-regulated Device Product: No

Additional relevant MeSH terms:

Diabetes Mellitus; Glucose Metabolism Disorders; Metabolic Diseases; Endocrine System Diseases