Technological Advances in Glucose Management in Older Adults (TANGO)

This is a study to assess the effectiveness of CGM (Continuous Glucose Monitor), enhanced by a diabetes management platform (DMP), collectively called enhanced CGM (eCGM), in the care of older patients with T1D. The DMP includes an automated data transfer from CGM, insulin-delivery devices, and activity tracker to a clinical decision support system (CDS) that provides dosing adjustment recommendations based on that data to the healthcare team. In addition, the DMP includes on-demand education for patients and caregivers, and an interface for communication between providers, patients, and their caregivers.

Hypoglycemia is a major and often devastating complication of T1D in the elderly. CGM has been shown to reduce the risk for hypoglycemia in adults with T1D including some more functional patients over 65 years old. However, the Medicare population is heterogeneous and may have age-related clinical and functional impairments that can impact self-care. These patients will require additional targeted guidance and support to fully realize the potential benefits of CGM. To address these age-specific barriers which could limit the effective use of CGM, in our planned RCT (Specific Aim 1) the use of CGM will be coupled with the DMP (Diabetes Management Platform), a tablet-based technology platform ( termed enhanced CGM (eCGM)). The CGM, insulin delivery, and activity data uploaded from the DMP will be analyzed by the clinical decision support system (CDS), which will provide insulin dosing recommendations to the study physicians, who will then accept or reject changes in therapy. The use of the DMP is expected to help the less technologically proficient Medicare patients to derive benefit from CGM. Specific Aim 2 will involve extensive mixed methods research (including semi-structured interviews of patients and caregivers) directed at making an in-depth assessment of barriers to the use of diabetes technology in older adults. This investigation will provide the evidence-base for future improvements in both the technology and clinical approach to the training of older adults and their caregivers. Specific Aim 3 will involve a cost-effectiveness analysis of the technology system (CGM with DMP = enhanced CGM [eCGM]) used in the trial as well as quality of life measures, providing a foundation for decision-making on coverage.

The intervention group will use eCGM (CGM with Diabetes Management Platform(DMP)) and an activity meter. The DMP will be pre-loaded with geriatric-specific education material, weblink to online education and surveys. The CGM, insulin delivery, and activity data uploaded from the DMP will be analyzed by the clinical decision support system (CDS), which will provide insulin dosing recommendations to the study physicians, who will then accept or reject changes in therapy. The DMP can also be configured to route the insulin regimen change approved by the study physician to the designated care providers of the patient. Blue-tooth unabled insulin pens will also provide additional data to verify if the patient is taking recommended insulin doses.

The attention control group will receive an android tablet pre-loaded with activity monitor devices, education material, and weblink to online education and surveys. However, the data will not be analyzed by CDS. An independent physician and a study staff member- only caring for the control group subjects will review the insulin and glucose data at in-person and remote study visits and make appropriate dosing adjustments based on self monitoring glucose levels

Glucose (CGM and Bluetooth BG meter), insulin (pump or Bluetooth insulin pen) and activity data will be automatically uploaded via the subjects' tablet computers, and analyzed by the CDS. The CDS will, if indicated generate adjustable insulin dosing recommendations that will compensate for different insulin requirements following high vs low activity days. The recommendations of the CDS will be used by the clinical team in their therapeutic decision-making about insulin dosing adjustments at the scheduled study follow up visits and the remote visits between these in-person visits. In addition, study staff will provide recommendations regarding hypoglycemic warning symptoms, causes, and appropriateness of treatment.

Minutes per day CGM < 70 mg/dL assessed over 2 weeks CGM use (blinded in the control arm; unblinded in the treatment arm)

change of duration of hypoglycemia(minutes/day) from baseline to 6 months between intervention and control groups

Incidence of severe biochemical hypoglycemia (defined as CGM below 54mg/dL for > than 20 minutes) assessed over 2 weeks CGM use (blinded in the control arm; unblinded in the treatment arm)

change in severe biochemical hypoglycemia (episodes per day) from baseline to 6 months between intervention and control groups

Incidence of clinically severe hypoglycemia (requiring third party assistance or loss of consciousness) measured by clinical history,

change in severe clinical hypoglycemia (episodes per day) from baseline to 6 months between intervention and control groups

cost-effectiveness and cost-utility of using eCGM versus usual care with self-monitoring glucose monitoring by calculating the incremental cost-effectiveness ratios

Mixed-method approach using semi-structured interviews to assess barriers and facilitators in those participants who fail the pretrial run-in and those who derive benefits from eCGM

Inclusion Criteria: Patients with age ≥ 65 years Community-living Clinical diagnosis of T1D On multiple insulin injections (≥3 injection/s day) or insulin pump. Exclusion Criteria: Use of real-time CGM in past 2 years A1c > 10% (since individuals with very poor glycemic control usually have barriers to optimal self-care that preclude effective use of technology) Use of insulin pump that cannot be uploaded for CDS Unable or unwilling to perform task needed for study participation during the run-in period Severe vision or hearing impairment that could interfere with study tasks Need to use acetaminophen on regular basis (since can interfere with CGM accuracy) Living in an institutional setting (e.g. group homes, nursing homes) Terminal diseases with life expectancy < 1 year (e.g. malignancy) Severe comorbidities that prevent completing outcome measurements (e.g. severe dementia, severe vision impairment, severe functional disabilities, inability to perform basic activities of daily living) Alcohol or other drug abuse Conditions that impact wear of CGM (e.g. CHF with edema, skin conditions); and End stage renal insufficiency (eGFR<30), or on dialysis (since impact of fluid shift on sensor lag not clearly understood).

Munshi M, Slyne C, Adam A, Davis D, Michals A, Atakov-Castillo A, Weinger K, Toschi E. Impact of Diabetes Duration on Functional and Clinical Status in Older Adults With Type 1 Diabetes. Diabetes Care. 2022 Mar 1;45(3):754-757. doi: 10.2337/dc21-2000.

Munshi M, Slyne C, Davis D, Michals A, Sifre K, Dewar R, Atakov-Castillo A, Toschi E. Use of Technology in Older Adults with Type 1 Diabetes: Clinical Characteristics and Glycemic Metrics. Diabetes Technol Ther. 2022 Jan;24(1):1-9. doi: 10.1089/dia.2021.0246.

Toschi E, Slyne C, Sifre K, O'Donnell R, Greenberg J, Atakov-Castillo A, Carl S, Munshi M. The Relationship Between CGM-Derived Metrics, A1C, and Risk of Hypoglycemia in Older Adults With Type 1 Diabetes. Diabetes Care. 2020 Oct;43(10):2349-2354. doi: 10.2337/dc20-0016. Epub 2020 May 27. Erratum In: Diabetes Care. 2021 Jan;44(1):299.

Toschi E, Munshi MN. Benefits and Challenges of Diabetes Technology Use in Older Adults. Endocrinol Metab Clin North Am. 2020 Mar;49(1):57-67. doi: 10.1016/j.ecl.2019.10.001. Epub 2019 Nov 18.