JMIR Publications

ABSTRACT

Background:

GLP-1 RAs and basal insulin both lower blood sugar, but while insulin puts people at risk for hypoglycemia and weight gain, GLP-1 RAs do not. In addition, GLP-1 RAs have additional cardiometabolic and renal benefits. For these reasons, when possible, many providers prefer their patients with well-controlled T2D be transitioned from basal insulin to a GLP-1 RA.This transition process can be labor-intensive, requiring multiple dosing adjustments and a watchful eye for (the short-term risk of) hypoglycemia. The MITI-GLP1 Program uses text message-based technology to support a streamlined and supervised transition process from basal insulin to a GLP-1 RA.

Objective:

Patients referred to MITI-GLP1 have fasting blood sugars that are already well-controlled. The objectives of the MITI-GLP1 Pilot Program are to increase the GLP-1 RA dose, lower the basal insulin dose, and avoid hypoglycemia during the transition.

Methods:

Patients are enrolled on a secure web platform that sends them a daily text message asking, “What was your fasting blood sugar this morning?" Each weekday, texted responses containing patients’ fasting blood sugars are checked for alarm values and once-weekly, patients are called and advised on if/how to lower their basal insulin and if/how to increase their GLP-1 RA. The program is co-run by General Internal Medicine Doctors and Nurses and continues until the patient has their insulin stopped completely and/or is maxed on their GLP-1 RA, or when 16 weeks is reached. Outcomes include those reflecting insulin reduction, GLP-1 RA escalation, avoidance of hypoglycemia during the transition, and patient engagement.

Results:

Seventy-two patients completed the pilot program. At baseline, they had an average age of 54 years, with 71% self-identifying as Hispanic or Latinx, 61% opting for texts in Spanish, and 61% having no health insurance. The average starting basal insulin dose was 20 units once-daily, the average starting BMI was 30 kg/m2, and the average starting fasting blood glucose was 126 mg/dL. Fifty-five patients (76%) had their basal insulin reduced by ≥ 50%; most of these (45 patients or 63%) had their insulin stopped completely. Forty five patients (63%) were discharged on the maximum dose of their GLP-1 RA. The average ending fasting blood glucose was 122 mg/dL. Regarding hypoglycemia, five out of 3,520 text messages (0.14%) had a value < 80 mg/dL. Patient engagement was high. Out of 3,671 text messages sent by the program, 3,520 (96%) received a reply from patients. Out of 719 cumulative once-weekly call attempts, 649 (90%) successfully reached patients.

Conclusions:

A General Internal Medicine-run MITI-GLP1 Pilot Program using text messaging and interdisciplinary teamwork between internists and nurses can safely and effectively transition people with well-controlled Type 2 Diabetes away from basal insulin and towards a GLP-1 RA.