Automated Insulin Delivery for INpatients With DysGlycemia (AIDING) Feasibility

This single-arm stepwise feasibility study will test initial deployment of hybrid closed-loop (HCL) automated insulin delivery (AID) using the Omnipod 5/Horizon HCL system with remote monitoring and device operation capabilities to hospitalized patients admitted to the general medical/surgical floor with diabetes (type 1 or type 2) requiring insulin therapy. All enrolled participants will be placed on HCL insulin therapy for 10 days or until hospital discharge (if less than 10 days) to determine functional operability of the system and its effect on glycemic control in the hospital setting. This study will generate preliminary data to inform the design of a large multi-institution randomized controlled trial to assess superiority of HCL compared to standard inpatient insulin therapy.

• The Omnipod 5/Horizon HCL system (Device): The Omnipod 5/Horizon HCL system, consists of a disposable insulin infusion pump (or "pod"), a built-in model predictive control (MPC) insulin dosing algorithm, and a remote Personal Diabetes Manager (PDM) interface, that interact with a Dexcom G6 continuous glucose monitor (CGM) to automatically control insulin delivery based upon real-time glucose values. The PDM component also enables remote interaction with the system, including glucose monitoring as well as insulin dosing management and adjustments.

Primary outcomes

• Percentage of Time Spent in HCL After CGM Sensor Meets Initial Validation Criteria [Up to 10 days (or hospital discharge if before 10 days)]: The percentage of time spent in HCL after CGM sensor met initial validation criteria of sensor glucose value being within ±20% of point of care (POC) values (for glucose levels ≥70 mg/dL) or ±20 mg/dL for POC glucose values <70 mg/dL.
• Percentage of Time Sensor Glucose is Within Target Glucose Range [Up to 10 days (or hospital discharge if before 10 days)]: The percentage of time that the sensor glucose measurement is within the target glucose range of 70-180 mg/dL.

Secondary outcomes

• Time From Enrollment to Start of HCL Therapy (After Initial CGM Validation) [Up to 10 days (or hospital discharge if before 10 days)]: The time from enrollment to start of HCL therapy, after initial CGM validation, was recorded.
• Percentage of Time With CGM Readings [Up to 10 days (or hospital discharge if before 10 days)]: The percentage of time during study participation with CGM readings was calculated.
• Percentage of CGM Values Meeting Accuracy Criteria for Bolus/Correction Insulin Dosing [Up to 10 days (or hospital discharge if before 10 days)]: The percentage of CGM values meeting accuracy criteria for bolus/correction insulin dosing was calculated.
• Percentage of CGM Readings Within %15/15 of POC Readings and Within %20/20 of POC Readings With the Cut Point at 70 mg/dL [Up to 10 days (or hospital discharge if before 10 days)]: The percentage of CGM values within 15% or 15 mg/dL (%15/15) and within 20% or 20 mg/dL (%20/20) of POC reference values for blood glucose, using a cut point at 70 mg/dL. The reference values are derived from a total of 597 paired CGM and reference capillary glucose values.
• Number of Hypoglycemic (<70 mg/dL) Episodes Per Patient [Up to 10 days (or hospital discharge if before 10 days)]: The number of hypoglycemic (<70 mg/dL) episodes per patient.
• Number of Hypoglycemic (<70 mg/dL) Episodes Per Patient-day [Up to 10 days (or hospital discharge if before 10 days)]: The number of hypoglycemic (<70 mg/dL) episodes per patient-day in hospital.
• Number Clinically Important Hypoglycemic (<54 mg/dL) Episodes Per Patient [Up to 10 days (or hospital discharge if before 10 days)]: The number clinically important hypoglycemic (<54 mg/dL) episodes per patient.
• Percent Time Below Range (TBR) of <70mg/dL [Up to 10 days (or hospital discharge if before 10 days)]: The percent time below range (TBR), defined as blood glucose <70mg/dL.
• Percent Time Below Range (TBR) of <54 mg/dL [Up to 10 days (or hospital discharge if before 10 days)]: The percentage of time below range (TBR) of <54 mg/dL.
• Percent Time Above Range (TAR) of >180 mg/dL [Up to 10 days (or hospital discharge if before 10 days)]: The percentage of time above range (TAR) of >180 mg/dL.
• Percent Time in Severe Hyperglycemia (>250 mg/dL) [Up to 10 days (or hospital discharge if before 10 days)]: The percentage of time in severe hyperglycemia, defined as >250 mg/dL.
• Frequency of Setting Overall Adjustments for Clinically-important Hypoglycemia (<54 mg/dL) [Up to 10 days (or hospital discharge if before 10 days)]: The frequency of setting adjustments across all participants for clinically-important hypoglycemia, defined as <54 mg/dL.
• Frequency of Setting Adjustments for Clinically-important Hypoglycemia (<54 mg/dL) to Basal Rate [Up to 10 days (or hospital discharge if before 10 days)]: The frequency of setting adjustments across all participants for clinically-important hypoglycemia (<54 mg/dL) to basal rate.
• Frequency of Setting Adjustments for Clinically-important Hypoglycemia (<54 mg/dL) to Insulin Carb Ratio (ICR) [Up to 10 days (or hospital discharge if before 10 days)]: The frequency of setting adjustments across all participants for clinically-important hypoglycemia (<54 mg/dL) to insulin carb ratio (ICR).
• Frequency of Setting Adjustments for Clinically-important Hypoglycemia (<54 mg/dL) to Insulin Sensitivity Factor (ISF) [Up to 10 days (or hospital discharge if before 10 days)]: The frequency of setting adjustments across all participants for clinically-important hypoglycemia (<54 mg/dL) to insulin sensitivity factor (ISF).
• Frequency of Setting Overall Adjustments for Prolonged Hyperglycemia (>250 mg/dL for >1 Hour) [Up to 10 days (or hospital discharge if before 10 days)]: The frequency of setting adjustments across all participants for prolonged hyperglycemia, defined as blood glucose >250 mg/dL for over one hour.
• Frequency of Setting Adjustments for Prolonged Hyperglycemia (>250 mg/dL for >1 Hour) to Basal Rate [Up to 10 days (or hospital discharge if before 10 days)]: The frequency of setting adjustments across all participants for prolonged hyperglycemia, defined as blood glucose >250 mg/dL for over one hour, to basal rate.
• Frequency of Setting Adjustments for Prolonged Hyperglycemia (>250 mg/dL for >1 Hour) to ICR [Up to 10 days (or hospital discharge if before 10 days)]: The frequency of setting adjustments across all participants for prolonged hyperglycemia, defined as >250 mg/dL for over one hour, to ICR.
• Frequency of Setting Adjustments for Prolonged Hyperglycemia (>250 mg/dL for >1 Hour) to ISF [Up to 10 days (or hospital discharge if before 10 days)]: The frequency of setting adjustments across all participants for prolonged hyperglycemia, defined as blood glucose >250 mg/dL for over one hour, to ISF.
• Total Daily Insulin (TDI) [Up to 10 days (or hospital discharge if before 10 days)]: The total daily insulin (TDI) was calculated.
• Total Daily Basal Insulin (TBI) [Up to 10 days (or hospital discharge if before 10 days)]: The total daily basal insulin (TBI) was calculated.
• Total Daily Bolus Meal/Correction [Up to 10 days (or hospital discharge if before 10 days)]: The total daily bolus meal/correction was recorded
• Number of Hypoglycemic Events That Required Assistance of Another Person [Up to 10 days (or hospital discharge if before 10 days)]: The number of hypoglycemic events across all participants that required assistance of another person due to altered consciousness to actively administer carbohydrate, glucagon, or other resuscitative actions.
• Number of Diabetic Ketoacidosis Events [Up to 10 days (or hospital discharge if before 10 days)]: The number of diabetic ketoacidosis (DKA) events across all participants.
• Patient Acceptability of HCL System [At time of device discontinuation (up to 10 days of use or hospital discharge if before 10 days)]: To assesses acceptability of the HCL system participants responded to the statement "Overall, I liked using the Omnipod 5/Horizon system to treat my blood sugar in the hospital" with five (5) options to choose from: Strongly Agree, Agree, Neither Agree nor Disagree, Disagree, and Strongly Disagree. The responses were not assigned a score, rather the number of participants for each response were examined.
• Patient Perceptions of HCL System Use [At time of device discontinuation (up to 10 days of use or hospital discharge if before 10 days)]: Participants were asked to provide their perceptions of the HCL system with four questions that were responded to with "Yes" or "No". Responses are assigned a score and a summary score is not calculated, rather the number of participants responding "Yes" or "No" to each of the individual questions is examined.

Sex: All, ages: 18 Years to any

Inclusion Criteria:

* Patients ≥18 years of age with insulin-treated T1 or Type 2 diabetes mellitus (T2DM) admitted to general (non-intensive care) medical-surgical hospital service requiring inpatient insulin therapy.

Exclusion Criteria:

* Patients admitted the ICU or anticipated to require ICU transfer
* Anticipated length of hospital stay <48 hours.
* Evidence of hyperglycemic crises (diabetic ketoacidosis or hyperosmolar hyperglycemic state) at enrollment
* Severely impaired renal function (eGFR < 30 ml/min/1.73m2) or clinically significant liver failure
* Severe anemia with hemoglobin <7 g/dL
* Evidence of hemodynamic instability
* Hypoxia (SpO2 <95% on supplemental oxygen)
* Pre-admission or inpatient total daily insulin dose >100 units
* Mental condition rendering the participant unable to consent or answer questionnaires
* Pregnant or breast feeding at time of enrollment
* Unable or unwilling to use rapid-acting insulin analogs (Humalog, Admelog, Novolog or Apidra) during the study
* Use of hydroxyurea or high-dose ascorbic acid (>1g/day)
* Coronavirus Disease 2019 (COVID-19) infection or person under investigation (PUI) on isolation precautions

Participant flow

Outcome results

• Percentage of Time Spent in HCL After CGM Sensor Meets Initial Validation Criteria (Primary), Mean, percentage of time: Hybrid Closed-loop (HCL) Automated Insulin Delivery (AID) 91 +/- 11
• Percentage of Time Sensor Glucose is Within Target Glucose Range (Primary), Mean, percentage of time in range (TIR): Hybrid Closed-loop (HCL) Automated Insulin Delivery (AID) 68 +/- 16
• Time From Enrollment to Start of HCL Therapy (After Initial CGM Validation) (Secondary), Median, hours: Hybrid Closed-loop (HCL) Automated Insulin Delivery (AID) 0 (0 to 0)
• Percentage of Time With CGM Readings (Secondary), Median, percentage of time: Hybrid Closed-loop (HCL) Automated Insulin Delivery (AID) 99 (97 to 100)
• Percentage of CGM Values Meeting Accuracy Criteria for Bolus/Correction Insulin Dosing (Secondary), Mean, percentage of CGM values: Hybrid Closed-loop (HCL) Automated Insulin Delivery (AID) 85 +/- 14
• Percentage of CGM Readings Within %15/15 of POC Readings and Within %20/20 of POC Readings With the Cut Point at 70 mg/dL (Secondary), Mean, percentage of CGM values: Hybrid Closed-loop (HCL) Automated Insulin Delivery (AID) 71 +/- 19
• Number of Hypoglycemic (<70 mg/dL) Episodes Per Patient (Secondary), Median, episodes per patient: Hybrid Closed-loop (HCL) Automated Insulin Delivery (AID) 0 (0 to 2)
• Number of Hypoglycemic (<70 mg/dL) Episodes Per Patient-day (Secondary), Median, episodes per patient-day: Hybrid Closed-loop (HCL) Automated Insulin Delivery (AID) 0 (0 to 0.3)
• Number Clinically Important Hypoglycemic (<54 mg/dL) Episodes Per Patient (Secondary), Median, episodes per patient: Hybrid Closed-loop (HCL) Automated Insulin Delivery (AID) 0 (0 to 2)
• Percent Time Below Range (TBR) of <70mg/dL (Secondary), Mean, percentage of time below range: Hybrid Closed-loop (HCL) Automated Insulin Delivery (AID) 0.17 +/- 0.3
• Percent Time Below Range (TBR) of <54 mg/dL (Secondary), Mean, percentage of time below range: Hybrid Closed-loop (HCL) Automated Insulin Delivery (AID) 0.06 +/- 0.2
• Percent Time Above Range (TAR) of >180 mg/dL (Secondary), Mean, percentage of time above range: Hybrid Closed-loop (HCL) Automated Insulin Delivery (AID) 25 +/- 11
• Percent Time in Severe Hyperglycemia (>250 mg/dL) (Secondary), Mean, percentage of time: Hybrid Closed-loop (HCL) Automated Insulin Delivery (AID) 6.9 +/- 7.0
• Frequency of Setting Overall Adjustments for Clinically-important Hypoglycemia (<54 mg/dL) (Secondary), Number, setting adjustments: Hybrid Closed-loop (HCL) Automated Insulin Delivery (AID) 3
• Frequency of Setting Adjustments for Clinically-important Hypoglycemia (<54 mg/dL) to Basal Rate (Secondary), Number, setting adjustments: Hybrid Closed-loop (HCL) Automated Insulin Delivery (AID) 0
• Frequency of Setting Adjustments for Clinically-important Hypoglycemia (<54 mg/dL) to Insulin Carb Ratio (ICR) (Secondary), Number, setting adjustments: Hybrid Closed-loop (HCL) Automated Insulin Delivery (AID) 3
• Frequency of Setting Adjustments for Clinically-important Hypoglycemia (<54 mg/dL) to Insulin Sensitivity Factor (ISF) (Secondary), Number, setting adjustments: Hybrid Closed-loop (HCL) Automated Insulin Delivery (AID) 0
• Frequency of Setting Overall Adjustments for Prolonged Hyperglycemia (>250 mg/dL for >1 Hour) (Secondary), Number, setting adjustments: Hybrid Closed-loop (HCL) Automated Insulin Delivery (AID) 6
• Frequency of Setting Adjustments for Prolonged Hyperglycemia (>250 mg/dL for >1 Hour) to Basal Rate (Secondary), Number, setting adjustments: Hybrid Closed-loop (HCL) Automated Insulin Delivery (AID) 6
• Frequency of Setting Adjustments for Prolonged Hyperglycemia (>250 mg/dL for >1 Hour) to ICR (Secondary), Number, setting adjustments: Hybrid Closed-loop (HCL) Automated Insulin Delivery (AID) 25
• 8 more outcome measures not shown.

Adverse events

Time frame: Information on adverse events was collected during exposure to study device, from the time individuals gave consent to participate (up to 10 days of use or hospital discharge if before 10 days).

• Study Protocol and Statistical Analysis Plan (PDF, 2020-11-23)

• Francisco Pasquel, M.D., M.P.H, Principal investigator, Emory University