The purpose of the CLEAR study is to determine the effect on counterregulatory responses (CRR) of intervening (by attempting to strictly avoid hypoglycemia) to improve awareness of hypoglycemic symptoms among adults with type 1 diabetes (T1D) who have impaired awareness of hypoglycemia (IAH). IAH affects 20-25% of adults with T1D, and rises with increasing duration of T1D.

Individuals with IAH exhibit blunted symptomatic and CR hormonal responses to hypoglycemia and, as such, have an impaired ability to respond to hypoglycemia. Thus, rates of severe hypoglycemia are up to 6-fold greater in those affected. Intensive management of T1D is necessary in preventing long-term complications, but can be complicated by recurrent episodes of hypoglycemia which lead to and sustain the CRR deficits of IAH. Technologies such as continuous glucose monitoring (CGM) and hybrid closed-loop (HCL) systems can reduce severe hypoglycemia (and also may reduce IAH) but the ability of technology to reverse impaired CRR (as assessed with experimental hypoglycemia clamp) remains unclear. Behavioral and psycho-educational interventions targeting knowledge/skills gaps, as well as particular cognitions and behaviors driving recurrent hypoglycemia, can also reduce severe hypoglycemia and improve awareness. No studies have compared technology with such behavioral interventions in terms of assessing their impact on IAH or the CRR (as a primary outcome). Unanswered questions include the degree of reduction in hypoglycemia required to restore awareness. Furthermore, participants may respond to different interventions according to their characteristics. For example, it remains unclear whether older individuals benefit from such interventions since they usually are excluded from studies. Therefore, there is an urgent need to determine effective interventions that can reverse IAH in a large representative population of adults with T1D and IAH. The investigators propose to study the effect of specific interventions aimed at restoring

• the CRR (tested via an experimental hypoglycemia clamp procedure)
• hypoglycemia awareness (self-reported via the Towler Questionnaire during the experimental hypoglycemia clamp procedure)

The study will use a Sequential Multiple Assignment Randomized Trial (SMART) design. At baseline, all participants who are HCL naïve will be randomized to HCL or Usual Care (UC) plus brief education (My HypoCOMPaSS) with a follow-up of two years. UC will consist of real-time continuous glucose monitoring (CGM) and insulin delivery via pump or multiple daily injections. Participants who fail to increase their CRR at 12 months will be randomized, or assigned, to a second intervention consisting of a small-group educational program focusing on motivations and unhelpful cognitions acting as barriers to hypoglycemia avoidance (HARPdoc). At baseline, all participants who are HCL non-naïve will be randomized to optimized HCL or HCL plus My HypoCOMPaSS; those with non-responsive CRR at 12 months will be randomized to either continue HCL (on the basis they need a longer period to reverse impaired CRR and total symptomatic responses) or to the HARPdoc intervention. Participants randomized to an HCL device are expected to wear the device continually, as well as a CGM. The My HypoCOMPaSS education requires 4-5 hours of training, whereas, the HARPdoc education requires four training sessions of seven hours each during weeks 1,2,3, and 6.

The specific aims and hypotheses are as follows:

Aim 1: To determine the effect on CRR (epinephrine increase ≥ 125 pg/ml over baseline) and total symptom responses (Towler Questionnaire increase ≥ 20% over baseline) during a hyperinsulinemic-hypoglycemic clamp procedure (glucose < 50 mg/dl) after 12 months of HCL versus Usual Care plus My HypoCOMPaSS Educational Intervention among adults with T1D and IAH who have never used HCL therapy previously.

Hypothesis 1: At 12 months, those allocated to Usual Care plus My HypoCOMPaSS will be more likely to have improved CRR and total symptomatic responses than those allocated to HCL.

Aim 2: To determine the effect on CRR and total symptom responses at 12 months of HCL plus My HypoCOMPaSS versus HCL alone among adults with T1D and IAH who are currently using HCL therapy prior to entering the study.

Hypothesis 2: At 12 months, those allocated to HCL plus My HypoCOMPaSS will be more likely to have improved hypoglycemic awareness and improved CRR than those using HCL alone.

Aim 3: To determine the durability of effect over 24 months of the intervention that improves CRR at 12 months among adults with type 1 diabetes and IAH at baseline.

Hypothesis 3: At 24 months, CRR will improve further among those who had restored CRR at 12 months.

Aim 4. To determine the effect on hypoglycemic awareness (Towler Questionnaire increase ≥ 20% over baseline) and CRR (epinephrine increase ≥ 125 pg/ml over baseline) during a hyperinsulinemic hypoglycemic clamp procedure at 24 months of an in-depth educational program (HARPdoc), initiated throughout months 12-24, among adults with T1D and IAH at baseline, for whom the intervention allocated at baseline did not restore CRR at 12 months.

Hypothesis 4: At 24 months, those allocated to HARPdoc for months 12-24 months will be more likely to have improved hypoglycemic awareness and CRR than those who continue with the therapy allocated at baseline.

Inclusion Criteria:

• Clinical diagnosis of type 1 diabetes
• Gold Score or Clarke Score ≥ 4 (highly associated with IAH)
• Random non-fasting C-peptide < 200 pmol/L
• Diabetes duration ≥ 10 years
• HbA1c < 10.5%
• Total Daily Insulin Dose of < 1 unit/kg
• Ability to read and speak English (because validated non-English versions of the cognitive tests and the educational interventions are not available)

Exclusion Criteria:

• Medical conditions that limit participation in study activities, as determined by the PI (including but not limited to cognitive dysfunction, reduced hearing, reduced vision, cancer under active treatment, untreated angina, organ failure)

• Active alcohol or drug abuse (as defined by DSM criteria of either 1) recurrent use of alcohol/drugs resulting in a failure to fulfill major role obligations at work, school, or home, 2) recurrent alcohol/drug use in situations in which it is physically hazardous, or 3) recurrent alcohol or drug-related legal problems)

• Social determinants of health that limit participation in study activities, as determined by the PI (including but not limited to homelessness, food insecurity, inadequate social support)

• Seizure disorder unrelated to hypoglycemia associated seizures, unless documented seizure-free for >12 months and on a stable regimen of anti-convulsant therapy

• Skin conditions that would preclude the use of a CGM

• Super-physiologic exposure to steroids within one month of enrollment

• eGFR < 45 mL/min/1.73 m2

• History of bariatric surgery that irreversibly alters gut innervation and structure

• Hyper- or hypokalemia (serum potassium >5.5 or <3.5 mmol/L)*

• Hemoglobin < 10 g/dL*

• Medical condition that requires intermittent or continuous use of glucocorticoids at greater than physiological replacement doses

• Pregnancy, plan for pregnancy, or breast feeding

• Abnormal thyroid function tests of clinical significance, as determined by PI*

• Liver transaminases > 3 times the upper limit of normal*

• Hospitalization for mental illness in last year

• History of adrenalectomy

  • At discretion of the PI, laboratory tests may be repeated once. If the participant is not eligible after the second attempt, then the participant. The participant may be screened again.

The release data sets will be stored at the NIDDK Central Repository.

The release data sets will de-identified, i.e., they will not contain names, social security numbers, addresses, phone numbers, health care records, and/or similar protected health information. In addition, the release data sets will have an anonymous study ID that is linked to an individual only in the participant records at the Biostatistics Research Center at the Penn State College of Medicine.

The timeline for data sharing from the Impaired Awareness of Hypoglycemia Consortium will meet NIH data sharing requirements, such as one year after the Consortium investigators have published the primary manuscript. The Biostatistics Research Center at the Penn State College of Medicine will make data available after the acceptance for publication of the main findings that address the specific aims of the study.

Investigators external to the Consortium will be able to request the data from the NIDDK Central Repository. They will be required to show proof of their local IRB approval.