While closed-loop insulin delivery (CLID) systems demonstrated safety and effectiveness in patients with unbalanced type 1 diabetes (T1D), no studies have included patients with highly and chronically unbalanced diabetes.

The investigators conduct a retrospective, observational, and single-center study to evaluate the acceptability, safety, and efficacy of a CLID system in patients living with T1D (≥2 years) with a HbA1c>11% in the past 12 months and a mean HbA1c >10% over the past three years. Efficacy was assessed using continuous glucose monitoring parameters.

If hybrid closed-loop systems (HCLS) ("artificial pancreas") for automated insulin administration have demonstrated their great interest in patients living with moderately unbalanced type 1 diabetes (T1D) (HbA1c < 8 .6% on average), no study has looked at their safety, acceptability and efficacy in patients living with highly and chronically unbalanced T1D (HbA1c > 11%). These patients, often young, present more or less advanced microangiopathic complications (retinopathy, nephropathy, neuropathy, etc.) or are highly exposed to the appearance of these complications, the prevention of which, whether primary or secondary, involves obtaining a good glycemic balance over time. These patients "escape" the possibilities of current insulin therapy, whether by multiple insulin injections or by insulin pump, more or less coupled to a continuous glucose monitoring system. Automated insulin delivery systems could be a solution to achieve less catastrophic glycemic control in some of these patients. The investigators are proposing a pilot study aimed at evaluating acceptability, safety and effectiveness of a HCLS currently available in France (the SMARTGUARD™ MINIMED™ 780G or the Control-IQ) in patients followed in the diabetology department of the Centre Hospitalier Sud-Francilien (CHSF), living with highly unbalanced T1D (HbA1c > 11%) in the context of usual follow-up.

Inclusion Criteria:

Patients with closed loop system :

• Adult patients
• With T1D for at least 2 years
• Follow-up in the diabetes department of the CHSF
• Having a laboratory HbA1c > 11% at least once during the past year and at least one other HbA1c > 10% in the previous 3 years
• Accepting the closed loop system as part of their support
• Subject informed of the study and not objecting to it

Patients without closed loop system :

Adult patients

• With T1D for at least 2 years
• Follow-up in the diabetes department of the CHSF
• Having a laboratory HbA1c > 11% at least once during the past year and at least one other HbA1c > 10% in the previous 3 years
• Refusing the closed loop system as part of their support or
• Patients for whom a multi-professional medical and paramedical collegial discussion (including diabetologist and nurse referents of the patient) within the service leads to the evaluation of an unfavorable benefit/risk ratio
• Subject informed of the study and not objecting to it

Exclusion Criteria:

Patients with closed loop system :

• Patients without an Internet connection, or without a smartphone or computer
• Patients for whom a multi-professional medical and paramedical collegial discussion (including diabetologist and patient referent nurse) within the department results in the assessment of an unfavorable benefit/risk ratio (patients with a history of interrupted pump treatment on medical decision, serious psychiatric disorders…)

Patients followed in the diabetes department of the Centre Hospitalier Sud Francilien