Randomized, controlled, multinational, multi-center, clinical trial to examine whether HbA1c can improve in type 1 diabetes patients who continuously use the Paradigm® REAL-Time system with alarm function as compared to patients on multiple injection therapy receiving one six-day period of continuous glucose monitoring - without alarm function (Guardian® REAL-Time Clinical).

1. Patients have been diagnosed with type 1 diabetes at least 12 months prior to study entry;

2. Patients are between 18 and 65 years of age, inclusive;

3. Patients are on multiple injection treatment, defined as a basal insulin analogue qd or bid and a rapid-acting insulin analogue used with every meal OR
Patients are on conventional MIT in whom previous treatment with long- and rapid-acting insulin has failed;

4. Patients are on multiple injection treatment at least 3 months prior to inclusion;

5. Patients have a baseline HbA1c ¡Ý 8.2%

1. Patient has hearing problems or impaired vision that might hinder recognition of the sensor alarm or screen alarms, respectively;

2. Alcohol or drug abuse other than nicotine;

3. Abdominal abnormalities, like lipodystrophia that might hinder either glucose measurement by the sensor or the continuous subcutaneous insulin infusion;

4. Current pharmaceutical treatment for any psychiatric disorder other than depression;

5. Treatment with CSII in the last six months prior to entry in the study;

6. Patients suffering from Cancer, Heart Failure, kidney disease (creatinin > 150 micromol/l) and other chronic debilitating conditions;

7. Patient is unwilling or unable to comply with the provisions of the protocol;

8. Patient has scheduled a vacation which will occur between Visit 1 and Visit 2;

9. Patient has planned trips when he/she will be out of telephone reach from the study medical care for >5 days or to a place where he/she cannot comply with study procedures;

10. Being pregnant, or the wish to become pregnant during the trial;

11. Patient is participating in another device or drug study.