HbA1C was measured in blood samples at a central laboratory.

Intent-to-treat population: All randomized participants who completed the study.

Participants measured their blood glucose at least 3-4 times daily throughout the study. The mean blood glucose level was calculated for each 3-day period from Baseline to Week 24 and the percentage of 3-day blood glucose levels with the target range of 70-180 mg/dL (3.9-10 mmol/L) was calculated for the 2 reporting periods of Screening to Baseline and Week 23 to Week 24.

Intent-to-treat population: All randomized participants who completed the study.

A symptomatic hypoglycemic episode was defined as an event with symptoms consistent with hypoglycemia which was confirmed by a blood glucose reading < 70 mg/dL (3.9 mmol/L). Symptoms might include but were not limited to sweating, dizziness, lightheadedness, tremors, nervousness, hunger, headaches, and weakness or tiredness.

Intent-to-treat population: All randomized participants who completed the study.

Approximately half of the investigational sites monitored glucose levels in participants enrolled in this study using the DexCom Seven® Plus Continuous Glucose Monitoring device. The device provides glucose measurements every 5 minutes for up to 7 days. The system contains a sensor, transmitter, and receiver. The sensor is a flexible round wire that goes under the skin to read glucose levels. The transmitter snaps into the sensor and wirelessly sends glucose readings to the receiver. Data was obtained from approximately one-third of participants and was used to calculate MAGE. Data were collected in the 3 days prior to Baseline and the Week 24 visit. The standard deviation of the blood glucose measurements in each 3-day period was calculated. For each glucose measurement, the difference from the previous reading was calculated. Absolute differences smaller than the standard deviation were discarded. MAGE is the mean of the remaining difference scores.

Intent-to-treat population: All randomized participants with continuous glucose monitoring data at Baseline and at Week 24 who completed the study.

Participants using the Accu-Chek® Aviva Expert blood glucose meter were encouraged to use the Bolus Advisor utility incorporated into the meter. A bolus opportunity occurs, for example, just before eating a meal.

Intent-to-treat population: All randomized participants.

Participants using the Accu-Chek® Aviva Expert blood glucose meter were encouraged to use the Bolus Advisor utility incorporated into the meter. A bolus opportunity occurs, for example, just before eating a meal.

Intent-to-treat population: All randomized participants.

Participants using the Aviva Nano blood glucose meter received individualized advice in how to use their insulin:carbohydrate ratio (I:CHO) and insulin sensitivity factor (ISF) values to determine their insulin dose. The I:CHO ratio advice was considered to have been used correctly if the meal bolus dose the participant indicated in his/her patient diary was in accordance with the dose which could be calculated based upon the participant's total carbohydrate intake and the I:CHO ratio. The ISF advice was considered to have been correctly used if the correction bolus dose the participant indicated in his/her patient diary was in accordance with the dose which could be calculated based upon the participant's blood glucose target, current blood glucose value, and the ISF.

Intent-to-treat population: All randomized participants who completed the study. Results are only reported for participants using the Aviva Nano blood glucose meter who had data available for analysis.

Participants were asked to assess the carbohydrate content (grams) of 10 standardized meals by using a set of Dose Adjustment for Normal Eating (DAFNE) plates, which provide standardized photographs of meals with known carbohydrate values. The mean meal error (MME), an indicator of accuracy, and mean meal absolute error (MMAE), an indicator of variability were calculated from their responses. The MME is defined as the mean of the differences between the estimated carbohydrate content and the actual carbohydrate content over the 10 DAFNE plates. A negative MME indicates an underestimation and a positive MME indicates an overestimation of the actual carbohydrate content. The MMAE is defined as the mean of the absolute value of the differences between the estimated carbohydrate content and the actual carbohydrate content over the 10 DAFNE plates. The MMAE is ≥ 0 with a lower value indicating a better ability to estimate the actual carbohydrate content.

Intent-to-treat population: All randomized participants who completed the study and had data available for analysis.

The Patient Health Questionnaire depression scale (PHQ-8) contains 8 items which can be rated from 0='not at all' to 3='nearly every day'. The total PHQ-8 score is the sum of the responses to the 8 items and ranges from 0 to 24. A lower score indicates less depression. A negative change score indicates improvement.

Intent-to-treat population: All randomized participants who had data available for analysis.

The Major Depression Disorder (MDD) scale is derived from the Patient Health Questionnaire depression scale (PHQ-8) and was used to categorize participants in regard to the severity of their depression. There are 5 categories on the MDD scale: None, mild, moderate, moderately severe, and severe. The category for each participant is determined from their PHQ-8 score. A total PHQ-8 score of 0 to 4 represents no significant depressive symptoms. A total PHQ-8 score of 5 to 9 represents mild depressive symptoms; 10 to 14, moderate; 15 to 19, moderately severe; and 20 to 24, severe. A higher score indicates more depression.

Intent-to-treat population: All randomized participants who completed the study.

The Problem Area in Diabetes (PAID) scale contains 20 items which can be rated from 0='not a problem' to 4='serious problem'. The total PAID scale score ranges from 0 to 80 with a higher score indicating more diabetes-related problems. A negative change score indicates improvement.

Intent-to-treat population: All randomized participants who completed the study and had data available for analysis.

The Hypoglycemia Fear Survey-II (HFS-II) contains 33 items (15 items regarding behavior and 18 items regarding worry) which can be rated from 0='never' to 4='always'). The total HFS-II score ranges from 0 to 132 with a higher score indicating more fear. A negative change score indicates improvement.

Intent-to-treat population: All randomized participants who completed the study and had data available for analysis.

The Diabetes Treatment Satisfaction Questionnaire at Baseline (DTSQs) contains 6 items which can be scored from 0='very bad' to 6='very good'. The total score is the sum of the scores of the 6 items and ranges from 0 to 36. A higher score indicates more satisfaction. This questionnaire was administered at Baseline only.

The Diabetes Treatment Satisfaction Questionnaire for change from Baseline (DTSQc) to study end contains 6 items which can be rated from -3='much worse now' to 3='much better now'). The total score is the sum of the scores of the 6 items and ranges from -18 to 18. A higher score indicates more satisfaction. This questionnaire was administered at the end of the study (Week 24) only.

Intent-to-treat population: All randomized participants who completed the study.