A Study to Determine the Efficacy and Safety of Albiglutide as Compared With Liraglutide.

HbA1c is a form of hemoglobin that is measured primarily to identify the average plasma glucose concentration over a 2- to 3-month period. The Baseline HbA1c value is defined as the last non-missing value before the start of treatment. Change from Baseline was calculated as the post-Baseline value minus the Baseline value. The LOCF method was used to impute missing data, in which the last non-missing post-Baseline on-treatment measurement was used to impute the missing measurement. If a participant had missing observation(s) immediately after Baseline, the Baseline observation was not carried forward and was left as missing. Participants were considered in the treatment week if they had received at least one dose in that treatment week.

The FPG test measures blood sugar levels after the participant has not eaten (fasted) for 12 to 14 hours. The Baseline FPG value is the last non-missing value before the start of treatment. Change from Baseline was calculated as the post-Baseline value minus the value at Baseline. The analysis was performed using an Analysis of Covariance (ANCOVA) model with treatment group, region, Baseline HbA1c category, history of prior myocardial infarction (yes versus no), and age category (\<65 years versus ≥65 years) as factors and Baseline FPG as a continuous covariate. The LOCF method was used to impute missing data, in which the last non-missing post-Baseline on-treatment measurement was used to impute the missing measurement. If a participant had missing observation(s) immediately after Baseline, the Baseline observation was not carried forward and was left as missing. Participants were considered in the treatment week if they had received at least one dose in that treatment week.

The FPG test measures blood sugar levels after the participant has not eaten (fasted) for 12 to 14 hours. The Baseline FPG value is the last non-missing value before the start of treatment. Change from Baseline was calculated as the post-Baseline value minus the value at Baseline. The LOCF method was used to impute missing data, in which the last non-missing post-Baseline on-treatment measurement was used to impute the missing measurement. If a participant had missing observation(s) immediately after Baseline, the Baseline observation was not carried forward and was left as missing. Participants were considered in the treatment week if they had received at least one dose in that treatment week.

Number of participants who achieved HbA1c response levels of \<6.5% and \<7.0% at Week 32 were assessed. The last observation carried forward (LOCF) method was used to impute missing data, in which the last non-missing post-Baseline on-treatment measurement was used to impute the missing measurement. If a participant had missing observation(s) immediately after Baseline, the Baseline observation was not carried forward and was left as missing.

Participants who experienced persistent hyperglycemia (high blood glucose) could have qualified for hyperglycemia rescue. The conditions for hyperglycemia rescue were as follows: fasting plasma glucose (FPG) \>=280 milligram/decilitre (mg/dL) \>= Week 2 and \< Week 4, FPG \>=250 mg/dL \>= Week 4 and \<Week 12, HbA1c ≥8.5% and ≤0.5% reduction from Baseline- \>= Week 12 and \<Week 26, or HbA1c ≥8.5% \>= Week 26. Time to hyperglycemia rescue is defined as the time between the date of the first dose of study medication and the date of hyperglycemia rescue plus one day, or the time between the date of the first dose of study medication and the date of the last visit during the active treatment period plus one day for participants not requiring rescue. This time was divided by 7 to express the result in weeks. All times extending beyond Week 32 relevant to hyperglycemia rescue were censored at Week 32.

The Baseline value is the last non-missing value before the start of treatment. Change from Baseline was calculated as the value at Week 32 minus the value at Baseline. The analysis was performed using an Analysis of Covariance (ANCOVA) model with treatment group, region, Baseline HbA1c category, history of prior myocardial infarction (yes versus no), and age category (\<65 years versus ≥65 years) as factors and Baseline weight as a continuous covariate. The LOCF method was used to impute missing data, in which the last non-missing post-Baseline on-treatment measurement was used to impute the missing measurement. If a participant had missing observation(s) immediately after Baseline, the Baseline observation was not carried forward and was left as missing. Weight values obtained after hyperglycemia rescue were treated as missing and replaced with pre-rescue values.