HbA1c is measured as a percentage. Thus, this change from baseline reflects the Week 24 HbA1c percent minus the baseline HbA1c percent. Means are treatment adjusted for baseline HbA1c and previous anti-diabetic medication.

The Full Analysis Set (FAS) included all treated and randomised patients with a baseline and at least one on-treatment HbA1c measurement available. Last observation carried forward (LOCF) was used as the imputation rule.

HbA1c is measured as a percentage. Thus, this change from baseline reflects the Week 6 HbA1c percent minus the baseline HbA1c percent. Means are treatment adjusted for baseline HbA1c and previous anti-diabetic medication.

The Full Analysis Set (FAS) included all treated and randomised patients with a baseline and at least one on-treatment HbA1c measurement available. Last observation carried forward (LOCF) was used as the imputation rule.

HbA1c is measured as a percentage. Thus, this change from baseline reflects the Week 12 HbA1c percent minus the baseline HbA1c percent. Means are treatment adjusted for baseline HbA1c and previous anti-diabetic medication.

The Full Analysis Set (FAS) included all treated and randomised patients with a baseline and at least one on-treatment HbA1c measurement available. Last observation carried forward (LOCF) was used as the imputation rule.

HbA1c is measured as a percentage. Thus, this change from baseline reflects the Week 18 HbA1c percent minus the baseline HbA1c percent. Means are treatment adjusted for baseline HbA1c and previous anti-diabetic medication.

The Full Analysis Set (FAS) included all treated and randomised patients with a baseline and at least one on-treatment HbA1c measurement available. Last observation carried forward (LOCF) was used as the imputation rule.

This change from baseline reflects the Week 24 FPG minus the baseline FPG. Means are treatment adjusted for baseline HbA1c, baseline FPG and previous anti-diabetic medication.

This population includes the FAS using the LOCF imputation, with the further restriction of patients with a baseline and post-baseline FPG value.

This change from baseline reflects the Week 2 FPG minus the baseline FPG. Means are treatment adjusted for baseline HbA1c, baseline FPG and previous anti-diabetic medication.

This population includes the FAS using the LOCF imputation, with the further restriction of patients with a baseline and post-baseline FPG value.

This change from baseline reflects the Week 6 FPG minus the baseline FPG. Means are treatment adjusted for baseline HbA1c, baseline FPG and previous anti-diabetic medication.

This population includes the FAS using the LOCF imputation, with the further restriction of patients with a baseline and post-baseline FPG value.

This change from baseline reflects the Week 12 FPG minus the baseline FPG. Means are treatment adjusted for baseline HbA1c, baseline FPG and previous anti-diabetic medication.

This population includes the FAS using the LOCF imputation, with the further restriction of patients with a baseline and post-baseline FPG value.

This change from baseline reflects the Week 18 FPG minus the baseline FPG. Means are treatment adjusted for baseline HbA1c, baseline FPG and previous anti-diabetic medication.

This population includes the FAS using the LOCF imputation, with the further restriction of patients with a baseline and post-baseline FPG value.

The percentage of patients with an HbA1c value below 7% at week 24 was calculated for each treatment arm. If a patient did not have an HbA1c value at week 24 they were considered a failure, so HbA1c above 7%.

FAS treated and randomised patients with baseline HbA1c >= 7.0%. Non-completers were considered as failure imputation (NCF).

The percentage of patients with an HbA1c value below 7% at week 24 was calculated for each treatment arm. If a patient did not have an HbA1c value at week 24 they were considered a failure, so HbA1c above 7%.

The FAS included all treated and randomised patients with a baseline and at least one on-treatment HbA1c measurement available. Non-completers were considered as failure imputation (NCF).

The percentage of patients with an HbA1c value below 6.5% at week 24 was calculated for each treatment arm. If a patient did not have an HbA1c value at week 24 they were considered a failure, so HbA1c above 6.5%. Only patients with baseline HbA1c =< 6.5%

FAS treated and randomised patients with baseline HbA1c >= 6.5%. Non-completers were considered as failure imputation (NCF).

The percentage of patients with an HbA1c value below 6.5% at week 24 was calculated for each treatment arm. If a patient did not have an HbA1c value at week 24 they were considered a failure, so HbA1c above 6.5%

The FAS included all treated and randomised patients with a baseline and at least one on-treatment HbA1c measurement available. Non-completers were considered as failure imputation (NCF).

The percentage of patients with an HbA1c reduction from baseline greater than 0.5% at week 24 was calculated for each treatment arm. If a patient did not have an HbA1c value at week 24 they were considered a failure, so HbA1c reduction less than 0.5%.

The FAS included all treated and randomised patients with a baseline and at least one on-treatment HbA1c measurement available. Non-completers were considered as failure imputation (NCF).

This change from baseline reflects the Week 24 2h PPG minus the baseline 2h PPG. Means are treatment adjusted for baseline HbA1c, baseline 2h PPG and previous anti-diabetic medication.

Meal tolerance test (MTT) set (patients with adequate MTT results available at the beginning and end of the randomised treatment period)

HbA1c is measured as a percentage. Thus, this change from baseline reflects the Week 24 HbA1c percent minus the baseline HbA1c percentage. Mean is unadjusted.

The open label set (OLS) included all patients with baseline HbA1c too high to be randomised and were assigned open-label medication, also with an on treatment HbA1c value. The Open label Set (OLS) included all patients treated with at least one dose on open-label L2.5+M1000 medication.

This change from baseline reflects the Week 24 FPG minus the baseline FPG. Mean is unadjusted.

The open label set (OLS) included all patients with baseline HbA1c too high to be randomised and were assigned open-label medication, also with a baseline on treatment HbA1c value. The Open label Set (OLS) included all patients treated with at least one dose on open-label L2.5+M1000 medication.

The use of rescue therapy (SUs, thiazolidinediones [TZDs], or insulin) was permitted only during the randomised treatment period of the trial (i.e. Visits 3 to 7), and was to be administered only if a patient had a 'confirmed' glucose level after an overnight fast.

Percentage of patients requiring rescue therapy