Structured population data were available for 3 of 3 supporting studies. 2 studies described participants with type 2 diabetes. 3 studies described adult participants.

Evidence uses heterogeneous comparison contexts: active_comparator, placebo, usual_care.

1 supporting study has high, serious, critical, unclear, or unavailable risk-of-bias information.

1 study contains conflicting p-value and significance fields; those significance claims are suppressed.

Basal Insulin (Glargine)

Placebo

Core clinician-reference fields are available or not applicable.

Participants received GLP-1 (Liraglutide) in combination with metformin and sulfonylurea.

GLP-1 (Liraglutide)

We considered three treatment arms: GLP-1 (Liraglutide), basal insulin (glargine), and placebo. After training, virtual trials were conducted by sampling 232 virtual patients per arm (according to the LEAD-5 inclusion criteria) and predicting post-treatment outcomes.

Participants received basal insulin (glargine) in combination with metformin and sulfonylurea.

We considered three treatment arms: GLP-1 (Liraglutide), basal insulin (glargine), and placebo.

Participants received placebo in combination with metformin and sulfonylurea.

Authoritative quality diagnostic: invalid or missing unit.

Authoritative quality diagnostic: invalid or missing value.

AI Emulation Shows GLP-1 Improves HbA1c in Type 2 Diabetes

Once-daily basal insulin comparator

Participants received once-weekly insulin icodec.

Once-weekly insulin icodec

Participants were randomized (1:1) to receive either once-weekly icodec or OD basal insulin comparator.

Participants received once-daily basal insulin comparators, which varied by trial.

Participants were randomized (1:1) to receive either once-weekly icodec or OD basal insulin comparator, which varied by trial.

Authoritative quality diagnostic: p value significance conflict.

Insulin icodec may improve glycaemic control with low hypoglycaemia risk

Basal plus group

Participants walked for 15–20 minutes after one meal per day for 6 weeks.

Post-meal walking group

In post-meal walking group, patients walked after meal for 15–20 minutes one meal a day every day for 6 weeks.

Participants received one prandial insulin injection before breakfast or main meal.

In prandial insulin (basal plus) group, one prandial insulin was injected before breakfast or main meal with rapid-acting insulin.

Post-meal walking may improve glycemic control in type 2 diabetes

Structured population data were available for 3 of 3 supporting studies. 1 study described participants with type 2 diabetes. 3 studies described adult participants.

1 supporting study does not have a displayable topic-matched quantitative effect.

Structured population data were available for 4 of 4 supporting studies. 3 studies described participants with type 2 diabetes. 4 studies described adult participants.

Upstream consistency scoring indicates mixed findings across the evidence.

Evidence uses heterogeneous comparison contexts: active_comparator, usual_care.

Structured population data were available for 8 of 8 supporting studies. 4 studies described participants with type 2 diabetes. 8 studies described adult participants.

1 relevant study has unresolved authoritative design evidence and does not contribute to design-specific or RoB-specific claims.

Evidence uses heterogeneous comparison contexts: baseline_within_group, unknown_unresolved, usual_care.

2 supporting studies have high, serious, critical, unclear, or unavailable risk-of-bias information.

Evidence uses heterogeneous comparison contexts: baseline_within_group, usual_care.

2 supporting studies do not have a displayable topic-matched quantitative effect.

Evidence uses heterogeneous comparison contexts: active_comparator, unknown_unresolved.

3 supporting studies have high, serious, critical, unclear, or unavailable risk-of-bias information.

Structured population data were available for 4 of 4 supporting studies. 4 studies described participants with type 2 diabetes. 4 studies described adult participants.

Structured population data were available for 5 of 5 supporting studies. Diabetes type was heterogeneous: 2 studies with type 1 diabetes, 1 studies with type 2 diabetes. Age groups were heterogeneous: 2 studies with adult participants, 1 studies with youth participants.

2 relevant studies have unresolved authoritative design evidence and do not contribute to design-specific or RoB-specific claims.

Evidence uses heterogeneous comparison contexts: baseline_within_group, other, usual_care.

4 of 5 studies with duration data have follow-up of 12 weeks or less.

Structured population data were available for 10 of 11 supporting studies. Diabetes type was heterogeneous: 5 studies with type 2 diabetes, 2 studies with type 1 diabetes. 8 studies described adult participants.

3 relevant studies have unresolved authoritative design evidence and do not contribute to design-specific or RoB-specific claims.

3 supporting studies do not have a displayable topic-matched quantitative effect.

Evidence uses heterogeneous comparison contexts: active_comparator, baseline_within_group, placebo, unknown_unresolved, usual_care.

1 supporting study lacks usable structured population information.

4 supporting studies have high, serious, critical, unclear, or unavailable risk-of-bias information.

Structured population data were available for 8 of 8 supporting studies. 7 studies described participants with type 2 diabetes. 7 studies described adult participants.

Evidence uses heterogeneous comparison contexts: active_comparator, baseline_within_group, placebo.

Structured population data were available for 4 of 4 supporting studies. 4 studies described participants with type 1 diabetes. 2 studies described adult participants.

Structured population data were available for 8 of 8 supporting studies. 6 studies described participants with type 2 diabetes. 7 studies described adult participants.

Evidence uses heterogeneous comparison contexts: active_comparator, unknown_unresolved, usual_care.

Structured population data were available for 8 of 8 supporting studies. Diabetes type was heterogeneous: 4 studies with type 2 diabetes, 1 studies with type 1 diabetes. Age groups were heterogeneous: 4 studies with adult participants, 1 studies with youth participants.

5 of 8 studies with duration data have follow-up of 12 weeks or less.

Structured population data were available for 3 of 3 supporting studies. 3 studies described participants with type 2 diabetes. 3 studies described adult participants.

3 of 3 studies with duration data have follow-up of 12 weeks or less.

Evidence uses heterogeneous comparison contexts: active_comparator, placebo.

Evidence uses heterogeneous comparison contexts: active_comparator, baseline_within_group.

Structured population data were available for 3 of 3 supporting studies. Diabetes type was heterogeneous: 1 studies with type 1 diabetes, 1 studies with type 2 diabetes. 2 studies described adult participants.

Structured population data were available for 13 of 13 supporting studies. 11 studies described participants with type 2 diabetes. 12 studies described adult participants.

4 relevant studies have unresolved authoritative design evidence and do not contribute to design-specific or RoB-specific claims.

Evidence uses heterogeneous comparison contexts: active_comparator, baseline_within_group, usual_care.

Evidence uses heterogeneous comparison contexts: unknown_unresolved, usual_care.

Structured population data were available for 3 of 3 supporting studies. Diabetes type was heterogeneous: 2 studies with type 2 diabetes, 1 studies with type 1 diabetes. Age groups were heterogeneous: 2 studies with adult participants, 1 studies with youth participants.

Structured population data were available for 12 of 12 supporting studies. Diabetes type was heterogeneous: 9 studies with type 2 diabetes, 1 studies with type 1 diabetes. 12 studies described adult participants.

9 supporting studies have high, serious, critical, unclear, or unavailable risk-of-bias information.

Structured population data were available for 5 of 5 supporting studies. 5 studies described participants with type 2 diabetes. 4 studies described adult participants.

Structured population data were available for 3 of 3 supporting studies. 2 studies described participants with type 2 diabetes. 2 studies described adult participants.

Structured population data were available for 2 of 3 supporting studies. 2 studies described participants with type 2 diabetes. 2 studies described adult participants.

Evidence uses heterogeneous comparison contexts: active_comparator, placebo, unknown_unresolved.

Structured population data were available for 10 of 10 supporting studies. Diabetes type was heterogeneous: 8 studies with type 2 diabetes, 2 studies with type 1 diabetes. Age groups were heterogeneous: 8 studies with adult participants, 1 studies with youth participants.

Evidence uses heterogeneous comparison contexts: active_comparator, other, usual_care.

3 studies contain conflicting p-value and significance fields; those significance claims are suppressed.

Structured population data were available for 4 of 4 supporting studies. 2 studies described participants with type 2 diabetes. 4 studies described adult participants.

Evidence uses heterogeneous comparison contexts: other, usual_care.

2 of 4 studies with duration data have follow-up of 12 weeks or less.

Structured population data were available for 4 of 4 supporting studies. Diabetes type was heterogeneous: 2 studies with type 2 diabetes, 1 studies with type 1 diabetes. 4 studies described adult participants.

Structured population data were available for 5 of 5 supporting studies. 5 studies described participants with type 2 diabetes. 5 studies described adult participants.

Structured population data were available for 11 of 11 supporting studies. 10 studies described participants with type 2 diabetes. 11 studies described adult participants.

Structured population data were available for 6 of 6 supporting studies. 6 studies described participants with type 2 diabetes. 6 studies described adult participants.

Evidence uses heterogeneous comparison contexts: active_comparator, baseline_within_group, unknown_unresolved, usual_care.

Structured population data were available for 7 of 7 supporting studies. Diabetes type was heterogeneous: 2 studies with type 2 diabetes, 1 studies with type 1 diabetes. 5 studies described adult participants.

Structured population data were available for 5 of 5 supporting studies. 3 studies described participants with type 2 diabetes. 4 studies described adult participants.

Structured population data were available for 5 of 5 supporting studies. Diabetes type was heterogeneous: 4 studies with type 2 diabetes, 1 studies with type 1 diabetes. 3 studies described adult participants.

Evidence uses heterogeneous comparison contexts: baseline_within_group, placebo, usual_care.

Basal insulin × HbA1c

Available supporting studies

AI Emulation Shows GLP-1 Improves HbA1c in Type 2 Diabetes

Insulin icodec may improve glycaemic control with low hypoglycaemia risk

Post-meal walking may improve glycemic control in type 2 diabetes