A Study Comparing LY900014 to Insulin Lispro (Humalog) in Children and Adolescents With Type 1 Diabetes

Change from baseline in HbA1c was analyzed using mixed model repeated measures (MMRM) and includes fixed class effects of treatment, strata (pooled country, type of basal insulin, and age group), visit, and treatment-by-visit interaction, as well as the continuous, fixed covariates of baseline value. An unstructured covariance structure will be used to model the within-participant errors.

The Efficacy estimand included data collected prior to permanent discontinuation of study drug through Week 26.

Change from baseline in HbA1c postprandial was analyzed using (MMRM and includes fixed class effects of treatment, strata (pooled country, type of basal insulin, and age group), visit, and treatment-by-visit interaction, as well as the continuous, fixed covariates of baseline value. An unstructured covariance structure will be used to model the within-participant errors.

The Efficacy estimand included data collected prior to permanent discontinuation of study drug through Week 26.

Documented post-dose hypoglycemia <54 milligrams per deciliter (mg/dL) and ≤ 70 mg/dL that occurred 1 and 2 hours after prandial dose.

Documented post-dose hypoglycemia event is an event of blood glucose of < 54 mg/dL and ≤70 mg/dL that occurred within 1 and 2 hours after the prandial dose. The rate of documented hypoglycemia was estimated by a negative binomial regression including treatment and age group as independent variable and number of episodes as dependent variables with log (exposure/365.25 days) as the offset in the model.

Documented hypoglycemia is defined as <54 mg/dL and ≤70 mg/dL, respectively.

Documented hypoglycemia is defined as a hypoglycemic event of blood glucose of ≤70 mg/dL or <54 mg/dL. The rate of documented hypoglycemia was estimated by negative binomial regression including treatment and age group as independent variables and number of episodes as dependent variable with log (exposure/365.25 days) as the offset in the model.

Severe hypoglycemia: during these episodes, participants have an altered mental status and cannot assist in their own care, may be semiconscious or unconscious, or experience coma with or without seizures, and require assistance of another person to actively administer carbohydrate, glucagon, or other resuscitative actions.

The rate of severe hypoglycemia per 100 years was calculated as: 100 times the total number of severe hypoglycemia episodes within the period divided by total exposure (in year) for all participants within the treatment group.

Change from baseline in insulin dose was analyzed using mixed model repeated measures (MMRM) and includes fixed class effects of treatment, strata (pooled country, type of basal insulin, age group, and HbA1c stratum (≤8.0%, >8.0%)), baseline value, visit and treatment-by-visit interaction. An unstructured covariance structure was used to model the within-participant errors.

Percentage of participants with HbA1c < 7.0% and <7.5% was analyzed using a longitudinal logistic regression with repeated measurements conducted by a generalized linear mixed model including independent variables of treatment, baseline HbA1c value, visit, baseline HbA1c-by-visit interaction, and treatment-by-visit interaction. An unstructured covariance structure was used.

Change from baseline in 7-point SMBG values were analyzed using MMRM and includes fixed class effects of treatment, strata (pooled country, type of basal insulin, and age group, and HbA1c stratum (≤8.0%, >8.0%)) baseline value, visit, and treatment-by-visit interaction. An unstructured covariance structure was used to model the within-participant errors.