Efficacy Analysis Sample Clauses

Efficacy Analysis. 4.4.1 Primary Efficacy Variable The primary end-point, that is, the proportion achieving “smoking reduction” at 24 weeks, will be analyzed with logistic regression modeling among eligible subjects using intent-to treat criteria and with participants who terminated the study prematurely for any reason considered to be failures.

Efficacy Analysis. Efficacy evaluation will be based on the full analysis set (see section 10.3) that includes all patients who received at least 1 dose of the study drug.

Efficacy Analysis. The key secondary endpoint, proportion of subjects achieving HiSCR50 at Week 12, will be analyzed using ▇▇▇▇▇▇’▇ exact test. Proportions and difference in proportions between treatment groups, and their associated 95% confidence intervals (CIs) will be reported. For subjects who discontinue study treatment prior to Week 12 due to any reason, their last observation will be used to impute response status. ▇▇▇▇▇▇’▇ exact test will also be used to analyze additional secondary endpoints which are dichotomized in nature, eg, proportion of subjects achieving HiSCR75 at Week 12, proportion of subjects with flare by Week 12, and other proportion endpoints. Change from baseline in NRS skin pain score, change from baseline in IHS4, and other continuous endpoints will be summarized using descriptive statistics. Change from baseline of these continuous endpoints will also be analyzed by a Mixed-effects model for repeated measures (MMRM). MMRM will include baseline as a covariate, treatment and visit as factors, and treatment-by-visit and baseline-by-visit interactions in the model.

Efficacy Analysis. The primary analysis of efficacy will be based on the evaluable treated subjects, hence, only those subjects who received a complete or partial study treatment and completed a follow-up visit at 30-days post-treatment #2 will be included in the analysis using the Last Value Carried Forward method. Analyses of safety will include all subjects who received (complete or incomplete) study treatment. Efficacy will be the proportion of treated subjects determined to be improved by the reduction in the number of sweat glands from baseline to 30-days post-treatment #2.

Efficacy Analysis. 10.7.2.1 Primary Efficacy Analysis (1) LS Means difference from stage 1 analysis (interim cohort analysis) (2) LS Means difference from stage 2 analysis (post-interim cohort analysis) 𝑟1: total number of patients planned to be included at the interim analysis divided by the total number of patients originally planned (planned sample size), i.e., 0.5 𝑁∗: final total number of patients divided by the total number of patients originally planned 𝑍2: combination test statistic at the end of the second stage obtained in equation 1 10.7.2.2 Sensitivity Analyses Three sensitivity analyses will be performed: • Multiple imputation methodology will be utilized to impute complete datasets with three monthly observations for all patients in the mITT population. This will be done within each treatment group conditioned on each patient’s non-missing data (including baseline), site, and the stratification factor. • A tipping point analysis will be performed using the above MI datasets and applying an increasing penalty to the active group until statistical significance is no longer achieved. • Finally, a missing-not-at-random multiple imputation approach will be utilized taking into account the reason for discontinuation; intermittent missing data will still be assumed to be missing at random. Full details of each analysis will be provided in the SAP.

Efficacy Analysis. A test for country-by-treatment interaction with respect to the primary efficacy endpoint, percent change from baseline in hepatic fat fraction at Week 12, as determined by MRI-PDFF, will be performed in the 50 mg cohort. An ANCOVA model with fixed effects for the stratification factor (diabetes presence/absence), country (US/China), treatment group (TVB-2640 50 mg and pooled placebo), and country-by-treatment interaction and with the baseline MRI-PDFF value as a covariate including all subjects from US and China will be performed. If there is no interaction detected (p≥0.05), the primary efficacy endpoint, percent change from baseline in hepatic fat fraction at Week 12, as determined by MRI PDFF, will be analyzed using an analysis of covariance (ANCOVA) model with fixed effects for the stratification factor (diabetes presence/absence) and treatment group (i.e., randomized TVB-2640 dose groups and pooled placebo) and with the baseline MRI-PDFF value as a covariate including all subjects from US and China. If the country-by-treatment interaction is significant (p<0.05) then the primary analysis will be conducted on subjects enrolled in the US and will maintain sufficient power with the 90 evaluable subjects in the US (30 active in each cohort and 30 combined placebo). A separate analysis would be performed on subjects enrolled in China in the 50 mg cohort. This analysis is not powered to detect a significant difference but would be supportive/descriptive in nature to evaluate trends. The primary efficacy analyses will be based on a F-test from the ANCOVA model in the mITT to compare each randomized TVB-2640 dose group individually with the placebo group. The fixed-sequence method, as described below, will be used to maintain an overall 0.05 Type I error rate. Data normality will be assessed and the appropriate data transformation will be applied to fulfill the underlying assumption of the ANCOVA model; if a suitable transformation cannot be determined the corresponding rank ANCOVA will become the primary analysis for this endpoint. Treatment interaction will be evaluated among the treatment groups, and if needed it will be considered supportive of the primary efficacy analyses. Summary statistics will be displayed by treatment group along with the difference in least squares means (and the associated 95% confidence interval and p-value) for each randomized TVB-2640 dose group to pooled placebo comparison. Differences between each randomized TVB-2640 dose gr...

Efficacy Analysis. The estimand for the primary efficacy analysis is defined as follows: 1. The treatment regimen for participants is: placebo or SAGE-217 for 14 days. 2. The target population is adult participants with a diagnosis with severe PPD (baseline HAM-D total score ≥26). 3. The variable of interest is the change from baseline in HAM-D total score at Day 15. 4. The population summary level deals with the difference between SAGE-217 and placebo treatments in mean change from baseline in HAM-D total score at Day 15. 5. The intercurrent events could be: a. The premature discontinuation of treatment for any reason, thus not having a Day 15 HAM-D total score available. This will be handled by a sensitivity analysis using multiple imputation technique (details will be provided in the SAP). b. Certain medications including, but not limited to, new antidepressants or benzodiazepines are prohibited in the protocol until Day 45 follow-up; however, the treatment policy strategy dictates that the results following these prohibited medication use will not be manipulated, but will rather be used ‘as is’ in analysis. Please note that this protocol does not specify any rescue medication, hence there is no rescue medication to be considered. Data will be analyzed using a mixed-effects model for repeated measures (MMRM); the model will include treatment, baseline HAM-D total score, stratification factor, assessment time point, and time point-by-treatment as explanatory variables. All explanatory variables will be treated as fixed effects. All postbaseline time points will be included in the model. Model-based point estimates (ie, least squares means, 95% confidence intervals, and p values) will be reported where applicable. An unstructured covariance structure will be used to model the within- participant errors. If there is a convergence issue with the unstructured covariance model, Toeplitz compound symmetry or Autoregressive (1) [AR(1)] covariance structure will be used, following this sequence until convergence is achieved. If the model still does not converge with AR(1) structure, no results will be reported. When the covariance structure is not UN, the sandwich estimator for the variance covariance matrix will be derived, using the EMPIRICAL option in the PROC MIXED statement in SAS. Similar to those methods described above for the primary endpoint, an MMRM will be used for the analysis of the change from baseline in other time points in HAM-D total score, MADRS total score, HA...