Types of Missingness: MCAR, MAR, and MNAR Explained
A practical guide to Missing Completely at Random (MCAR), including the formal definition, examples, diagnostics, implications for clinical trials, complete-case analysis, multiple imputation, likelihood-based methods, and how MCAR differs from MAR and MNAR.
Tipping Point Analysis for Sensitivity to Missing Data
A practical guide to tipping point analysis for sensitivity to missing data in clinical trials, including missing-not-at-random assumptions, reference-based imputation, controlled pattern-mixture models, delta adjustments, treatment effects, and R implementation.
The ICH E9(R1) Estimand Framework Explained
A practical guide to the ICH E9(R1) estimand framework, including treatment, population, variable, intercurrent events, summary measure, estimand strategies, estimators, estimates, and worked clinical trial examples.
Sensitivity Analyses for Estimands
A practical guide to sensitivity analyses for clinical trial estimands, including ICH E9(R1), intercurrent events, treatment policy, hypothetical, while-on-treatment, composite, principal stratum strategies, missing data assumptions, tipping point analyses, and worked examples.
Reference-Based Multiple Imputation Methods
A practical guide to reference-based multiple imputation methods for clinical trials, including jump to reference, copy increments in reference, copy reference, assumptions, implementation, Rubin
Pattern-Mixture Models for Missing Data
A practical guide to pattern-mixture models for missing data in clinical trials, including missingness patterns, MAR versus MNAR, pattern-specific distributions, delta adjustment, tipping-point analysis, sensitivity analysis, and implementation in R and SAS.
Multiple Imputation: A Practical Guide
A practical guide to multiple imputation for missing data, covering MCAR, MAR, and MNAR assumptions, the imputation-analysis-pooling framework, MICE, predictive mean matching, Rubin
Multiple Imputation in R: A Step-by-Step Guide
A practical step-by-step guide to multiple imputation in R using mice, including missing-data assumptions, imputation models, chained equations, diagnostics, pooling, regression analysis, sensitivity analysis, and clinical-trial examples.
Last Observation Carried Forward: Why It’s Outdated
A practical guide to why Last Observation Carried Forward (LOCF) is generally outdated for handling missing longitudinal clinical trial data, including its assumptions, biases, alternatives such as MMRM and multiple imputation, regulatory considerations, and practical R examples.
Defining Intercurrent Events in Clinical Trials
A practical guide to defining intercurrent events in clinical trials, including treatment discontinuation, rescue medication, death, treatment switching, estimands, ICE strategies, hypothetical strategies, treatment policy, composite strategies, while-on-treatment strategies, and principal stratum strategies.