This page contain links to blog posts, papers, presentations and software on the topic of ‘estimands’ in clinical trials. The estimand refers to how the treatment effect in a clinical trial is defined and quantified in light of so called intercurrent events which may occur. These are events which affect the existence or interpretation of patient outcomes after they occur. For further details on the estimand framework, see the ICH E9 addendum on estimands, linked to below.
Our work in this area is funded by a UK Medical Research Council grant (MR/T023953/1)
ICH E9 addendum on estimands
Blog posts
- Estimating hypothetical estimands with causal inference and missing data estimators in a diabetes trial
- Is the ICH E9 estimand addendum compatible with model-based estimands?
- G-formula for causal inference via multiple imputation
- Causal (in)validity of the trimmed means estimand
- Hypothetical estimands – a unification of causal inference and missing data methods
- Comments on FDA guidance ‘Adjusting for Covariates in Randomized Clinical Trials for Drugs and Biological Products’
Papers
- Estimating hypothetical estimands with causal inference and missing data estimators in a diabetes trial – arXiv preprint
- G-formula for causal inference via multiple imputation – arXiv preprint
- Hypothetical estimands in clinical trials: a unification of causal inference and missing data methods – Statistics in Biopharmaceutical Research 2022
Talks
- Hypothetical estimands – a unification of causal inference and missing data methods – Australian Trials Methodology Conference 2023
- Hypothetical estimands – a unification of causal inference and missing data methods – Basel Biometric Society event – Addressing intercurrent events: Treatment policy and hypothetical strategies 2022
- An introduction to covariate adjustment in trials – PSI Webinar: Covariate Adjustment – Considerations and Examples 2022
- Causal (in)validity of the trimmed means estimand in clinical trials – International Biometric Society conference 2022
- Hypothetical estimands in clinical trials – a unification of causal inference and missing data methods – Danish Society for Biopharmaceutical Statistics 2022
- Hypothetical estimands in clinical trials – a unification of causal inference and missing data methods – European Statistical Forum 2021
Software
- gFormulaMI – an R package for implementing G-formula by leveraging multiple imputation methods and software