Related Experiment Video
Updated: Jun 1, 2026
Inverse Probability of Treatment Weighting (Propensity Score) using the Military Health System Data Repository and National Death Index
Published on: January 8, 2020
Potential bias in efficacy estimates due to intercurrent treatment discontinuations
Bjorn Redfors1,2, Jacky Choi1, Mario Gaudino3
1Department of Population Health Sciences, Weill Cornell Medicine, 425 East 61st Street, NewYork, NY 10065, USA.
Insights
Treatment discontinuation in randomized clinical trials (RCTs) is common and can bias results. Addressing it as an intercurrent event improves trial validity and interpretation, especially in antithrombotic studies.
Area of Science:
- Clinical Trials Methodology
- Biostatistics
- Pharmacovigilance
Background:
- Treatment discontinuation is a frequent, underrecognized issue in randomized clinical trials (RCTs) that can introduce bias into treatment effect estimates.
- Antithrombotic trials frequently experience treatment discontinuation due to bleeding, often in patients susceptible to ischemic events, highlighting the need for careful analysis.
Purpose of the Study:
- To assess the reporting and analytical handling of treatment discontinuation in antithrombotic RCTs.
- To examine the impact of discontinuation patterns on efficacy estimates using simulations under various scenarios.
- To compare treatment policy and while-on-treatment strategies within the ICH E9(R1) framework.
Main Methods:
- Review of large antithrombotic RCTs published between 2020-2025 in The New England Journal of Medicine.
- Simulation studies modeling discontinuation in treatment and control arms under different conditions (e.g., similar patterns, outcome-associated).
- Comparison of intention-to-treat (ITT) and per-protocol (PP) analyses, considering treatment policy and while-on-treatment approaches.
Main Results:
- Approximately one-third of patients discontinued treatment in trials exceeding six months, with greater imbalance in open-label studies.
- Over half of discontinuations were safety-related; however, timing and temporary interruptions were infrequently reported.
- Per-protocol analyses, assuming non-informative censoring, generally produced larger efficacy estimates than intention-to-treat analyses. Simulations revealed bias in treatment policy and while-on-treatment strategies when discontinuation was outcome-associated.
Conclusions:
- Treatment discontinuation is a common, prognostically significant event often oversimplified in analyses, potentially compromising RCT validity.
- Explicitly defining discontinuation as an intercurrent event within the estimand framework is crucial for robust trial interpretation.
- Utilizing stratified outcome curves, such as Simon-Makuch curves, can help identify prognostic discontinuation patterns and improve the interpretation of trial findings.
Aims:
Treatment discontinuation is a common intercurrent event in randomized clinical trials (RCTs) that can bias treatment effect estimates, yet it remains under recognized and inconsistently addressed. Antithrombotic trials provide a relevant case, as bleeding-related discontinuation is common and more often occur in patients at increased risk also for ischaemic events.
Methods And Results:
We reviewed large antithrombotic RCTs published in The New England Journal of Medicine (2020-2025) to assess reporting and analytic handling of treatment discontinuation. We also conducted simulations to examine how discontinuation patterns influence efficacy estimates under three scenarios: discontinuation only in the treatment arm, in both arms with similar patterns, and in both arms but outcome-associated only in the treatment arm. Treatment policy (intention-to-treat) and while-on-treatment strategies were compared within the ICH E9(R1) framework. Among fourteen reviewed antithrombotic RCTs, about one-third of patients discontinued treatment in trials with more than 6 months follow-up, with higher imbalance in open-label studies. Over half of discontinuations were safety-related, yet timing and temporary interruptions were rarely reported. Most reported both ITT and per-protocol (PP) analyses, with PP assuming non-informative censoring and yielding larger efficacy estimates. In simulations, treatment policy and while-on-treatment strategies were biased when discontinuation was outcome-associated, while Simon-Makuch curves helped identify prognostic discontinuation differing by treatment arm.
Conclusion:
Discontinuation is common, prognostically meaningful, and often oversimplified in analysis, threatening the validity of RCT findings. Explicitly addressing discontinuation as an intercurrent event within the estimand framework, along with stratified outcome curves, can improve interpretation and support more robust trial conclusions.
Related Concept Videos
01:29Bias in Epidemiological Studies
02:07Blind Procedures
01:23Types of Biopharmaceutical Studies: Controlled and Non-Controlled Approaches
Non-controlled studies, commonly employed for initial exploration, lack a control group, rendering them susceptible to biases and external influences. In contrast, controlled...
01:54The Placebo Effect
01:22Bias
In statistics, a sampling bias is created when a sample is collected from a population, and some members of the population are not as likely to be chosen as others (remember, each member...
01:11Blinding