Related Experiment Video
Updated: Jun 18, 2025
05:37
An R-Based Landscape Validation of a Competing Risk Model
Published on: September 16, 2022
2.0K
Comparison between asymptotic and re-randomisation tests under non-proportional hazards in a randomised controlled
Ryusei Kimura1,2, Shogo Nomura3, Kengo Nagashima4
1Biostatistics Unit, Clinical and Translational Research Center, Keio University Hospital, Tokyo, 160-8582, Japan. kimura.ryusei@keio.jp.
BMC Medical Research Methodology
|July 30, 2024
Summary
Re-randomisation tests, particularly the stratified MaxCombo test, offer robust power for analysing survival data in randomised controlled trials with minimisation, especially when non-proportional hazards are present.
Area of Science:
- Biostatistics
- Clinical Trials
- Survival Analysis
Background:
- Pocock-Simon's minimisation is a common method for balancing treatment assignments in randomised controlled trials (RCTs).
- Traditional asymptotic tests for survival outcomes can be conservative or have inflated type I error rates, especially in small samples or with non-proportional hazards (non-PH).
- Existing re-randomisation tests are limited in scenarios with non-PH, potentially reducing statistical power.
Purpose of the Study:
- To propose and evaluate novel re-randomisation tests for RCTs using minimisation, specifically addressing non-PH scenarios.
- To compare the performance of these new tests against existing asymptotic and re-randomisation methods.
- To assess the statistical power and type I error rates under various non-PH conditions.
Main Methods:
- Developed two re-randomisation tests: a maximum combination of weighted log-rank tests (MaxCombo) and difference in restricted mean survival time (dRMST).
- Compared these tests with log-rank and Cox PH models using simulated data with non-PH (delayed, crossing, diminishing effects).
- Evaluated performance across different sample sizes (50, 100, 500) and allocation ratios (1:1).
Main Results:
- Re-randomisation tests maintained nominal type I error rates across null scenarios.
- Unadjusted asymptotic tests were overly conservative; adjusted asymptotic tests (Cox PH, dRMST) showed inflated type I error rates with small sample sizes (n=50).
- The stratified MaxCombo-based re-randomisation test demonstrated consistently robust statistical power in all tested scenarios.
Conclusions:
- Re-randomisation tests are a valuable alternative for RCTs employing minimisation, particularly in the presence of non-PH.
- The stratified MaxCombo test offers superior and robust power across diverse non-PH situations.
- These findings support the use of advanced re-randomisation techniques for more reliable survival analysis in complex trial designs.
Keywords:
MaxCombo testMinimisationNon-proportional hazardsRe-randomisation testRestricted mean survival timeSurvival analysisWeighted log-rank testMore Related Videos
Related Concept Videos
01:20Comparing the Survival Analysis of Two or More Groups
168
Survival analysis is a cornerstone of medical research, used to evaluate the time until an event of interest occurs, such as death, disease recurrence, or recovery. Unlike standard statistical methods, survival analysis is particularly adept at handling censored data—instances where the event has not occurred for some participants by the end of the study or remains unobserved. To address these unique challenges, specialized techniques like the Kaplan-Meier estimator, log-rank test, and...
168
01:19The Mantel-Cox Log-Rank Test
336
The Mantel-Cox log-rank test is a widely used statistical method for comparing the survival distributions of two groups. It tests whether a statistically significant difference exists in survival times between the groups without assuming a specific distribution for the survival data, making it a non-parametric test. This flexibility makes the log-rank test particularly valuable in medical research and other fields where the timing of an event, such as death or disease recurrence, is of...
336
01:13Randomized Experiments
6.8K
The randomization process involves assigning study participants randomly to experimental or control groups based on their probability of being equally assigned. Randomization is meant to eliminate selection bias and balance known and unknown confounding factors so that the control group is similar to the treatment group as much as possible. A computer program and a random number generator can be used to assign participants to groups in a way that minimizes bias.
Simple randomization
Simple...
Simple randomization
Simple...
6.8K
01:23Types of Biopharmaceutical Studies: Controlled and Non-Controlled Approaches
124
Biopharmaceutical studies constitute a vital field aiming to enhance drug delivery methods and refine therapeutic approaches, drawing upon diverse interdisciplinary knowledge. In research methodologies, the choice between controlled and non-controlled studies significantly influences the study's reliability and accuracy.
Non-controlled studies, commonly employed for initial exploration, lack a control group, rendering them susceptible to biases and external influences. In contrast,...
Non-controlled studies, commonly employed for initial exploration, lack a control group, rendering them susceptible to biases and external influences. In contrast,...
124
01:15Assumptions of Survival Analysis
117
Survival models analyze the time until one or more events occur, such as death in biological organisms or failure in mechanical systems. These models are widely used across fields like medicine, biology, engineering, and public health to study time-to-event phenomena. To ensure accurate results, survival analysis relies on key assumptions and careful study design.
117
01:24Testing a Claim about Population Proportion
3.3K
A complete procedure for testing a claim about a population proportion is provided here.
There are two methods of testing a claim about a population proportion: (1) Using the sample proportion from the data where a binomial distribution is approximated to the normal distribution and (2) Using the binomial probabilities calculated from the data.
The first method uses normal distribution as an approximation to the binomial distribution. The requirements are as follows: sample size is large...
There are two methods of testing a claim about a population proportion: (1) Using the sample proportion from the data where a binomial distribution is approximated to the normal distribution and (2) Using the binomial probabilities calculated from the data.
The first method uses normal distribution as an approximation to the binomial distribution. The requirements are as follows: sample size is large...
3.3K