Effects of simulated sample sizes on mortality estimates-Protocol for a study in 3 randomised ICU trials

Area of Science:

• Clinical Trials Methodology
• Critical Care Medicine
• Observational Studies

Background:

• A growing trend exists for clinical trials to be terminated earlier than initially planned.
• Concerns have been raised that premature trial cessation may lead to an overestimation of treatment effects.

Purpose of the Study:

• To simulate the consequences of stopping clinical trials before reaching their planned sample size.
• To assess the impact of early trial termination on mortality estimates in critical care.

Main Methods:

• Utilizing data from three international randomized clinical trials (RCTs) in critical care: the 6S, TRISS, and SUP-ICU trials.
• Estimating relative risks and 95% confidence intervals for 90-day mortality as a function of patient inclusion.
• Graphical presentation of mortality outcomes against the number of included patients.

Main Results:

• Analysis of three critical care RCTs (6S, TRISS, SUP-ICU) to assess early stopping effects.
• Estimation of relative risks for 90-day mortality based on patient enrollment.
• Graphical representation of outcome data as a function of sample size.

Conclusions:

• The study will offer crucial insights into the effects of early termination of critical care trials.
• Findings will have significant implications for patients, clinicians, researchers, and policymakers involved in critical care research and decision-making.