Investigating started sample size, completed sample size and drop‐out rate in 10 252 phase III clinical trials: Insights from ClinicalTrials.gov
Authors:
- 20 February 2025
- journal article
- Published by Wiley in British Journal of Clinical Pharmacology
- Vol. 91 (7) , 1957-1964
Abstract
Aims: This study investigated the started sample size, completed sample size and drop‐out rate of 10 252 published and unpublished phase III clinical trials registered on ClinicalTrials.gov over the past 20 years. Methods: We conducted a comprehensive search on ClinicalTrials.gov for all phase III clinical trials with registered results before 26 May 2023. We retrieved and downloaded 10 252, which covered a period of more than 20 years. Results: Out of the 10 252 trials analysed, 889 (8.7%) were sponsored by the US National Institutes of Health/US Federal agencies (NIH/US Fed), 8429 (82.2%) were sponsored by industry. The overall median started sample size was 302. The overall median completed sample size was 228. The median completed sample sizes were 258 for industry‐sponsored trials, and 194 for NIH/US Fed‐sponsored trials. The median completed sample sizes were 321 for biological interventions and 223 for drug interventions. The overall median drop‐out rate was 11%. Statistically significant differences were observed in sample sizes between industry‐sponsored trials and NIH/US Fed‐sponsored trials ( P < 0.0001). Neither the started sample sizes nor the completed sample sizes have shown any change over the course of the past 20 years. Conclusions: Our findings indicate that a majority of the analysed trials lack sufficient statistical power (80%) to detect small effect sizes, and approximately half of the trials did not have 80% power to detect medium effect sizes. Notably, trials funded by NIH/US Fed exhibit significantly smaller sample sizes compared to trials sponsored by industry. When conducting sample size calculations for clinical trials, it is crucial to consider the anticipated effect size, variability and drop‐out rate.Keywords
This publication has 21 references indexed in Scilit:
- Novel Effect Size Interpretation Guidelines and an Evaluation of Statistical Power in Rehabilitation ResearchArchives of Physical Medicine and Rehabilitation, 2020
- Overall bias and sample sizes were unchanged in ICU trials over time: a meta-epidemiological studyJournal of Clinical Epidemiology, 2019
- Effect Size Guidelines, Sample Size Calculations, and Statistical Power in GerontologyInnovation in Aging, 2019
- The Results of Randomized Controlled Trials in Emergency Medicine Are Frequently FragileAnnals of Emergency Medicine, 2019
- The fragility of statistically significant results in otolaryngology randomized trialsAmerican Journal of Otolaryngology, 2019
- The fragility of statistically significant findings in randomised controlled anaesthesiology trials: systematic review of the medical literatureBritish Journal of Anaesthesia, 2018
- Fragility of Results in Ophthalmology Randomized Controlled TrialsOphthalmology, 2018
- The fragility of significant results underscores the need of larger randomized controlled trials in nephrologyKidney International, 2017
- Sample size calculations in surgery: Are they done correctly?Surgery, 2003
- Statistical Power and Effect Sizes of Clinical Neuropsychology ResearchJournal of Clinical and Experimental Neuropsychology, 2001