Statistical Fragility of Findings From Randomized Phase 3 Trials in Pediatric Oncology
The statistical conclusions of pediatric oncology phase 3 trials hinge on a relatively small number and proportion of patients. Despite the sample size limitations of low prevalence diseases, pediatric cancer trials are similarly or less fragile than adult oncology trials. Smaller trials do not appear more statistically fragile than larger trials. Statistical fragility appears to have remained constant over the four decades evaluated. We recommend reporting FI or SIFI, in conjunction with p-values, for all phase 3 pediatric oncology trials.
Authors
- Steven G. DuBois (ORCID: https://orcid.org/0000-0003-0882-738X)
- Wendy B. London (ORCID: https://orcid.org/0000-0003-3571-6538)
- David S. Shulman (ORCID: https://orcid.org/0000-0002-5994-2604)
- Pei‐Chi Kao
- Hannah E. Olsen (ORCID: https://orcid.org/0000-0002-4804-5684)
- Caleb Richmond
Institutions
- Boston Children's Hospital (US)
- Harvard University (US)
- Dana-Farber/Boston Children's Cancer and Blood Disorders Center (US)
Publication Details
- Journal
- Cancer Medicine
- Published
- 2024-12-01
- DOI
- https://doi.org/10.1002/cam4.70356
- Citations
- 1
- Primary Topic
- Childhood Cancer Survivors' Quality of Life
- Type
- article
- Field-Weighted Citation Impact
- 1.11
Funders
- Conquer Cancer Foundation
- Alex's Lemonade Stand Foundation for Childhood Cancer
- Harvard University
- Harvard Catalyst