Combined P-Value Functions for Compatible Effect Estimation and Hypothesis Testing in Drug Regulation.
Uloženo v:
| Název: | Combined P-Value Functions for Compatible Effect Estimation and Hypothesis Testing in Drug Regulation. |
|---|---|
| Autoři: | Pawel S; Epidemiology, Biostatistics and Prevention Institute (EBPI), Center for Reproducible Science (CRS), University of Zurich, Zurich, Switzerland., Roos M; Epidemiology, Biostatistics and Prevention Institute (EBPI), Center for Reproducible Science (CRS), University of Zurich, Zurich, Switzerland., Held L; Epidemiology, Biostatistics and Prevention Institute (EBPI), Center for Reproducible Science (CRS), University of Zurich, Zurich, Switzerland. |
| Zdroj: | Statistics in medicine [Stat Med] 2025 Oct; Vol. 44 (23-24), pp. e70224. |
| Způsob vydávání: | Journal Article |
| Jazyk: | English |
| Informace o časopise: | Publisher: Wiley Country of Publication: England NLM ID: 8215016 Publication Model: Print Cited Medium: Internet ISSN: 1097-0258 (Electronic) Linking ISSN: 02776715 NLM ISO Abbreviation: Stat Med Subsets: MEDLINE |
| Imprint Name(s): | Original Publication: Chichester ; New York : Wiley, c1982- |
| Výrazy ze slovníku MeSH: | Meta-Analysis as Topic* , Clinical Trials as Topic*/statistics & numerical data , Drug and Narcotic Control*, Humans ; Confidence Intervals ; Data Interpretation, Statistical ; Models, Statistical ; Computer Simulation |
| Abstrakt: | The two-trials rule in drug regulation requires statistically significant results from two pivotal trials to demonstrate efficacy. However, it is unclear how the effect estimates from both trials should be combined to quantify the drug effect. Fixed-effect meta-analysis is commonly used but may yield confidence intervals that exclude the value of no effect even when the two-trials rule is not fulfilled. We systematically address this by recasting the two-trials rule and meta-analysis in a unified framework of combined p-value functions, where they are variants of Wilkinson's and Stouffer's combination methods, respectively. This allows us to obtain compatible combined p-values, effect estimates, and confidence intervals, which we derive in closed-form. Additionally, we provide new results for Edgington's, Fisher's, Pearson's, and Tippett's p-value combination methods. When both trials have the same true effect, all methods can consistently estimate it, although some show bias. When true effects differ, the two-trials rule and Pearson's method are conservative (converging to the less extreme effect), Fisher's and Tippett's methods are anti-conservative (converging to the more extreme effect), and Edgington's method and meta-analysis are balanced (converging to a weighted average). Notably, Edgington's confidence intervals always asymptotically include the individual trial effects, while meta-analytic confidence intervals shrink to a point at the weighted average effect. We conclude that all of these methods may be appropriate depending on the estimand of interest. We implement combined p-value function inference for two trials in the R package twotrials, allowing researchers to easily perform compatible hypothesis testing and effect estimation. (© 2025 The Author(s). Statistics in Medicine published by John Wiley & Sons Ltd.) |
| References: | Ther Innov Regul Sci. 2023 Mar;57(2):316-320. (PMID: 36289189) Stat Med. 2024 Nov 20;43(26):5023-5042. (PMID: 38573319) Pharm Stat. 2023 Jan;22(1):96-111. (PMID: 36054079) Lancet Respir Med. 2019 Mar;7(3):213-226. (PMID: 30658914) Psychol Bull. 1951 May;48(3):156-8. (PMID: 14834286) Eur Respir J. 2018 Jan 25;51(1):. (PMID: 29371393) Stat Med. 2005 Jun 15;24(11):1619-37; discussion 1639-56. (PMID: 15568207) Biostatistics. 2009 Jan;10(1):1-2. (PMID: 18550565) Control Clin Trials. 1999 Feb;20(1):16-39. (PMID: 10027498) Eur Respir J. 2018 Jan 25;51(1):. (PMID: 29371383) R Soc Open Sci. 2024 Aug 28;11(8):240149. (PMID: 39205991) J Biopharm Stat. 2002 May;12(2):107-19. (PMID: 12413234) JAMA Netw Open. 2019 Oct 2;2(10):e1912869. (PMID: 31596493) Biometrics. 1999 Dec;55(4):1286-90. (PMID: 11315085) Biom J. 2005 Apr;47(2):237-47. (PMID: 16389921) Nat Commun. 2022 Aug 31;13(1):5126. (PMID: 36045130) Food Drug Law J. 2017;72(4):595-635. (PMID: 30294197) J Am Stat Assoc. 2020;115(529):393-402. (PMID: 33012899) JAMA Netw Open. 2020 Apr 1;3(4):e203284. (PMID: 32315070) Res Synth Methods. 2010 Apr;1(2):97-111. (PMID: 26061376) Stat Med. 1999 Jul 30;18(14):1833-48. (PMID: 10407255) PLoS One. 2017 Mar 8;12(3):e0173184. (PMID: 28273140) Elife. 2017 Jan 19;6:. (PMID: 28100398) Biometrics. 1994 Dec;50(4):1029-41. (PMID: 7786985) Stat Med. 2019 Sep 20;38(21):4189-4197. (PMID: 31270842) Stat Med. 2002 Jun 30;21(12):1727-42. (PMID: 12111908) Eur Respir J. 2018 Jan 25;51(1):. (PMID: 29371384) Stat Med. 2024 Mar 15;43(6):1271-1289. (PMID: 38205556) |
| Contributed Indexing: | Keywords: confidence interval; estimand; median estimate; meta‐analysis; two‐trials rule |
| Entry Date(s): | Date Created: 20251023 Date Completed: 20251023 Latest Revision: 20251124 |
| Update Code: | 20251125 |
| PubMed Central ID: | PMC12548019 |
| DOI: | 10.1002/sim.70224 |
| PMID: | 41128100 |
| Databáze: | MEDLINE |
Full Text Finder 
Nájsť tento článok vo Web of Science | Abstrakt: | The two-trials rule in drug regulation requires statistically significant results from two pivotal trials to demonstrate efficacy. However, it is unclear how the effect estimates from both trials should be combined to quantify the drug effect. Fixed-effect meta-analysis is commonly used but may yield confidence intervals that exclude the value of no effect even when the two-trials rule is not fulfilled. We systematically address this by recasting the two-trials rule and meta-analysis in a unified framework of combined p-value functions, where they are variants of Wilkinson's and Stouffer's combination methods, respectively. This allows us to obtain compatible combined p-values, effect estimates, and confidence intervals, which we derive in closed-form. Additionally, we provide new results for Edgington's, Fisher's, Pearson's, and Tippett's p-value combination methods. When both trials have the same true effect, all methods can consistently estimate it, although some show bias. When true effects differ, the two-trials rule and Pearson's method are conservative (converging to the less extreme effect), Fisher's and Tippett's methods are anti-conservative (converging to the more extreme effect), and Edgington's method and meta-analysis are balanced (converging to a weighted average). Notably, Edgington's confidence intervals always asymptotically include the individual trial effects, while meta-analytic confidence intervals shrink to a point at the weighted average effect. We conclude that all of these methods may be appropriate depending on the estimand of interest. We implement combined p-value function inference for two trials in the R package twotrials, allowing researchers to easily perform compatible hypothesis testing and effect estimation.<br /> (© 2025 The Author(s). Statistics in Medicine published by John Wiley & Sons Ltd.) |
|---|---|
| ISSN: | 1097-0258 |
| DOI: | 10.1002/sim.70224 |