Optimal allocations for two treatment comparisons within the proportional odds cumulative logits model.

Szczegóły
Abstrakt

Tytuł:: Optimal allocations for two treatment comparisons within the proportional odds cumulative logits model.
Autorzy:: Moerbeek M; Department of Methodology and Statistics, Utrecht University, Utrecht, The Netherlands.
Źródło:: PloS one [PLoS One] 2021 Apr 21; Vol. 16 (4), pp. e0250119. Date of Electronic Publication: 2021 Apr 21 (Print Publication: 2021).
Typ publikacji:: Comparative Study; Journal Article
Język:: English
Imprint Name(s):: Original Publication: San Francisco, CA : Public Library of Science
MeSH Terms:: Logistic Models*
Data Interpretation, Statistical ; Humans ; Models, Statistical ; Probability ; Sample Size
References:: Stat Med. 2001 Apr 15;20(7):1089-95. (PMID: 11276038)
Stat Med. 1997 Nov 30;16(22):2599-618. (PMID: 9403959)
Clin Trials. 2005;2(2):152-62. (PMID: 16279137)
Stat Med. 2008 Jan 15;27(1):36-46. (PMID: 17634969)
Stat Med. 1987 Mar;6(2):119-25. (PMID: 3589241)
Behav Res Methods. 2011 Dec;43(4):1014-22. (PMID: 21512873)
Psychol Methods. 2011 Dec;16(4):373-90. (PMID: 22040372)
Stat Med. 1999 Jan 15;18(1):87-91. (PMID: 9990694)
Biometrics. 2017 Sep;73(3):916-926. (PMID: 28182835)
Biometrics. 1988 Sep;44(3):815-21. (PMID: 3203130)
Stat Methods Med Res. 2015 Oct;24(5):494-512. (PMID: 24030312)
Stat Med. 2018 Sep 20;37(21):3027-3046. (PMID: 29888393)
J Biopharm Stat. 2011 Jan;21(1):125-35. (PMID: 21191859)
Stat Med. 1990 Jan-Feb;9(1-2):65-71; discussion 71-2. (PMID: 2345839)
Stat Methods Med Res. 2017 Feb;26(1):399-413. (PMID: 25125453)
Clin Trials. 2018 Apr;15(2):169-177. (PMID: 29316807)
Stat Med. 2007 Aug 15;26(18):3385-97. (PMID: 17149799)
Stat Med. 1983 Jul-Sep;2(3):387-96. (PMID: 6417752)
Stat Med. 2018 Jan 15;37(1):12-27. (PMID: 28948651)
Stat Med. 1993 Dec 30;12(24):2257-71. (PMID: 8134732)
Stat Med. 2000 Mar 15;19(5):639-48. (PMID: 10700736)
Stat Methods Med Res. 2015 Oct;24(5):557-73. (PMID: 25519890)
Stat Med. 2019 Jul 20;38(16):2905-2927. (PMID: 31049999)
Biometrics. 1990 Dec;46(4):1171-8. (PMID: 2085632)
Stat Methods Med Res. 2015 Oct;24(5):574-93. (PMID: 25911332)
Stat Med. 2008 Sep 30;27(22):4581-95. (PMID: 18563794)
Stat Med. 2016 May 10;35(10):1616-36. (PMID: 26670388)
Stat Med. 2008 Jul 10;27(15):2850-64. (PMID: 17960589)
Entry Date(s):: Date Created: 20210421 Date Completed: 20210928 Latest Revision: 20210928
Update Code:: 20240104
PubMed Central ID:: PMC8059828
DOI:: 10.1371/journal.pone.0250119
PMID:: 33882086
: Czasopismo naukowe

Pełny tekst

This paper studies optimal treatment allocations for two treatment comparisons when the outcome is ordinal and analyzed by a proportional odds cumulative logits model. The variance of the treatment effect estimator is used as optimality criterion. The optimal design is sought so that this variance is minimal for a given total sample size or a given budget, meaning that the power for the test on treatment effect is maximal, or it is sought so that a required power level is achieved at a minimal total sample size or budget. Results are presented for three, five and seven ordered response categories, three treatment effect sizes and a skewed, bell-shaped or polarized distribution of the response probabilities. The optimal proportion subjects in the intervention condition decreases with the number of response categories and the costs for the intervention relative to those for the control. The relation between the optimal proportion and effect size depends on the distribution of the response probabilities. The widely used balanced design is not always the most efficient; its efficiency as compared to the optimal design decreases with increasing cost ratio. The optimal design is highly robust to misspecification of the response probabilities and treatment effect size. The optimal design methodology is illustrated using two pharmaceutical examples. A Shiny app is available to find the optimal treatment allocation, to evaluate the efficiency of the balanced design and to study the relation between budget or sample size and power.
Competing Interests: The author has declared that no competing interests exist.

Informacja