Bayesian Analysis of an Optimal Five Period Cross-over Design
Journal of Scientific Research and Reports,
A crossover design is a repeated measurements design such that each experimental unit receives different treatments during the different time periods. In a majority of bioequivalence studies, design and analysis of cross-over using classical methods such as analysis of variance (ANOVA) and test are normally associated with erroneous results. The Bayesian method is desirable in the analysis of crossover designs to eliminate errors associated with carryover effects. The objective of this study was to compare the Bayesian and the - test analysis methods on treatments and carryover effects for an optimal two treatments, five periods and four sequence C (2, 5, 4) design. The treatments and residual estimates were obtained using Best Linear Unbiased Estimation (BLUE) method. In the Bayesian method of analysis, the posterior quantities were obtained for the mean intervals of treatments and carry-over effects and the highest posterior density (HPD) graphs were plotted and interpreted using conditional probability statements. For validation purposes, the Bayesian method results were compared with the existing -tests results. From the Bayesian analysis, the probability of significant treatment difference in the presence of carryover effects was 1, while from the -test, the calculated value of 11.73 was greater than the two sided tabulated value at 95 level of significance. The two analysis methods implied significant differences in the treatment effects. In conclusion, it was established that Bayesian method of analysis can be used for bioequivalence analysis even when the carry-over effects are present and hence it is highly recommended for bioequivalence studies.
- carryover effects
- Bayesian methods
How to Cite
Reed III J. Extension of grizzle's classic cross-over design. Journal of Modern Applied Statistical Methods. 2011;10(1): 29.
Reed III J. Four period cross-over designs. Journal of Modern Applied Statistical Methods. 2012;11(1):25.
Grizzle J. The two period change-over design and it's use in clinical trials, Biometrics. 1965;1(1):467-480.
Novick M, Grizzle J. A Bayesian approach to the analysis of data from clinical trials. Journal of the American Statistical Association. 1965;60(309);81-96.
Westlake W. The design and analysis of comparative blood level trials. Swarbick J.Ed. current concepts in the pharmaceutical sciences, dosage form design and bioavailability. Philadelphia: Lea and Febier,149-179. 1973;1(1):149-179.
Westlake W. Symmetrical confidence interval for bioequivalence trials. Biometrics. 1976;1(1):741-744.
Patil V. Approximation to the Behren's fisher distributions. Biometrika. 165;52(1/2):267-271.
Metzler C. Bioavailability -A problem in Equivalence. Biometrics. 1965;1(1):309-3171.
Patil V. The Beherens Fisher Problem andits Bayesian Solution. Journal of the Indian Statistical Association. 1964;2(1):21-31.
Nyakundi CN, Koske JK, Mutiso JM, Tum IK. Mathematical Analysis of a Five periods Cross-over design for two treatments. Journal of Biostatistics and Epidemiology. 2021;7(1):48-59.
Abstract View: 64 times
PDF Download: 25 times