Meta-analysis is the process of merging summary data from related but independent studies. The main aim of meta-analysis is to assess the level of heterogeneity between studies. The degree of heterogeneity in a meta-analysis influences the difficulty of making inferences. This may be obtained by calculating the between-study variance, but the interpretation is then bound to a certain treatment effect metric. The objective of this paper is to calculate the odds ratio in clinical trials between treatment and control groups. In addition, this work focuses on the sample size of each study and the number of zeros that appear in the study arms, which are regarded dataset characteristics and contribute to the significance of this analysis. The emphasis is on analytical approaches for estimating the parameters of interest. MCMC has been used to conduct a comparative study on individual, overall odds ratio and heterogeneity for eighteen datasets using the Binomial-normal and normal-normal models through Bayesian. All of the datasets point estimates and interval estimates have been computed, and a comparative study between the balanced and imbalanced groups has been performed.

Bayesian inference, Binomial-normal, heterogeneity, imbalanced, MCMC, normal-normal, sample size, odds ratio.

Received: January 20, 2022; Revised: July 1, 2022;  Accepted: July 18, 2022; Published: August 17, 2022

How to cite this article: S. Sumathi and B. Senthil Kumar, An empirical study of odds ratio estimation employing Binomial-normal and normal-normal models, JP Journal of Biostatistics 21 (2022), 67-87. http://dx.doi.org/10.17654/0973514322021

This Open Access Article is Licensed under Creative Commons Attribution 4.0 International License

References:

[1] A. Agresti, Categorical Data Analysis, 3rd ed., Wiley, 2013.
[2] J. Bowden, J. F. Tierney, A. J. Copas and S. Burdett, Quantifying, displaying and accounting for heterogeneity in the meta-analysis of RCTs using standard and generalised Q statistics, BMC Med. Res. Methodol. 11(1) (2011), 41. doi: 10.1186/1471-2288-11-41.
[3] N. A. Card, Applied Meta-analysis for Social Science Research, Guilford Publications, New York, 2012.
[4] J. B. Carlin, Meta-analysis for tables: a Bayesian approach, Stat. Med. 11(2) (1992), 141-158.
[5] Y. Chu and Y. Yuan, A Bayesian basket trial design using a calibrated Bayesian hierarchical model, Clinical Trials 15(2) (2018), 149-158.
[6] J. Davis, K. Mengersen, S. Bennett and L. Mazerolle, Viewing systematic reviews and meta-analysis in social research through different lenses, Springer Plus 3(1) (2014), 511.
[7] B. Efron, Empirical Bayes methods for combining likelihoods, J. Amer. Statist. Assoc. 91(434) (1996), 538-550.
[8] E. A. Engels, C. H. Schmid, N. Terrin, I. Olkin and J. Lau, Heterogeneity and statistical significance in meta-analysis: an empirical study of 125 meta-analyses, Stat. Med. 19(13) (2000), 1707-1728.
[9] L. M. Friedman, C. D. Furberg, D. L. DeMets, D. M. Reboussin and C. B. Granger, Fundamentals of Clinical Trials, Springer, 2015.
[10] L. Frison and S. J. Pocock, Repeated measures in clinical trials: analysis using mean summary statistics and its implications for design, Stat. Med. 11(13) (1992), 1685-1704.
[11] A. Gelman, J. B. Carlin, H. S. Stern, D. B. Dunson, A. Vehtari and D. B. Rubin, Bayesian Data Analysis, CRC Press, 2013.
[12] P. E. Green and T. Park, Bayesian methods for contingency tables using Gibbs sampling, Statist. Papers 45(1) (2004), 33-50.
[13] J. Higgins and S. G. Thompson, Quantifying heterogeneity in a meta-analysis, Stat. Med. 21(11) (2002), 1539-1558.
[14] B. P. Hobbs and B. P. Carlin, Practical Bayesian design and analysis for drug and device clinical trials, Journal of Biopharmaceutical Statistics 18(1) (2007), 54-80.
[15] J. R. Johnson, G. Williams and R. Pazdur, End points and United States Food and Drug Administration approval of oncology drugs, Journal of Clinical Oncology 21(7) (2003), 1404-1411.
[16] D. Langan J. Higgins and M. Simmonds, Comparative Performance of Heterogeneity Variance Estimators in Meta-analysis: a Review of Simulation Studies, Research Synthesis Methods, 2016.
[17] S. Lewis and M. Clarke, Forest plots: trying to see the wood and the trees, BMJ: British Medical Journal 322 (2001), 1479.
[18] M. Khalilia, S. Chakraborty and M. Popescu, Predicting disease risks from highly imbalanced data using random forest, BMC Medical Informatics and Decision Making 11(1) (2011), 1-13.
[19] S. Piantadosi, Clinical Trials: A Methodologic Perspective, John Wiley & Sons, 2017.
[20] H. Pezeshk and J. Gittins, A fully Bayesian approach to calculating sample sizes for clinical trials with binary responses, Drug Information Journal 36(1) (2002), 143-150.
[21] R Core Team, R: A Language and Environment for Statistical Computing, R Foundation for Statistical Computing, Vienna, Austria, 2016.
[22] R. D. Riley, J. P. T. Higgins and J. J. Deeks, Interpretation of random effects meta-analyses, BMJ 342 (2011), d549. doi: 10.1136/bmj.d549.
[23] D. Sorensen and D. Gianola, Likelihood, Bayesian, and MCMC Methods in Quantitative Genetics, 2002.
[24] M. Subbiah and M. R. Srinivasan, Classification of sparse data sets with zero cells, Statist. Probab. Lett. 78(18) (2008), 3212-3215.
[25] L. Tian, T. Cai, N. Piankov, P.-Y. Cremieux and L. J. Wei, Effectively Combining Independent Tables for Valid Inferences in Meta Analysis with all Available Data but no Artificial Continuity Corrections for Studies with Zero Events and its Application to the Analysis of Rosiglitazone’s Cardiovascular Disease Related Event Data, 2007.
[26] W. Viechtbauer, Confidence intervals for the amount of heterogeneity in meta-analysis, Stat. Med. 26(1) (2007), 37-52.
[27] W. Viechtbauer, Conducting meta-analyses in r with the meta for package, Journal of Statistical Software 36(3) (2010), 1-48.
[28] Y. Wu, H. Tjelmeland and M. West, Bayesian CART: prior specification and posterior simulation, Journal of Computational and Graphical Statistics 16(1) (2007), 44-66.
[29] G. Yin, Y. Li and Y. Ji, Bayesian dose-finding in phase I/II clinical trials using toxicity and efficacy odds ratios, Biometrics 62(3) (2006), 777-787.