Biomedical Statistics and Informatics

| Peer-Reviewed |

A Gaussian Copula Regression Approach for Modelling Repeated Data in Medical Research

Received: 27 June 2023    Accepted: 19 July 2023    Published: 31 July 2023
Views:       Downloads:

Share This Article

Abstract

In repeated measures data, the observations tend to be correlated within each subject, and such data are often analyzed using Generalized Estimating Equations (GEE), which are robust to assumptions that many methods hold. The main limitation of GEE is that its method of estimation is quasi-likelihood. The recent framework of the copula is very popular for handling repeated data. The maximum likelihood-based analysis for repeated data can be obtained using Gaussian copula regression. The purpose of this study is to show the handling and analysis of the repeated data using the Gaussian copula regression approach and compare the findings with GEE. The prospective, double-blinded, randomized controlled trial data for this study was obtained from the Department of Anesthesia, Christian Medical College, and Vellore. ASA I and II patients were randomized into three groups. Hemodynamic parameters were obtained for 88 patients at thirteen-time points. The outcome of interest was mean arterial pressure. Both GEE and Gaussian copula regression were compared assuming four different correlation structures. The optimal correlation structures were selected with the Akaike Information Criterion (AIC) and Correlation Information Criterion (CIC) goodness of fit criteria according to the method of estimation of Gaussian copula regression and GEE, respectively. The correlation structures, unstructured and autoregressive, were found to be optimal for Gaussian copula regression and GEE based on AIC and CIC criteria values respectively. A comparison between the estimated values of the selected models showed no major differences. Gaussian copula regression found that intrathecal morphine has a significant reduction in MAP over time, this significance is considered important as the study uses randomized controlled trial data. Both methods have almost similar results, but Gaussian copula regression provides better results by identifying significant findings associated with the outcome using maximum likelihood estimation that GEE fails to identify using quasi-likelihood estimation.

DOI 10.11648/j.bsi.20230802.11
Published in Biomedical Statistics and Informatics (Volume 8, Issue 2, June 2023)
Page(s) 22-30
Creative Commons

This is an Open Access article, distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution and reproduction in any medium or format, provided the original work is properly cited.

Copyright

Copyright © The Author(s), 2024. Published by Science Publishing Group

Keywords

Correlation Structures, Gaussian Copula Regression, Generalized Estimating Equations, Repeated Data

References
[1] A D. An Introduction to Generalized Linear Models Florida: Chapman & Hall/CRC. 2002.
[2] PJ D, P H, K L, et al. Analysis of Longitudinal Data. Second Edition. Oxford:: Oxford University Press 2002.
[3] Lee U, Garcia TP, Carroll RJ, et al. Analysis of repeated measures data in nutrition research. Front Biosci (Landmark Ed) 2019; 24: 1377–89.
[4] K. Song PX. Correlated Data Analysis: Modeling, Analytics, and Applications. 2007.
[5] Lu B, Preisser JS, Qaqish BF, et al. A comparison of two bias-corrected covariance estimators for generalized estimating equations. Biometrics 2007; 63: 935–41. doi: 10.1111/j.1541-0420.2007.00764.x.
[6] Crowder M. On the Use of a Working Correlation Matrix in Using Generalised Linear Models for Repeated Measures. Biometrika 1995; 82: 407–10. doi: 10.2307/2337417.
[7] Zeger SL, Liang KY. Longitudinal data analysis for discrete and continuous outcomes. Biometrics 1986; 42: 121–30.
[8] Girden ER. ANOVA: Repeated measures: Sage. 1992.
[9] Horton NJ, Bebchuk JD, Jones CL, et al. Goodness-of-fit for GEE: an example with mental health service utilization. Stat Med 1999; 18: 213–22. doi: 10.1002/(sici)1097-0258(19990130)18:2<213::aid-sim999>3.0.co;2-e.
[10] Barnhart HX, Williamson JM. Goodness-of-fit tests for GEE modeling with binary responses. Biometrics 1998; 54: 720–9.
[11] Heagerty PJ, Zeger SL. Marginalized multilevel models and likelihood inference (with comments and a rejoinder by the authors). Statistical Science 2000; 15: 1–26. doi: 10.1214/ss/1009212671.
[12] Shen C, Weissfeld L. A copula model for repeated measurements with non-ignorable non-monotone missing outcome. Stat Med 2006; 25: 2427–40. doi: 10.1002/sim.2355.
[13] Sklar M. Fonctions de repartition an dimensions et leurs marges. Publ inst statist univ Paris. 1959.
[14] Han Z, De Oliveira V. Maximum Likelihood Estimation of Gaussian Copula Models for Geostatistical Count Data. Commun Stat Simul Comput 2020; 49: 1957–81. doi: 10.1080/03610918.2018.1508705.
[15] Ganjali M, Baghfalaki T. A Copula Approach to Joint Modeling of Longitudinal Measurements and Survival Times Using Monte Carlo Expectation-Maximization with Application to AIDS Studies. J Biopharm Stat 2015; 25: 1077–99. doi: 10.1080/10543406.2014.971584.
[16] Suresh K, Taylor JMG, Tsodikov A. A Gaussian copula approach for dynamic prediction of survival with a longitudinal biomarker. Biostatistics 2021; 22: 504–21. doi: 10.1093/biostatistics/kxz049.
[17] Kim JM, Ju H, Jung Y. Copula Approach for Developing a Biomarker Panel for Prediction of Dengue Hemorrhagic Fever. Annals of Data Science 2020; 7: 697–712. doi: 10.1007/s40745-020-00293-x.
[18] Zhang A, Fang J, Hu W, et al. A Latent Gaussian Copula Model for Mixed Data Analysis in Brain Imaging Genetics. IEEE/ACM Trans Comput Biol Bioinform 2021; 18: 1350–60. doi: 10.1109/TCBB.2019.2950904.
[19] Li J, Zhu X, Lee C-F, et al. On the aggregation of credit, market and operational risks. Rev Quant Finan Acc 2015; 44: 161–89. doi: 10.1007/s11156-013-0426-0.
[20] Masseran N. Modeling the Characteristics of Unhealthy Air Pollution Events: A Copula Approach. Int J Environ Res Public Health 2021; 18: 8751. doi: 10.3390/ijerph18168751.
[21] Nelsen RB. An Introduction to Copulas. Second Edition. USA:: Springer 2006.
[22] Masarotto G VC. Gaussian copula marginal regression. Electronic Journal of Statistics 2012; 6: 1517–49.
[23] Ferreira PH, Fiaccone RL, Lordelo JS, et al. Bivariate Copula-based Linear Mixed-effects Models: An Application to Longitudinal Child Growth Data. TEMA (São Carlos) 2019; 20: 37–59. doi: 10.5540/tema.2019.020.01.037.
[24] Cristiano Varin GM. Gaussian Copula Regression in R. Journal of Statistical Software 2017; 77: 1–26.
[25] Johnson JPJ, Arumugam R, Karuppusami R, et al. Intraoperative administration of systemic/epidural/intrathecal morphine on the quality of recovery following substitutional urethroplasty with buccal mucosal graft: A randomized control trial. Journal of Anaesthesiology Clinical Pharmacology 2022; 38: 537. doi: 10.4103/joacp.JOACP_589_20.
[26] Zeger SL, Liang KY. An overview of methods for the analysis of longitudinal data. Stat Med 1992; 11: 1825–39. doi: 10.1002/sim.4780111406.
[27] David J H, Martin J C. Practical longitudinal data analysis. 1st edn. London: Chapman and Hall: 1996.
[28] Hubbard AE, Ahern J, Fleischer NL, et al. To GEE or not to GEE: comparing population average and mixed models for estimating the associations between neighborhood risk factors and health. Epidemiology 2010; 21: 467–74. doi: 10.1097/EDE.0b013e3181caeb90.
[29] Twisk, J. Applied Longitudinal Data Analysis for Epidemiology: A Practical Guide. (2nd ed.). Cambridge:: Cambridge University Press 2013. doi: 10.1017/CBO9781139342834.
[30] Cnaan A, Laird NM, Slasor P. Using the general linear mixed model to analyse unbalanced repeated measures and longitudinal data. Stat Med 1997; 16: 2349–80. doi: 10.1002/(sici)1097-0258(19971030)16:20<2349::aid-sim667>3.0.co;2-e.
[31] Ballinger GA. Using Generalized Estimating Equations for Longitudinal Data Analysis. Organizational Research Methods 2004; 7: 127–50. doi: 10.1177/1094428104263672.
[32] Ghisletta P, Spini D. An Introduction to Generalized Estimating Equations and an Application to Assess Selectivity Effects in a Longitudinal Study on Very Old Individuals. Journal of Educational and Behavioral Statistics 2004; 29 (4): 421–37.
[33] Westgate PM, Burchett WW. A Comparison of Correlation Structure Selection Penalties for Generalized Estimating Equations. Am Stat 2017; 71: 344–53. doi: 10.1080/00031305.2016.1200490.
[34] Scott JM, deCamp A, Juraska M, et al. Finite-sample corrected generalized estimating equation of population average treatment effects in stepped wedge cluster randomized trials. Stat Methods Med Res 2017; 26: 583–97. doi: 10.1177/0962280214552092.
[35] Zeger SL, Liang KY, Albert PS. Models for longitudinal data: a generalized estimating equation approach. Biometrics 1988; 44: 1049–60.
[36] Sun J, Frees E, Rosenberg M. Heavy-tailed longitudinal data modeling using copulas. Mathematics and Economics 2008; 42: 817–30. doi: 10.1016/J.INSMATHECO.2007.09.009.
[37] Madsen L, Fang Y. Joint Regression Analysis for Discrete Longitudinal Data. Biometrics 2011; 67: 1171–5. doi: 10.1111/j.1541-0420.2010.01494.x.
[38] Escarela G, Rodríguez CE, Núñez-Antonio G. Copula modeling of receiver operating characteristic and predictiveness curves. Stat Med 2020; 39: 4252–66. doi: 10.1002/sim.8723.
[39] Jaman A, Latif MAHM, Bari W, et al. A determinant-based criterion for working correlation structure selection in generalized estimating equations. Stat Med 2016; 35: 1819–33. doi: 10.1002/sim.6821.
Cite This Article
  • APA Style

    Reka Karuppusami, Gomathi Sudhakar, Juliya Pearl Joseph Johnson, Ramamani Mariappan, Jansi Rani, et al. (2023). A Gaussian Copula Regression Approach for Modelling Repeated Data in Medical Research. Biomedical Statistics and Informatics, 8(2), 22-30. https://doi.org/10.11648/j.bsi.20230802.11

    Copy | Download

    ACS Style

    Reka Karuppusami; Gomathi Sudhakar; Juliya Pearl Joseph Johnson; Ramamani Mariappan; Jansi Rani, et al. A Gaussian Copula Regression Approach for Modelling Repeated Data in Medical Research. Biomed. Stat. Inform. 2023, 8(2), 22-30. doi: 10.11648/j.bsi.20230802.11

    Copy | Download

    AMA Style

    Reka Karuppusami, Gomathi Sudhakar, Juliya Pearl Joseph Johnson, Ramamani Mariappan, Jansi Rani, et al. A Gaussian Copula Regression Approach for Modelling Repeated Data in Medical Research. Biomed Stat Inform. 2023;8(2):22-30. doi: 10.11648/j.bsi.20230802.11

    Copy | Download

  • @article{10.11648/j.bsi.20230802.11,
      author = {Reka Karuppusami and Gomathi Sudhakar and Juliya Pearl Joseph Johnson and Ramamani Mariappan and Jansi Rani and Belavendra Antonisamy and Prasanna S. Premkumar},
      title = {A Gaussian Copula Regression Approach for Modelling Repeated Data in Medical Research},
      journal = {Biomedical Statistics and Informatics},
      volume = {8},
      number = {2},
      pages = {22-30},
      doi = {10.11648/j.bsi.20230802.11},
      url = {https://doi.org/10.11648/j.bsi.20230802.11},
      eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.bsi.20230802.11},
      abstract = {In repeated measures data, the observations tend to be correlated within each subject, and such data are often analyzed using Generalized Estimating Equations (GEE), which are robust to assumptions that many methods hold. The main limitation of GEE is that its method of estimation is quasi-likelihood. The recent framework of the copula is very popular for handling repeated data. The maximum likelihood-based analysis for repeated data can be obtained using Gaussian copula regression. The purpose of this study is to show the handling and analysis of the repeated data using the Gaussian copula regression approach and compare the findings with GEE. The prospective, double-blinded, randomized controlled trial data for this study was obtained from the Department of Anesthesia, Christian Medical College, and Vellore. ASA I and II patients were randomized into three groups. Hemodynamic parameters were obtained for 88 patients at thirteen-time points. The outcome of interest was mean arterial pressure. Both GEE and Gaussian copula regression were compared assuming four different correlation structures. The optimal correlation structures were selected with the Akaike Information Criterion (AIC) and Correlation Information Criterion (CIC) goodness of fit criteria according to the method of estimation of Gaussian copula regression and GEE, respectively. The correlation structures, unstructured and autoregressive, were found to be optimal for Gaussian copula regression and GEE based on AIC and CIC criteria values respectively. A comparison between the estimated values of the selected models showed no major differences. Gaussian copula regression found that intrathecal morphine has a significant reduction in MAP over time, this significance is considered important as the study uses randomized controlled trial data. Both methods have almost similar results, but Gaussian copula regression provides better results by identifying significant findings associated with the outcome using maximum likelihood estimation that GEE fails to identify using quasi-likelihood estimation.},
     year = {2023}
    }
    

    Copy | Download

  • TY  - JOUR
    T1  - A Gaussian Copula Regression Approach for Modelling Repeated Data in Medical Research
    AU  - Reka Karuppusami
    AU  - Gomathi Sudhakar
    AU  - Juliya Pearl Joseph Johnson
    AU  - Ramamani Mariappan
    AU  - Jansi Rani
    AU  - Belavendra Antonisamy
    AU  - Prasanna S. Premkumar
    Y1  - 2023/07/31
    PY  - 2023
    N1  - https://doi.org/10.11648/j.bsi.20230802.11
    DO  - 10.11648/j.bsi.20230802.11
    T2  - Biomedical Statistics and Informatics
    JF  - Biomedical Statistics and Informatics
    JO  - Biomedical Statistics and Informatics
    SP  - 22
    EP  - 30
    PB  - Science Publishing Group
    SN  - 2578-8728
    UR  - https://doi.org/10.11648/j.bsi.20230802.11
    AB  - In repeated measures data, the observations tend to be correlated within each subject, and such data are often analyzed using Generalized Estimating Equations (GEE), which are robust to assumptions that many methods hold. The main limitation of GEE is that its method of estimation is quasi-likelihood. The recent framework of the copula is very popular for handling repeated data. The maximum likelihood-based analysis for repeated data can be obtained using Gaussian copula regression. The purpose of this study is to show the handling and analysis of the repeated data using the Gaussian copula regression approach and compare the findings with GEE. The prospective, double-blinded, randomized controlled trial data for this study was obtained from the Department of Anesthesia, Christian Medical College, and Vellore. ASA I and II patients were randomized into three groups. Hemodynamic parameters were obtained for 88 patients at thirteen-time points. The outcome of interest was mean arterial pressure. Both GEE and Gaussian copula regression were compared assuming four different correlation structures. The optimal correlation structures were selected with the Akaike Information Criterion (AIC) and Correlation Information Criterion (CIC) goodness of fit criteria according to the method of estimation of Gaussian copula regression and GEE, respectively. The correlation structures, unstructured and autoregressive, were found to be optimal for Gaussian copula regression and GEE based on AIC and CIC criteria values respectively. A comparison between the estimated values of the selected models showed no major differences. Gaussian copula regression found that intrathecal morphine has a significant reduction in MAP over time, this significance is considered important as the study uses randomized controlled trial data. Both methods have almost similar results, but Gaussian copula regression provides better results by identifying significant findings associated with the outcome using maximum likelihood estimation that GEE fails to identify using quasi-likelihood estimation.
    VL  - 8
    IS  - 2
    ER  - 

    Copy | Download

Author Information
  • Department of Biostatistics, Christian Medical College, Vellore, India

  • Department of Biostatistics, Christian Medical College, Vellore, India

  • Departments of Anaesthesia, Christian Medical College, Vellore, India

  • Department of Neuroanaesthesia, Christian Medical College, Vellore, India

  • Department of Biostatistics, Christian Medical College, Vellore, India

  • Department of Biostatistics, Christian Medical College, Vellore, India

  • Department of Biostatistics, Christian Medical College, Vellore, India

  • Sections