Journal List > Transl Clin Pharmacol > v.22(2) > 1082598

TOOLS
Choi, Hong, and Lee: On comparison of SAS codes with GLM and MIXED for the crossover studies with QT interval data
Original Article

Translational and Clinical Pharmacology 2014;22(2):78-82.

Published online: 30 December 2014

DOI: https://doi.org/10.12793/tcp.2014.22.2.78

On comparison of SAS codes with GLM and MIXED for the crossover studies with QT interval data

Kyungmee Choi1, ∗, Taegon Hong2, 3, Jongtae Lee2, 3

1Divison of Mathematics, College of Science and Technology, Hongik University at Sejong, Sejong 339–701, Korea

2Department of Clinical Pharmacology and Therapeutics, Seoul St. Mary's Hospital, The Catholic University of Korea, Seoul 137–701, Korea

3PIPET (Pharmacometrics Institute for Practical Education & Training), Seoul 137–701, Korea

∗Correspondence: K. M. Choi; Tel: +82-44-860-2237, Fax: +82-44-860-2648, E-mail: kmchoi@hongik.ac.kr

Received 6 November 201427 November 2014       Accepted 29 November 2014

Copyright © 2014 Korean Society for Clinical Pharmacology and Therapeutics

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

The structural complexity of crossover studies for bioequivalence test confuses analysts and leaves them a hard choice among various programs. Our study reviews PROC GLM and PROC MIXED in SAS and compares widely used SAS codes for crossover studies. PROC MIXED based on REML is more recommended since it provides best linear unbiased estimator of the random between-subject effects and its variance. Our study also considers the covariance structure within subject over period which most PK/PD studies and crossover studies ignore. The QT interval data after the administration of moxifloxacin for a fixed time point are analyzed for the comparison of representative SAS codes for crossover studies.

Keywords: QT interval, PROC GLM, PROC MIXED, Crossover studies, repeated measures

References

1. Rosner B. Fundamentals of Biostatistics. 7th ed.Cengage Learning;2000. p. 666–673.
2. Patterson S, Jones B. In: Keidling N (ed) Bioequivalence and Statistics in Clinical Pharmacology. Chapman and Hall/CRC, Boca Raton. 2006. 39–77.
3. Wallenstein S, Fisher AC. The analysis of two-period repeated measurements crossover design with application to clinical trials. Biometrics. 1977; 33:261–269.
4. Woods JR, Williams JG, Tavel M. The two-period crossover design in medical research. Ann Intern Med. 1989; 110:560–566.
crossref
5. Cheng J, Olbricht G, Gunaratna N, Kendall R, Lipka A, Paul S, et al. Mixed Models,. http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.136.77. 49&rep=rep1&type=pdf, Accessed 29 October. 2014.
6. Jennrich RI, Schluchter MD. Unbalanced repeated-measures models with structured covariance matrices. Biometrics. 1986; 42:805–820.
crossref
7. Wolfinger RM, Chang M. Comparing the SAS® GLM and MIXED Procedures for Repeated Measures,. http://www.ats.ucla.edu/stat/sas/library/mixedglm.pdf. Accessed 29 October. 2014.
8. Montgomery DC. Design and Analysis of Experiment, John Wiley & Sons. 1984. 71.
9. Park SH. Design of Experiments, Min-Young-SA. 2006. 58.
10. Meyers LS, Gamst G, Guarino AJ. Applied Multivariate Research, Design and Interpretation, Sage. 2006. 279–314.
11. Owen JS, Fiedler-Kelly J. Introduction to Population Pharmacokinetic/Pharmacodynamic Analysis with Nonlinear Mixed Effects Models, Wiley, New York. 2014. 9–29.
12. Brunelle R. Review various methods to perform the analysis of a 2 treatment, 2 period crossover study. http://www.math.iupui.edu/∼indyasa/crosover.pdf. Accessed 29 October. 2014.
13. SAS. 2014. ).,. http://support.sas.com/onlinedoc/913/getDoc/en/statug.hlp/glm_sect34.htm. Accessed 29 October 2014.
14. Saarinen F. Using mixed models in a crossover study with repeated measurements within periods,. http://www2.math.su.se/matstat/reports/se-rieb/2004/rep22/report.pdf. Accessed 29 October. 2014.
15. OehLert GW. A few words about REML. http://users.stat.umn.edu/∼corbett/classes/5303/REML.pdf. Accessed 29 October. 2014.
16. Pinheiro CJ, Bates DM. Mixed-Effects Models in S and S-PLUS. Springer;2004. p. 57–82.
17. Kiernan K, Tao J, Gibbs P. Tips and strategies for mixed modeling with SAS/Stat procedures, SAS global forum 2012,. http://support.sas.com/resources/papers/proceedings12/332–2012.pdf. Accessed 29 October. 2014.
18. Frison L, Pocock SJ. Repeated measures in clinical trials: Analysis using mean summary statistics and its implications for design. Stat Med. 1992; 11:1685–1704.
crossref
19. Sheiner LB, Beal SL. Evaluation of methods for estimating population pharmacokinetic parameters III. Monoexponential model: routine clinical pharmacokinetic data. J Pharmacokinet Biopharm. 1983; 11:303–319.
crossref
20. Mauchly JW. Significance test for sphericity of a normal n-variate distribution. Annals of Mathematical Statistics. 1940; 11:204–209.
crossref
21. Anderson TW. An Introduction to Multivariate Statistical Analysis. New York: John Wiley & Sons, Inc.;1984. p. 440–442.
22. Huynh H, Feldt LS. Conditions under Which Mean Square Ratios in Repeated Measurements Designs Have Exact F-Distribution. Journal of the American Statistical Association. 1970; 65:1582–1589.
23. Sakamoto Y, Ishiguro M, Kitagawa G. Akaike information criterion statistics. Psychometrika. 1986; 54:539–541.
24. Schwarz G. Estimating the dimension of a model. Ann Statistics. 1978; 6:461–464.
crossref

Table 1.
AIC and BIC for programs
Program AIC BIC
2 179.7 181.2
3 177.7 178.7
5 with cs 177.7 178.7
5 with un 181.9 184.8

AIC, information criteria of Akaike; BIC, information criteria of Schwar

Table 2.
Type 3 tests of fixed effects
Effect Numertor DF Denomenator DF F P-value
QTcFb 1 16 3.79 0.0694
seq 5 5 1.08 0.4674
period 2 16 0.78 0.4766
Trt 2 16 33.18 <0.0001

DF, degree of freedom

Table 3.
Differences of least squares means
TRT TRT Estimate SE DF t Adjusted p Upper CI Lower CI
A B –15.0223 3.0916 16 –4.86 0.0005 –20.4199 –9.6247
A C –25.0443 3.0927 16 –8.10 <0.0001 –30.4438 –19.6448
B C –10.0220 3.0761 16 –3.26 0.0129 –15.3925 –4.6514

SE, standard error; DF, degree of freedom; CI, confidence interval

TOOLS
Similar articles