Abstract
Beta regression models have been recommended for continuous bounded outcome scores that are often collected in clinical studies. Implementing beta regression in NONMEM presents difficulties since it does not provide gamma functions required by the beta distribution density function. The objective of the study was to implement mixed-effects beta regression models in NONMEM using Nemes’ approximation to the gamma function and to evaluate the performance of the NONMEM implementation of mixed-effects beta regression in comparison to the commonly used SAS approach. Monte Carlo simulations were conducted to simulate continuous outcomes within an interval of (0, 70) based on a beta regression model in the context of Alzheimer’s disease. Six samples per subject over a 3 years period were simulated at 0, 0.5, 1, 1.5, 2, and 3 years. One thousand trials were simulated and each trial had 250 subjects. The simulation–reestimation exercise indicated that the NONMEM implementation using Laplace and Nemes’ approximations provided only slightly higher bias and relative RMSE (RRMSE) compared to the commonly used SAS approach with adaptive Gaussian quadrature and built-in gamma functions, i.e., the difference in bias and RRMSE for fixed-effect parameters, random effects on intercept, and the precision parameter were <1–3 %, while the difference in the random effects on the slope was <3–7 % under the studied simulation conditions. The mixed-effect beta regression model described the disease progression for the cognitive component of the Alzheimer’s disease assessment scale from the Alzheimer’s Disease Neuroimaging Initiative study. In conclusion, with Nemes’ approximation of the gamma function, NONMEM provided comparable estimates to those from SAS for both fixed and random-effect parameters. In addition, the NONMEM run time for the mixed beta regression models appeared to be much shorter compared to SAS, i.e., 1–2 versus 20–40 s for the model and data used in the manuscript.
References
Frisoni GB, Caroli A (2007) Neuroimaging outcomes for clinical trials. J Nutr Health Aging 11(4):348–352
Gelinas I et al (1999) Development of a functional measure for persons with Alzheimer’s disease: the disability assessment for dementia. Am J Occup Ther 53(5):471–481
Swearingen CJ et al (2011) Modeling percentage outcomes: the % Beta_Regression macro. SAS® global forum proceedings, Paper 335, pp 1–12
Cribari-Neto F, Zeileis A (2010) Beta regression in R. J Stat Softw 34(2):1–24
Ferrari SP, Cribari-Neto F (2004) Beta regression for modelling rates and proportions. J Appl Stat 31(7):799–815
Smithson M, Verkuilen J (2006) A better lemon squeezer? maximum-likelihood regression with beta-distributed dependent variables. Psychol Methods 11(1):54–71
Rogers JA et al (2011) A longitudinal dose–response model for the progression of Alzheimers disease, based on a combination of summary-level and patient-level data. In: 4th annual bayesian biostatistics conference, Houston
Rogers JA et al (2012) Combining patient-level and summary-level data for Alzheimer’s disease modeling and simulation: a beta regression meta-analysis. J Pharmacokinet Pharmacodyn 39(5):479–498
Verkuilen J, Smithson M (2012) Mixed and mixture regression models for continuous bounded responses using the beta distribution. J Educ Behav Stat 37(1):82–113
Nemes G (2010) New asymptotic expansion for the gamma function. Arch Math 95(2):161–169
Pregibon D (1980) Goodness of link tests for generalized linear models. Appl Stat 29(1):15–24
Chen KF, Wu JW (1994) Testing goodness of fit for a parametric family of link functions. J Am Stat Assoc 89(426):657–664
McCullagh P, Nelder JA (1989) Generalized linear models, monographs on statistics and applied probability, vol 37, 2nd edn. Chapman and Hall, New York, p 511
Nemes G (2008) New asymptotic expansion for the gamma function. http://dx.doi.org/10.3247/sl2math08.005
Samtani MN et al (2012) An improved model for disease progression in patients from the Alzheimer’s disease neuroimaging initiative. J Clin Pharmacol 52(5):629–644
Ito K et al (2010) Disease progression meta-analysis model in Alzheimer’s disease. Alzheimers Dement 6(1):39–53
Ito K et al (2011) Disease progression model for cognitive deterioration from Alzheimer’s disease neuroimaging initiative database. Alzheimers Dement 7(2):151–160
Espinheira PL, Ferrari SLP, Cribari-Neto F (2008) On beta regression residuals. J Appl Stat 35(4):407–419
Karlsson MO, Holford N, (2008) A tutorial on visual predictive checks,p 17, Abstr 1434 [www.page-meeting.org/?abstract=1434], PAGE
Ospina R, Ferrari SP (2010) Inflated beta distributions. Stat Pap 51:111–126
Ospina R, Ferrari SP (2012) A general class of zero-or-one inflated beta regression models. Comput Stat Data Anal 56(6):1609–1623
Hutmacher MM et al (2012) Estimating transformations for repeated measures modeling of continuous bounded outcome data. Stat Med 30(9):935–949
Ito K, Hutmacher MM, Corrigan BW (2012) Modeling of functional assessment questionnaire (FAQ) as continuous bounded data from the ADNI database. J Pharmacokinet Pharmacodyn 39(6):601–618
Molas M, Lesaffre E (2008) A comparison of three random effects approaches to analyze repeated bounded outcome scores with an application in a stroke revalidation study. Stat Med 27(30):6612–6633
Hu C et al (2011) Bounded outcome score modeling: application to treating psoriasis with ustekinumab. J Pharmacokinet Pharmacodyn 38(4):497–517
Pinheiro JC, Bates DM (1995) Approximations to the log-likelihood function in the nonlinear mixed-effects model. J Comput Graph Stat 4(1):12–35
Acknowledgments
All authors are employees of Janssen Research & Development. Steven Xu is an adjunct assistant professor in the School of Public Health at the University of Medicine and Dentistry of New Jersey. Data collection and sharing for this project was funded by the Alzheimer’s Disease Neuroimaging Initiative (ADNI) (National Institutes of Health Grant U01 AG024904). ADNI is funded by the National Institute on Aging, the National Institute of Biomedical Imaging and Bioengineering, and through generous contributions from the following: Abbott; Alzheimer’s Association; Alzheimer’s Drug Discovery Foundation; Amorfix Life Sciences Ltd.; AstraZeneca; Bayer HealthCare; BioClinica, Inc.; Biogen Idec Inc.; Bristol-Myers Squibb Company; Eisai Inc.; Elan Pharmaceuticals Inc.; Eli Lilly and Company; F. Hoffmann-La Roche Ltd and its affiliated company Genentech, Inc.; GE Healthcare; Innogenetics, N.V.; IXICO Ltd.; Janssen Alzheimer Immunotherapy Research & Development, LLC.; Johnson & Johnson Pharmaceutical Research & Development LLC.; Medpace, Inc.; Merck & Co., Inc.; Meso Scale Diagnostics, LLC.; Novartis Pharmaceuticals Corporation; Pfizer Inc.; Servier; Synarc Inc.; and Takeda Pharmaceutical Company. The Canadian Institutes of Health Research is providing funds to support ADNI clinical sites in Canada. Private sector contributions are facilitated by the Foundation for the National Institutes of Health (www.fnih.org). The grantee organization is the Northern California Institute for Research and Education, and the study is coordinated by the Alzheimer’s disease Cooperative Study at the University of California, San Diego. ADNI data are disseminated by the Laboratory for Neuro Imaging at the University of California, Los Angeles. This research was also supported by NIH Grants P30 AG010129 and K01 AG030514.
Author information
Authors and Affiliations
Consortia
Corresponding author
Additional information
The Alzheimer’s Disease Neuroimaging Initiative—Data used in preparation of this article were obtained from the Alzheimer’s disease Neuroimaging Initiative (ADNI) database (adni.loni.ucla.edu). As such, the investigators within the ADNI contributed to the design and implementation of ADNI and/or provided data but did not participate in analysis or writing of this report. A complete listing of ADNI investigators can be found at: http://adni.loni.ucla.edu/wp-content/uploads/how_to_apply/ADNI_Acknowledgement_List.pdf
Electronic supplementary material
10928_2013_9318_MOESM1_ESM.ppt
Supplementary Fig. 1. Histograms for the simulated data from a randomly selected simulation at different times when the precision parameter for the mixed-effects beta regression (τ) was set to 3 (1a), 5 (1b), and 7 (1c)
Appendix
Appendix
A sample NONMEM code for the mixed effects beta regression model for the ADNI data is presented.
Code | Comments |
---|---|
$PROB beta regression | ; Data specification |
$INPUT $DATA ADNI.csv | |
$PRED | ; Mean model |
B0 = THETA(1) + ETA(1) B1 = THETA(2) + ETA(2) | ; Define parameters for baseline and slope |
LINP = B0 + B1 * TIME MU = EXP(LINP)/(1 + EXP(LINP)) | ; Linear predictor on logit scale ; Conditional mean by anti-logit transforming the linear predictor |
LTAU = THETA(3) TAU = EXP(LTAU) | ; precision parameter |
; specify the log likelihood based on the density function for beta distribution (Eq. 2 ) | |
X1 = TAU X2 = MU * TAU X3 = (1−MU) * TAU | |
LG1 = 0.5 * (LOG(2 * 3.1415) − LOG(X1)) + X1 * (LOG(X1) − 1) + (5/4) * X1 * (LOG (1 + (1/(15 * X1 * 2)))) LG2 = 0.5 * (LOG(2 * 3.1415) − LOG(X2)) + X2 * (LOG(X2) − 1) + (5/4) * X2 * (LOG (1 + (1/(15 * X2 * 2)))) LG3 = 0.5 * (LOG(2 * 3.1415) − LOG(X3)) + X3 * (LOG(X3) − 1) + (5/4) * X3 * (LOG (1 + (1/(15 * X3 * 2)))) | ; Approximation of the log(gamma) function (Eq. 4 ) ; The first part of the log likelihood, 0.5 * (LOG(2 * 3.1415), can be omitted if computation of full likelihood is not required. ; Log Likelihood of the beta distribution (Eq. 2 ) |
LOGL = LG1 − LG2 − LG3 + (MU * TAU−1) * LOG(DV) + ((1−MU) * TAU−1) * LOG(1−DV) Y = −2 * LOGL | |
SOR = (DV − MU)/SQRT(MU * (1−MU)/(1 + TAU)) | ; Pearson residuals (standardized ordinary residuals) |
$THETA | ; specify initial values |
$OMEGA | |
$EST MAX = 9999 PRINT = 5 METHOD = COND − 2LOGLIK LAPLACIAN NOABORT $COV PRINT = E MATRIX = R $TABLE | ; estimation step |
Rights and permissions
About this article
Cite this article
Xu, X.S., Samtani, M.N., Dunne, A. et al. Mixed-effects beta regression for modeling continuous bounded outcome scores using NONMEM when data are not on the boundaries. J Pharmacokinet Pharmacodyn 40, 537–544 (2013). https://doi.org/10.1007/s10928-013-9318-0
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10928-013-9318-0