In medical sciences, we often encounter longitudinal temporal relationships that are non-linear in nature. The influence of risk factors may also change across longitudinal follow-up. A system of multiphase non-linear mixed effects model is presented to model temporal patterns of longitudinal continuous measurements, with temporal decomposition to identify the phases and risk factors within each phase. Application of this model is illustrated using spirometry data after lung transplantation using readily available statistical software. This application illustrates the usefulness of our flexible model when dealing with complex non-linear patterns and time-varying coefficients.
BanburyMKCosgroveDMThomasJD. Hemodynamic stability during 17 Years of the Carpentier-Edwards aortic pericardial bioprosthesis. Ann Thorac Surg2002; 73: 1460–1465.
2.
DigglePJHeagertyPJLiangK. Analysis of longitudinal data, 2nd ed. New York: Oxford University press, 2002.
3.
LairdNMWareJH. Random-effects models for longitudinal data. Biometrics1982; 38: 963–974.
4.
LeeYNelderJA. Conditional and marginal models: another view. Stat Sci2004; 19: 219–238.
McCullochCE. Maximum likelihood algorithm for generalized linear mixed models. J Am Stat Assoc1997; 92: 162–170.
7.
WuHDingA. Population HIV-1 dynamics in vivo: applicable models and inferential tools for virological data from AIDS clinical trials. Biometrics1999; 55: 410–418.
8.
SheinerLBBealSL. Evaluation of methods for estimating population pharmacokinetic parameters. I.Michaelis-Menten Model: Routine clinical pharmacokinetic data. J Pharm Biopharm1980; 8: 553–571.
9.
DavidianMGiltinanDM. Nonlinear models for repeated measurement data, New York: Chapman & Hall, 1995.
10.
VoneshEFChinchilliVG. Linear and nonlinear models for the analysis of repeated measurements, London: Chapman and Hall, 1997.
11.
PinheiroJCBatesDM. Approximations to the log-likelihood function in a nonlinear mixed-effects model. J Comput Graph Stat1995; 4: 12–35.
12.
VoneshEF. A note on the use of Laplace’s approximation for nonlinear mixed-effects models. Biometrika1996; 83: 447–452.
13.
BlackstoneEHNaftelDCTurnerMEJr. The decomposition of time-varying hazard into phases, each incorporating a separate stream of concomitant information. J Am Stat Assoc1986; 81: 615–624.
14.
GelmanABoisFJiangLM. Physiology pharmacokinetic analysis using population modeling and informative prior distributions. J Am Stat Assoc1996; 91: 1400–1412.
15.
MullerPRosnerGL. A Bayesian population model with hierarchical mixture prior applied to blood count data. J Am Stat Assoc1997; 92: 1279–1292.
16.
MikulichSKZerbeGOJonesRH. Comparing linear and nonlinear mixed model approaches to Cosinor analysis. Stat Med2003; 22: 3195–3211.
17.
WuL. A joint model for nonlinear mixed-effects models with censoring and covariates measured with error, with application to AIDS studies. J Am Stat Assoc2002; 97: 955–964.
18.
FahrmeirLKneibTLangS. Penalized structured additive regression for space time data: a Bayesian perspective. Stat Sinica2004; 14: 731–761.
19.
WuCOYuKF. Nonparametric varying-coefficient models for the analyses of longitudinal data. Int Stat Rev2000; 3: 373–393.
20.
WuHZhangJT. Nonparametric regression methods for longitudinal data analysis, New Jersey: John Wiley & Sons, 2006.
21.
HazelrigJBTurnerMEJrBlackstoneEH. Parametric survival analysis combining longitudinal and cross-sectional censored and interval-censored data with concomitant information. Biometrics1982; 39: 1–5.
22.
WolfingerRD. Laplace approximation for non-linear mixed models. Biomatrika1993; 80: 791–795.
23.
SundaresanRSShiraishiYTrulockEP. Single or bilateral lung transplantation for emphysema?J Thorac Cardiovasc Surg1996; 112: 1485–1494.
24.
HankinsonJLOdencrantzJRFedanKB. Spirometry reference values from a sample of the general U.S. population. Am J Respir Crit Care Med1999; 159: 179–187.
25.
Akaike H. Information theory as an extension of the maximum likelihood principle. In: Second international symposium on information theory, Budapest, 1973, pp. 267–281.
26.
VoneshEFChincillVMPuK. Goodness-of-fit in generalized nonlinear mixed-effects models. Biometrics1996; 52: 572–587.
27.
LinLI. A concordance correlation coefficient to evaluate reproducibility. Biometrics1989; 45: 255–268.
28.
CantoniEFieldCFlemmingJM. Longitudinal variable selection by cross-validation in the case of many covariates. Stat Med2007; 26: 919–930.
29.
PinheiroJCChaoEC. Efficient Laplacian and adaptive Gaussian quadrature algorithms for multilevel generalized linear mixed models. J Comput Graph Stat2006; 15: 58–81.
30.
Harry Joe. Accuracy of Laplace approximation for discrete response mixed models. Comput Data Anal2008; 52: 5066–5074.
31.
FieuwsSVerbekeGMollenbergsG. Random-effects models for multivariate repeated measures. Stat Methods Med Res2007; 16: 387–397.
32.
VerbekeGFieuwsSMollenbergsG. The analysis of multivariate longitudinal data: a review. Stat Methods Med Res2014; 23: 42–59.
33.
FieuwsSVerbekeG. Joint modelling of multivariate longitudinal profiles: pitfalls of the random-effects approach. Stat Med2004; 23: 3093–3104.
34.
GueorguievaR. A multivariate generalized linear mixed model for joint modelling of clustered outcomes in the exponential family. Stat Model2001; 1: 177–193.
35.
GueorguievaRSanacoraG. Joint analysis of repeatedly observed continuous and ordinal measures of disease severity. Stat Med2006; 25: 1307–1322.
36.
FitzmauriceGDavidianMVerbekeG. Longitudinal data analysis, Boca Raton, FL: Chapman & Hall/CRC Taylor & Francis Group, 2009, pp. 317–391.
