In empirical studies, multiple outcomes are often measured repeatedly over time, and interest frequently lies in studying the association between these longitudinal outcomes and a time-to-event outcome. Therefore, shared-parameter joint models for longitudinal and time-to-event outcomes have been developed. However, while such joint models in theory also allow for multiple longitudinal outcomes, they are often restricted to a limited number of outcomes due to computational complexity when fitting the models. To address this problem, we propose a new joint model, which is based on correlated instead of shared random effects, and for which a pairwise-modelling strategy can be used. In this approach, the longitudinal outcomes are modelled with (generalized) linear mixed models and the survival outcome with a Weibull proportional hazards frailty model. Instead of fitting the full joint model, this approach involves fitting all possible bivariate models, and inference is based on pseudo-likelihood theory. The main advantage of our approach is that there is no restriction on the number of longitudinally measured outcomes that are jointly modelled with the time-to-event outcome.
AertsM, MolenberghsG, RyanLM and GeysH (2002) Topics in modelling of clustered data. Chapman and Hall/CRC.
2.
AlbertPS and ShihJH (2010) An approach for jointly modeling multivariate longitudinal measurements and discrete time-to-event data. The Annals of Applied Statistics, 4, 1517.
3.
ArnoldBC and StraussD (1991) Pseudolikelihood estimation: some examples. Sankhyā: The Indian Journal of Statistics, Series B, 233–43.
4.
CollettD (2023) Modelling survival data in medical research. CRC press.
5.
CrowtherMJ, AbramsKR and LambertPC (2013) Joint modeling of longitudinal and survival data. The Stata Journal, 13, 165–84.
6.
DuchateauL and JanssenP (2008) The frailty model. Springer.
7.
ElbersC and RidderG (1982) True and spurious duration dependence: the identifiability of the proportional hazard model. The Review of Economic Studies, 49, 403–09.
8.
FeddagM-L and BacciS (2009) Pairwise likelihood for the longitudinal mixed rasch model. Computational Statistics & Data Analysis, 53, 1027–37.
9.
FieuwsS and VerbekeG (2004) Joint modelling of multivariate longitudinal profiles: pitfalls of the random-effects approach. Statistics in Medicine, 23, 3093–3104.
10.
FieuwsS and VerbekeG (2005) Evaluation of the pairwise approach for fitting joint linear mixed models: a simulation study. Technical report, Technical Report TR0527, Biostatistical Centre, Katholieke Universiteit.
11.
FieuwsS and VerbekeG (2006) Pairwise fitting of mixed models for the joint modeling of multivariate longitudinal profiles. Biometrics, 62, 424–31.
12.
FieuwsS, VerbekeG, BoenF and DelecluseC (2006) High dimensional multivariate mixed models for binary questionnaire data. Journal of the Royal Statistical Society Series C: Applied Statistics, 55, 449–60.
13.
FieuwsS, VerbekeG and MolenberghsG (2007) Random-effects models for multivariate repeated measures. Statistical Methods in Medical Research, 16, 387–97.
14.
GeysH, MolenberghsG and RyanLM (1997) Pseudo-likelihood inference for clustered binary data. Communications in Statistics-Theory and Methods, 26, 2743–67.
15.
GeysH, MolenberghsG and RyanLM (1999) Pseudolikelihood modeling of multivariate outcomes in developmental toxicology. Journal of the American Statistical Association, 94, 734–45.
16.
HermansL, NassiriV, MolenberghsG, KenwardMG, Van der ElstW, AertsM and VerbekeG (2018) Clusters with unequal size: maximum likelihood versus weighted estimation in large samples. Statistica Sinica, 28, 1107–32.
17.
HickeyGL, PhilipsonP, JorgensenA and Kolamunnage-DonaR (2018) joinerml: a joint model and software package for time-to-event and multivariate longitudinal outcomes. BMC Medical Research Methodology, 18, 1–14.
18.
HuangX, LiG, ElashoffRM and PanJ (2011) A general joint model for longitudinal measurements and competing risks survival data with heterogeneous random effects. Lifetime Data Analysis, 17, 80–100.
19.
IvanovaA, MolenberghsG and VerbekeG (2016) Mixed models approaches for joint modeling of different types of responses. Journal of Biopharmaceutical Statistics, 26, 601–18.
20.
LairdNM and WareJH (1982) Random-effects models for longitudinal data. Biometrics, 963–74.
21.
Lawrence GouldA, BoyeME, CrowtherMJ, IbrahimJG, QuarteyG, MicallefS and BoisFY (2015) Joint modeling of survival and longitudinal non-survival data: current methods and issues. Report of the Dia Bayesian joint modeling working group. Statistics in Medicine, 34, 2181–95.
22.
LiN, ElashoffRM and LiG (2009) Robust joint modeling of longitudinal measurements and competing risks failure time data. Biometrical Journal: Journal of Mathematical Methods in Biosciences, 51, 19–30.
23.
LipsitzSR, FitzmauriceGM, IbrahimJG, GelberR and LipshultzS (2002) Parameter estimation in longitudinal studies with outcome-dependent follow-up. Biometrics, 58, 621–30.
24.
LittleRJ and RubinDB (2019) Statistical analysis with missing data, 793. John Wiley & Sons.
25.
MauffK, SteyerbergE, KardysI, BoersmaE and RizopoulosD (2020) Joint models with multiple longitudinal outcomes and a time-to-event outcome: a corrected two-stage approach. Statistics and Computing, 30, 999–1014.
26.
MauffK, ErlerNS, KardysI and RizopoulosD (2021) Pairwise estimation of multivariate longitudinal outcomes in a Bayesian setting with extensions to the joint model. Statistical Modelling, 21, 115–36.
27.
MolenberghsG and VerbekeG (2005) Models for discrete longitudinal data. ISBN 978-0-387-25144-8. doi: 10.1007/0-387-28980-1.
28.
MolenberghsG, VerbekeG, DemétrioCG and VieiraAM (2010) A family of generalized linear models for repeated measures with normal and conjugate random effects. Statistical Science, 25, 325–47.
29.
MolenberghsG, KenwardMG, VerbekeG and BirhanuT (2011) Pseudo-likelihood estimation for incomplete data. Statistica Sinica, 187–206.
30.
MolenberghsG, VerbekeG, EfendiA, BraekersR and DemétrioCG (2015) A combined gamma frailty and normal random-effects model for repeated, overdispersed time-to-event data. Statistical Methods in Medical Research, 24, 434–52.
31.
MolyneauxPL, Willis-OwenSA, CoxMJ, JamesP, CowmanS, LoebingerM, BlanchardA, EdwardsLM, StockC, DaccordC, . (2017). Host–microbial interactions in idiopathic pulmonary fibrosis. American Journal of Respiratory and Critical Care Medicine, 195, 1640–50.
32.
MurtaughPA, DicksonER, Van DamGM, MalinchocM, GrambschPM, LangworthyAL and GipsCH (1994) Primary biliary cirrhosis: prediction of short-term survival based on repeated patient visits. Hepatology, 20, 126–34.
33.
NelsonKP, LipsitzSR, FitzmauriceGM, IbrahimJ, ParzenM and StrawdermanR (2006) Use of the probability integral transformation to fit nonlinear mixed-effects models with nonnormal random effects. Journal of Computational and Graphical Statistics, 15, 39–57.
34.
NguyenHT, VasconcellosHD, KeckK, ReisJP, LewisCE, SidneyS, Lloyd-JonesDM, SchreinerPJ, GuallarE, WuCO, . (2023) Multivariate longitudinal data for survival analysis of cardiovascular event prediction in young adults: insights from a comparative explainable study. BMC Medical Research Methodology, 23, 23.
35.
PapageorgiouG, MauffK, TomerA and RizopoulosD (2019) An overview of joint modeling of time-to-event and longitudinal outcomes. Annual Review of Statistics and its Application, 6, 223–40.
36.
RenardD, MolenberghsG and GeysH (2004) A pairwise likelihood approach to estimation in multilevel probit models. Computational Statistics & Data Analysis, 44, 649–67.
37.
RizopoulosD (2012) Joint models for longitudinal and time-to-event data: with applications in R. CRC Press.
38.
RizopoulosD (2016) The R package JMbayes for fitting joint models for longitudinal and time-to-event data using mcmc. Journal of Statistical Software, 72, 1–45. doi: 10.18637/ jss.v072.i07.
39.
RousseeuwPJ and MolenberghsG (1993) Transformation of nonpositive semidefinite correlation matrices. Communications in Statistics–Theory and Methods, 22, 965–84.
40.
SchwartzD, HelmersR, GalvinJ, Van FossenD, FreesK, DaytonC, BurmeisterL and HunninghakeG (1994) Determinants of survival in idiopathic pulmonary fibrosis. American Journal of Respiratory and Critical Care Medicine, 149, 450–54.
41.
SunJ, Herazo-MayaJD, MolyneauxPL, MaherTM, KaminskiN and ZhaoH (2019) Regularized latent class model for joint analysis of high-dimensional longitudinal biomarkers and a time-to-event outcome. Biometrics, 75, 69–77.
42.
TherneauT and GrambschP (2013) Modeling survival data: extending the cox model. Statistics for Biology and Health. Springer New York. ISBN 9781475732948. URLhttps://books.google.be/books?id=oj0mBQAAQBAJ.
43.
TibaldiF, AbrahantesJC, MolenberghsG, RenardD, BurzykowskiT, BuyseM, ParmarM, StijnenT and WolfingerR (2003) Simplified hierarchical linear models for the evaluation of surrogate endpoints. Journal of Statistical Computation and Simulation, 73, 643–58.
44.
TsiatisAA and DavidianM (2004) Joint modeling of longitudinal and time-to-event data: an overview. Statistica Sinica, 809–34.
45.
VarinC, ReidN and FirthD (2011) An overview of composite likelihood methods. Statistica Sinica, 5–42.
46.
VerbekeG and MolenberghsG (2000) Linear mixed models for longitudinal data. Springer.
47.
VerbekeG, FieuwsS, MolenberghsG and DavidianM (2014) The analysis of multivariate longitudinal data: a review. Statistical Methods in Medical Research, 23, 42–59.
48.
WuMC and CarrollRJ (1988) Estimation and comparison of changes in the presence of informative right censoring by modeling the censoring process. Biometrics, 175–88.
