This tutorial article demonstrates how time-to-event data can be modelled in a
very flexible way by taking advantage of advanced inference methods that have
recently been developed for generalized additive mixed models. In particular, we
describe the necessary pre-processing steps for transforming such data into a
suitable format and show how a variety of effects, including a smooth nonlinear
baseline hazard, and potentially nonlinear and nonlinearly time-varying effects,
can be estimated and interpreted. We also present useful graphical tools for
model evaluation and interpretation of the estimated effects. Throughout, we
demonstrate this approach using various application examples. The article is
accompanied by a new R-package called
pammtools implementing all of the tools described
here.
AndersenPKBorganOGillR and KeidingN (1992)
Statistical Models Based on Counting
Processes. Berlin and New York,
NY:
Springer-Verlag.
2.
ArgyropoulosC and UnruhML (2015)
Analysis of time to event outcomes in randomized controlled
trials by generalized additive models. PLoS
ONE, 10, e0123784.
doi: 10.1371/journal.pone.0123784
3.
BatesDMächlerMBolkerB and WalkerS (2015)
Fitting linear mixed-effects models using
lme4. Journal of Statistical
Software, 67,
1–48. doi:
10.18637/jss.v067.i01
4.
BenderA and ScheiplF (2017,
November 14). adibender/pammtools: v0.0.3.2 (Version
v0.0.3.2). Zenodo. URL
http://doi.org/10.5281/zenodo.1048832
5.
BenderAGrollA and ScheiplF (2018,
January 14). adibender/pammtutorial-smj: Release v1.0.1
(Version v1.0.1). Zenodo. URL
http://doi.org/10.5281/zenodo.1147058
6.
BenderAScheiplFKüchenhoffHDayAG and HartlW (2016)
Modeling exposure-lag-response associations with
penalized piece-wise exponential models (Technical report
108). Ludwig-Maximilians-University URL
https://epub.ub.uni-muenchen.de/32010/.
7.
BergerM and SchmidM (2018)
Semiparametric regression for discrete time-to-event
data. Statistical
Modelling, 18322–345.
8.
ClaytonDG (1983)
Fitting a general family of failure-time distributions using
GLIM. Journal of the Royal Statistical
Society. Series C (Applied Statistics),
32, 102–109. doi:
10.2307/2347288
9.
CoxDR (1972)
Regression models and life tables (with
discussion). Journal of the Royal
Statistical Society, B 34,
187–220.
10.
DemarquiFNLoschiRH and
ColosimoEA (2008)
Estimating the grid of time-points for the piecewise
exponential model. Lifetime Data
Analysis, 14,
333–356. doi:
10.1007/s10985-008-9086-0
11.
EilersPHC (1998)
Hazard smoothing with B-splines.
Proceedings of the 13th International Workshop on
Statistical Modelling, New Orleans, La,
200–207.
12.
EilersPHC and MarxBD (1996)
Flexible smoothing with B-splines and
penalties. Statistical
Science, 11,
89–121. doi:
10.1214/ss/1038425655
13.
FoxJ and
WeisbergHS (2011)
An R Companion to Applied Regression.
Thousand Oaks, CA: SAGE. ISBN
978-1-4129-7514-8.
14.
FriedmanM (1982)
Piecewise exponential models for survival data with
covariates. The Annals of
Statistics, 10,
101–113.
15.
FriedmanJHastieT and
TibshiraniR (2010)
Regularization paths for generalized linear models via
coordinate descent. Journal of Statistical
Software, 33,
1–22.
16.
FrumentoP (2016)
pch: Piecewise constant hazards models for censored and
truncated data. R package version 1.3. URL
https://CRAN.R-project.org/package=pch
17.
GasparriniAScheiplFArmstrongB and KenwardMG (2017)
A penalized framework for distributed lag non-linear
models. Biometrics.
doi: 10.1111/biom.12645
18.
GerdsTAKattanMWSchumacherM and YuC (2013)
Estimating a time-dependent concordance index for survival
prediction models with covariate dependent censoring.
Statistics in Medicine,
32, 2173–2184. doi:
10.1002/sim.5681
19.
GrollAHastieT and TutzG (2017)
Selection of effects in Cox frailty models by regularization
methods. Biometrics,
73, 846–856.
HastieT and
TibshiraniR (1993)
Varying-coefficient models.
Journal of the Royal Statistical Society. Series B
(Methodological), 55,
757–796. doi:
10.2307/2345993
22.
HolfordTR (1980)
The analysis of rates and of survivorship using log-linear
models. Biometrics,
36, 299–305. doi:
10.2307/2529982
23.
HothornT and
BühlmannP (2006)
Model-based boosting in high dimensions.
Bioinformatics, 22,
2828–2829.
24.
HothornTBühlmannPKneibTSchmidM and HofnerB (2016)
mboost: Model-based boosting.
R package version 2.7-0. URL
https://CRAN.R-project.org/package=mboost
25.
HurvichCMSimonoffJS and TsaiC (1998)
Smoothing parameter selection in non-parametric regression
using an improved Akaike information criterion.
Journal of the Royal Statistical Society: Series B
(Statistical Methodology), 60,
271–293.
26.
KalbfleischJ and
PrenticeR (1980)
The Statistical Analysis of Failure Time
Data. New York, NY:
Wiley.
27.
KleinJP and
MoeschbergerML (1997)
Survival Analysis: Techniques for Censored and Truncated
Data.New York, NY:
Springer.
28.
LairdN and OlivierD (1981)
Covariance analysis of censored survival data using
log-linear analysis techniques. Journal of
the American Statistical Association,
76, 231–240. doi:
10.2307/2287816
29.
MarraG and WoodSN (2011)
Practical variable selection for generalized additive
models. Computational Statistics & Data
Analysis, 55,
2372–2387. doi:
10.1016/j.csda.2011.02.004
30.
MartinussenT and ScheikeTH (2006)
Dynamic Regression Models for Survival
Data. New York, NY:
Springer.
31.
MayrA and HofnerB (2018)
Boosting for statistical modelling: A non-technical
introduction. Statistical
Modelling, 18365–384.
32.
MeierLVan deGeer S and
BühlmannP (2008)
The group LASSO for logistic regression.
Journal of the Royal Statistical
Society, B 70,
53–71.
33.
R Core Team
(2016) R: A Language
and Environment for Statistical Computing.
Vienna: R Foundation for Statistical Computing URL
https://www.R-project.org/
34.
Rodríguez-GirondoMKneibTCadarso-SuárezC and
Abu-AssiE (2013)
Model building in nonproportional hazard
regression. Statistics in
Medicine, 32,
5301–5314. doi:
10.1002/sim.5961
35.
RossiPHBerkRA and
LenihanKJ (1980)
Money, Work, and Crime: Experimental
Evidence.New York: Academic
Press.
36.
RuppertDWandMP and
CarrollRJ (2003)
Semiparametric Regression.
Cambridge: Cambridge
University Press.
37.
Sennhenn-ReulenH and KneibT (2016)
Structured fusion LASSO penalized multi-state
models. Statistics in
Medicine. doi:
10.1002/sim.7017
38.
SimonNFriedmanJHastieT and
TibshiraniR (2011)
Regularization paths for Cox's proportional hazards model via
coordinate descent. Journal of Statistical
Software, 39,
1.
39.
SylvestreM-P and
AbrahamowiczM (2009)
Flexible modeling of the cumulative effects of time-dependent
exposures on the hazard. Statistics in
Medicine, 28,
3437–3453. doi:
10.1002/sim.3701
40.
TherneauTM (2015)
A package for survival analysis in S.
R package version 2.38. URL
http://cran.us.r-project.org/web/packages/survival/index.html
41.
ThomasL and ReyesEM (2014)
Tutorial: Survival estimation for Cox regression models with
time-varying coefficients using SAS and R.
Journal of Statistical Software, Code
Snippets, 61. URL
https://www.jstatsoft.org/article/view/v061c01
42.
WhiteheadJ (1980)
Fitting Cox's regression model to survival data using
GLIM. Journal of the Royal Statistical
Society. Series C (Applied Statistics),
29, 268–275. doi:
10.2307/2346901
43.
WoodSN (2011)
Fast stable restricted maximum likelihood and marginal
likelihood estimation of semiparametric generalized linear
models. Journal of the Royal Statistical
Society: Series B (Statistical Methodology),
73, 3–36.
44.
WoodSN (2012)
On p-values for smooth components of an extended generalized
additive model.
Biometrika, 100,
221–228. doi:
10.1093/biomet/ass048
WoodSNLiZShaddickG and
AugustinNH (2016)
Generalized additive models for gigadata: Modelling the UK
black smoke network daily data. Journal of
the American Statistical Association,
1–40. doi:
10.1080/01621459.2016.1195744
47.
WoodSNPyaN and SaefkenB (2016)
Smoothing parameter and model selection for general smooth
models. Journal of the American Statistical
Association, 111,
1548–1563. doi:
10.1080/01621459.2016.1180986
