Frailtypack


Welcome to FRAILTYPACK, an online modelling and prediction tool designed to help clinicians, epidemiologists and statisticians. Different modelling for clustered or recurrent failure times data are proposed, with also the possibility to make prediction of the future of the patients in terms of survival or risk of recurrence accounting for the history of the patient.


Development of the software was a collaborative project in the INSERM Biostatistical team.


We welcome any feedback you may have about FRAILTYPACK. If you have questions about its development or there are features you would like to have added to the model please let us know by emailing us at virginie.rondeau@inserm.fr


Cox Model - Modelisation


Positive smoothing parameter :

Cox Model - Prediction

Choose profile of the patient(s)
(right clik on the left block to add a line)

Cox Model


The proportional hazards Cox model is a frailty model without random effect. With frailtypack, it is possible to fit such a model with parameters estimated by penalized likelihood maximization.


To fit a such model, you need first to load your data file (text, csv and excel are allowed), then you have to set the following parameters :


Time : name of the column corresponding to the time.

Censoring indicator name of the column corresponding to the censoring indicator.

Co-variables name of the column corresponding to the co-variables.


The following parameters are set by default but you are free to modifie them as you need :


Hazard function : type of hazard functions: Splines for semiparametric hazard functions using equidistant intervals or Splines-per using percentile with the penalized likelihood estimation, Piecewise-per for piecewise constant hazard function using percentile (not available for interval-censored data), Piecewise-equi for piecewise constant hazard function using equidistant intervals, Weibull for parametric Weibull functions. Default is Splines.

Number of knots : integer giving the number of knots to use.(only for splines or splines-per hazard function)

Positive smoothing parameter : positive smoothing parameter in the penalized likelihood estimation.(only for splines or splines-per hazard function)

Use cross validation : Logical value. Is cross validation procedure used for estimating smoothing parameter in the penalized likelihood estimation ? If so a search of the smoothing parameter using cross validation is done, with kappa as the seed. .(only for splines or splines-per hazard function)

Number of time interval : integer giving the number of time intervals.(only for Piecewise-per or Piecewise-equi hazard function)


If you want to change the datafile after having fit a model, you need to press the button 'Change data file' after select a new file.

Shared Frailty Model - Modelisation


Positive smoothing parameter :

Shared Frailty Model - Prediction

Choose profile of the patient(s)
(right clik on the left block to add a line)

Shared Model


Fit a shared gamma or log-normal frailty model using a semiparametric Penalized Likelihood estimation or parametric estimation on the hazard function. Left-truncated, right-censored data, interval-censored data and strata (up to 6 levels) are allowed. It allows to obtain a non-parametric smooth hazard of survival function. This approach is different from the partial penalized likelihood approach of Therneau et al.


To fit a such model, you need first to load your data file (text, csv and excel are allowed), then you have to set the following parameters :


The variables : Time, Censoring indicator, Cluster, Co-variables : set with name of the column corresponding.


The following parameters are set by default but you are free to modifie them as you need :


Hazard function : type of hazard functions: Splines for semiparametric hazard functions using equidistant intervals or Splines-per using percentile with the penalized likelihood estimation, Piecewise-per for piecewise constant hazard function using percentile (not available for interval-censored data), Piecewise-equi for piecewise constant hazard function using equidistant intervals, Weibull for parametric Weibull functions. Default is Splines.

Number of knots : integer giving the number of knots to use.(only for splines or splines-per hazard function)

Positive smoothing parameter : positive smoothing parameter in the penalized likelihood estimation.(only for splines or splines-per hazard function)

Use cross validation : Logical value. Is cross validation procedure used for estimating smoothing parameter in the penalized likelihood estimation ? If so a search of the smoothing parameter using cross validation is done, with kappa as the seed. .(only for splines or splines-per hazard function)

Number of time interval : integer giving the number of time intervals.(only for Piecewise-per or Piecewise-equi hazard function)

Init theta : initial value for variance of the frailties.

Type of random effect distribution :'Gamma' for a gamma distribution, 'LogN'for a log-normal distribution. Default is 'Gamma'.


If you want to change the datafile after having fit a model, you need to press the button 'Change data file' after select a new file.

Additive Frailty Model - Modelisation


Additive Model


Fit an additive frailty model using a semiparametric penalized likelihood estimation or a parametric estimation. The main issue in a meta-analysis study is how to take into account the heterogeneity between trials and between the treatment effects across trials. Additive models are proportional hazard model with two correlated random trial effects that act either multiplicatively on the hazard function or in interaction with the treatment, which allows studying for instance meta-analysis or multicentric datasets. Right-censored data are allowed, but not the left-truncated data. A stratified analysis is possible (maximum number of strata = 2). This approach is different from the shared frailty models.


To fit a such model, you need first to load your data file (text, csv and excel are allowed), then you have to set the following parameters :


The variables : Time, Censoring indicator, Cluster, Co-variables, Co-variable slope : set with name of the column corresponding.


The following parameters are set by default but you are free to modifie them as you need :


Hazard function : type of hazard functions: Splines for semiparametric hazard functions using equidistant intervals or Splines-per using percentile with the penalized likelihood estimation, Piecewise-per for piecewise constant hazard function using percentile (not available for interval-censored data), Piecewise-equi for piecewise constant hazard function using equidistant intervals, Weibull for parametric Weibull functions. Default is Splines.

Number of knots : integer giving the number of knots to use.(only for splines or splines-per hazard function)

Positive smoothing parameter : positive smoothing parameter in the penalized likelihood estimation.(only for splines or splines-per hazard function)

Use cross validation : Logical value. Is cross validation procedure used for estimating smoothing parameter in the penalized likelihood estimation ? If so a search of the smoothing parameter using cross validation is done, with kappa as the seed. .(only for splines or splines-per hazard function)

Number of time interval : integer giving the number of time intervals.(only for Piecewise-per or Piecewise-equi hazard function)

The random effect are correlated :yes or no


If you want to change the datafile after having fit a model, you need to press the button 'Change data file' after select a new file.

Nested Frailty Model - Modelisation


Positive smoothing parameter :

Nested Model


Data should be ordered according to cluster and subcluster Fit a nested frailty model using a Penalized Likelihood on the hazard function or using a para metric estimation. Nested frailty models allow survival studies for hierarchically clustered data by including two iid gamma random effects. Left-truncated and right-censored data are allowed. Stratification analysis is allowed (maximum of strata = 2).


To fit a such model, you need first to load your data file (text, csv and excel are allowed), then you have to set the following parameters :


The variables : Time, Censoring indicator, Cluster,Subcluster, Co-variables : set with name of the column corresponding.


The following parameters are set by default but you are free to modifie them as you need :


Hazard function : type of hazard functions: Splines for semiparametric hazard functions using equidistant intervals or Splines-per using percentile with the penalized likelihood estimation, Piecewise-per for piecewise constant hazard function using percentile (not available for interval-censored data), Piecewise-equi for piecewise constant hazard function using equidistant intervals, Weibull for parametric Weibull functions. Default is Splines.

Number of knots : integer giving the number of knots to use.(only for splines or splines-per hazard function)

Positive smoothing parameter : positive smoothing parameter in the penalized likelihood estimation.(only for splines or splines-per hazard function)

Use cross validation : Logical value. Is cross validation procedure used for estimating smoothing parameter in the penalized likelihood estimation ? If so a search of the smoothing parameter using cross validation is done, with kappa as the seed. .(only for splines or splines-per hazard function)

Number of time interval : integer giving the number of time intervals.(only for Piecewise-per or Piecewise-equi hazard function)

Init eta : initial value for parameter eta.


If you want to change the datafile after having fit a model, you need to press the button 'Change data file' after select a new file.

Joint Frailty Model - Modelisation


Positive smoothing parameter :
Positive smoothing parameter :

Joint Frailty Model - Prediction

Choose profile of the patient(s)
(right clik on the left block to add a line)

Joint Model


Fit a joint model either with gamma or log-normal frailty model for recurrent and terminal events using a penalized likelihood estimation on the hazard function or a parametric estimation. Right-censored data and strata (up to 6 levels) for the recurrent event part are allowed. Left-truncated data is not possible. Joint frailty models allow studying, jointly, survival processes of recurrent and terminal events, by considering the terminal event as an informative censoring.


To fit a such model, you need first to load your data file (text, csv and excel are allowed), then you have to set the following parameters :


The variables : Time, Censoring indicator, Cluster, Co-variables for recurrents and terminal event, terminal event : set with name of the column corresponding.


The following parameters are set by default but you are free to modifie them as you need :


Hazard function : type of hazard functions: Splines for semiparametric hazard functions using equidistant intervals or Splines-per using percentile with the penalized likelihood estimation, Piecewise-per for piecewise constant hazard function using percentile (not available for interval-censored data), Piecewise-equi for piecewise constant hazard function using equidistant intervals, Weibull for parametric Weibull functions. Default is Splines.

Number of knots : integer giving the number of knots to use.(only for splines or splines-per hazard function)

Positive smoothing parameter : positive smoothing parameter in the penalized likelihood estimation.(only for splines or splines-per hazard function)

Use cross validation : Logical value. Is cross validation procedure used for estimating smoothing parameter in the penalized likelihood estimation ? If so a search of the smoothing parameter using cross validation is done, with kappa as the seed. .(only for splines or splines-per hazard function)

Number of time interval : integer giving the number of time intervals.(only for Piecewise-per or Piecewise-equi hazard function)

Init theta : initial value for variance of the frailties.

Init alpha : initial value for parameter alpha.

Type of random effect distribution :'Gamma' for a gamma distribution, 'LogN'for a log-normal distribution. Default is 'Gamma'.


If you want to change the datafile after having fit a model, you need to press the button 'Change data file' after select a new file.

Joint Longitudinal Frailty Model - Modelisation


Joint Longi Frailty Model - Prediction

Choose profile of the patient(s)
(right clik on the left block to add a line)
---terminal event
---biomarker observations

Joint Longi Model


Fit a joint model for longitudinal data and a terminal event using a semiparametric penalized likeli hood estimation or a parametric estimation on the hazard function. We consider that the longitudinal outcome can be a subject to a quantification limit, i.e. some observations, below a level of detection s cannot be quantified (left-censoring).


To fit a such model, you need first to load your data file (text, csv and excel are allowed), then you have to set the following parameters :


The variables : Time, Censoring indicator, Co-variables for the terminal event, Biomarker, Co-variables for the longitudinal outcome, Variables for the random effects of the longitudinal outcome, Variable representing the individuals : set with name of the column corresponding.


The following parameters are set by default but you are free to modifie them as you need :


Hazard function : type of hazard functions: Splines for semiparametric hazard functions using equidistant intervals or Splines-per using percentile with the penalized likelihood estimation, Weibull for parametric Weibull functions. Default is Splines.

Number of knots : integer giving the number of knots to use.(only for splines or splines-per hazard function)

Positive smoothing parameter : positive smoothing parameter in the penalized likelihood estimation.(only for splines or splines-per hazard function)

Use cross validation : Logical value. Is cross validation procedure used for estimating smoothing parameter in the penalized likelihood estimation ? If so a search of the smoothing parameter using cross validation is done, with kappa as the seed. .(only for splines or splines-per hazard function)

Init eta : initial value for parameter eta.

Type of link function for the dependence between the biomarker and death :'Random-effects' for the association directly via the random effects of the biomarker, 'Current-level' for the association via the true current level of the biomarker. The option 'Current-level' can be chosen only if the biomarker random effects are associated with the intercept and time (following this order). The default is 'Random-effects' .


Method for the Gauss-Hermite quadrature : 'Standard' for the standard non- adaptive Gaussian quadrature, 'Pseudo-adaptive' for the pseudo-adaptive Gaus sian quadrature and 'HRMSYM' for the algorithm for the multivariate non-adaptive Gaussian quadrature (see Details). The default is 'Standard' ..


If you want to change the datafile after having fit a model, you need to press the button 'Change data file' after select a new file.

Joint Trivariate Model - Modelisation


Positive smoothing parameter :

Joint Trivariate Model - Prediction

Choose profile of the patient(s)
(right clik on the left block to add a line)
---terminal event
---biomarker observations

Trivariate Joint Model


Fit a trivariate joint model for longitudinal data, recurrent events and a terminal event using a semi- parametric penalized likelihood estimation or a parametric estimation on the hazard functions. We consider that the longitudinal outcome can be a subject to a quantification limit, i.e. some observations, below a level of detection s cannot be quantified (left-censoring)


To fit a such model, you need first to load your data file (text, csv and excel are allowed), then you have to set the following parameters :


The variables : Time, Censoring indicator,Cluster,Co-variables for the recurrent event, Co-variables for the terminal event,Terminal event, Biomarker, Co-variables for the longitudinal outcome, Variables for the random effects of the longitudinal outcome, Variable representing the individuals : set with name of the column corresponding.


The following parameters are set by default but you are free to modifie them as you need :


Hazard function : type of hazard functions: Splines for semiparametric hazard functions using equidistant intervals or Splines-per using percentile with the penalized likelihood estimation, Weibull for parametric Weibull functions. Default is Splines.

Number of knots : integer giving the number of knots to use.(only for splines or splines-per hazard function)

Positive smoothing parameter : positive smoothing parameter in the penalized likelihood estimation.(only for splines or splines-per hazard function)

Use cross validation : Logical value. Is cross validation procedure used for estimating smoothing parameter in the penalized likelihood estimation ? If so a search of the smoothing parameter using cross validation is done, with kappa as the seed. .(only for splines or splines-per hazard function)

Init eta : initial value for parameter eta.

Type of link function for the dependence between the biomarker and death :'Random-effects' for the association directly via the random effects of the biomarker, 'Current-level' for the association via the true current level of the biomarker. The option 'Current-level' can be chosen only if the biomarker random effects are associated with the intercept and time (following this order). The default is 'Random-effects' .


Method for the Gauss-Hermite quadrature : 'Standard' for the standard non- adaptive Gaussian quadrature, 'Pseudo-adaptive' for the pseudo-adaptive Gaus sian quadrature and 'HRMSYM' for the algorithm for the multivariate non-adaptive Gaussian quadrature (see Details). The default is 'Standard' ..


If you want to change the datafile after having fit a model, you need to press the button 'Change data file' after select a new file.

Joint Non Linear Trivariate Model - Modelisation


Positive smoothing parameter :

Joint Non Linear Trivariate Model - Prediction

Choose profile of the patient(s)
(right clik on the left block to add a line)
---terminal event
---biomarker observations

Non Linear Trivariate Joint Model


Fit a non-linear trivariate joint model for a longitudinal biomarker, recurrent events and a terminal event using a semiparametric penalized likelihood estimation or a parametric estimation on the hazard functions. We consider that the longitudinal outcome can be a subject to a quantification limit, i.e. some observations, below a level of detection s cannot be quantified (left-censoring)


The variables : Time, Censoring indicator,Cluster,Co-variables for the recurrent event, Co-variables for the terminal event, Terminal event, Biomarker, Times of biomarker measurements, Co-variables for the biomarker growth, Co-variables for the biomarker drug-induced decline, Parameters with random effects included in the model, Variable representing the individuals, Drug concentration indicator : set with name of the column corresponding.


The following parameters are set by default but you are free to modify them as you need :


Hazard function : type of hazard functions: Splines for semiparametric hazard functions using equidistant intervals or Splines-per using percentile with the penalized likelihood estimation, Weibull for parametric Weibull functions. Default is Splines.

Number of knots : integer giving the number of knots to use.(only for splines or splines-per hazard function)

Positive smoothing parameter : positive smoothing parameter in the penalized likelihood estimation.(only for splines or splines-per hazard function)

Method for the Gauss-Hermite quadrature : 'Standard' for the standard non- adaptive Gaussian quadrature, 'Pseudo-adaptive' for the pseudo-adaptive Gaus sian quadrature and 'HRMSYM' for the algorithm for the multivariate non-adaptive Gaussian quadrature (see Details). The default is 'Standard' ..

Init alpha : initial value for parameter alpha.

Number of nodes : Number of nodes for the Gauss-Hermite quadrature. Default is 9.

Biomarker left-censored below a treshold : No if there is no left-censoring, yes otherwise, with the corresponding treshold.


If you want to change the datafile after having fit a model, you need to press the button 'Change data file' after select a new file.

Joint multivariate Model - Modelisation


Positive smoothing parameter :

Joint Surrogate Model - Modelisation


Iterations before re-estimation :
Theta initial value :
Sigma.ss initial value :
Sigma.tt initial value :
Sigma.st value :
Gamma initial value :
Alpha initial value :
Zeta initial value :
Beta.s initial value :
Beta.t initial value :

Joint Surrogate - Prediction

(Optional)

Choose profile of the patient(s)
(right clik on the left block to add a line)
A dataset cannot be added for the moment. It is under construction

Prediction

Joint Surrogate Model


Fit the one-step Joint surrogate model for the evaluation of a canditate surrogate endpoint, with different integration methods on the random effects, using a semiparametric penalized likelihood estimation.


The data must contain at least 7 variables entitled : patienID, trialID, timeS, statusS, timeT, statusT, trt .


patienID : A numeric, that represents the patient’s identifier, must be unique

trialID : A numeric, that represents the trial in which each patient wasrandomized

timeS : The follow up time associated with the surrogate endpoint

statusS : The event indicator associated with the surrogate endpoint. Normally 0 = no event, 1 = event

timeT : The follow up time associated with the true endpoint

statusT : The event indicator associated with the true endpoint. Normally0 = no event, 1 = event

trt : The treatment indicator for each patient, with 1 = treated, 0 = un-treated

The following parameters are set by default but you are free to modify them as you need :


Number of knots : An integer giving the number of knots to use. (only for splines or splines-per hazard function)

Kappa use : A numeric, that indicates how to manage the smoothing parameters in case of convergence issues. If set to 1, the given smoothing parameters or those obtained by cross-validation are used. If set to 3, the associated smoothing parameters are successively divided by 10, in case of convergence issues until 5 times. If set to 4, the management of the smoothing parameter is as in case 1, follows by the successive division as described in case 3 and preceded by the changing of the number of nodes for the Gauss-Hermite quadrature.

Integration method : 0 for Monte carlo, 1 for Gaussian-Hermite quadrature, 2 for a combination of both Gaussian-Hermite quadrature to integrate over the individual-level random effects and Monte carlo to integrate over the trial-level random effects, 4 for a combination of both Monte carlo to integrate over the individual-level random effects and Gaussian-Hermite quadrature to integrate over the trial-level random effects.

Number of samples in the Monte-Carlo integration : A value between 100 and 300 most often gives good results. However, beyond 300, the program takes a lot of time to estimate the parameters. The default is 300.

Number of nodes for the Gauss-Hermite quadrature : Default is 32.

Number of nodes for the Gauss-Hermite quadrature in case of non-convergence : Default is 20.

Adaptatif : A binary, indicates whether the pseudo adaptive Gaussian-Hermite quadrature (1) or the classical Gaussian-Hermite quadrature (0) is used.

Frail base : Considered the heterogeneity between trial on the baseline risk (1), using the shared cluster specific frailties (u_i), or not (0).

Indicator alpha : A binary, indicates whether the power's parameter ζ should be estimated (1) or not (0). If 0, ζ will be set to 1 during estimation. The default is 1. This parameter can be seted to 0 in case of convergence and identification issues.

Indicator zeta : A binary, indicates whether the power's parameter α should be estimated (1) or not (0). If 0, α will be set to 1 during estimation.


If you want to change the datafile after having fit a model, you need to press the button 'Change data file' after select a new file.