adjsurv: Computation of direct adjusted survival estimates based on a...

In mexhaz: Mixed Effect Excess Hazard Models

Computation of direct adjusted survival estimates based on a mexhaz model

Description

Function for computing direct adjusted survival estimates from a model fitted with the mexhaz. It can be used to obtain direct adjusted survival estimates for one or two populations. In the latter case, survival difference estimates are also computed. Corresponding variance estimates are based on the Delta Method (based on the assumption of multivariate normality of the model parameter estimates). When the model includes a random effect, two types of predictions can be made: (i) marginal predictions (obtained by integration over the random effect distribution) or (ii) conditional predictions either for a particular cluster (using the corresponding shrinkage estimate) or for the value 0 of the random effect.

Usage

adjsurv(object, time.pts, data, data.0 = NULL, weights = NULL,
clust.name = NULL, marginal = TRUE, level = 0.95)

Arguments

object	an object of class mexhaz, corresponding to a hazard-based regression model fitted with the mexhaz function.
time.pts	a vector of numerical values representing the time points at which predictions are requested. Time values greater than the maximum follow-up time on which the model estimation was based are discarded.
data	a data.frame containing the values of the covariates of the population for which direct adjusted estimates are to be calculated.
data.0	an optional data.frame containing the values of the covariates of a second population for which direct adjusted estimates can also be calculated (and compared with those of the first population). The default value is set to NULL.
clust.name	name of the variable in data (and data.0) containing information on cluster membership to be used to obtain conditional predictions from a mixed effect model. Standard errors of prediction are obtained with the approximation of the conditional mean squared error of prediction (without the second-order bias correction term) described in Booth and Hobert. This argument is not used if the model is a fixed effect model. The default value is NULL: this corresponds to marginal predictions (if marginal is set to TRUE, the preferred option), or to conditional predictions for the value 0 of the random effect (if marginal is set to FALSE).
weights	optional argument specifying the weights to be associated with each row of data (and data.0). the default value is set to NULL which corresponds to attributing to each row of the dataset(s) a weight equal to one over the total number of rows.
marginal	logical value controlling the type of predictions returned by the function when the model includes a random intercept. When TRUE, marginal predictions are computed. The marginal survival is obtained by integrating the predicted survival over the distribution of the random effect. When FALSE (default value), conditional predictions depending on the value of the cluster argument are calculated.
level	a number in (0,1) specifying the level of confidence for computing the confidence intervals of the hazard and the survival. By default, level=0.95.

Value

An object of class resMexhaz that can be used by the function plot.resMexhaz to produce graphics of the direct adjusted survival curve. It contains the following elements:

results	a data.frame consisting of: the time points at which the direct adjusted survival values have been calculated; the direct ajusted survival values with their confidence limits for population data; the direct ajusted survival values with their confidence limits for population data.0; the direct adjusted survival difference estimates with their confidence limits.
type	type of results returned by the function. The value is used by plot.resMexhaz and lines.resMexhaz, and set to "as" (adjusted survival).
multiobs	value used by plot.resMexhaz and lines.resMexhaz, and set to FALSE (computation of the adjusted survival at several time points for one vector of covariates).
ci.method	method used to compute confidence limits. Currently set to "delta" as only the Delta Method is implemented.
level	level of confidence used to compute confidence limits.

Author(s)

Hadrien Charvat, Aurelien Belot

References

Charvat H, Remontet L, Bossard N, Roche L, Dejardin O, Rachet B, Launoy G, Belot A; CENSUR Working Survival Group. A multilevel excess hazard model to estimate net survival on hierarchical data allowing for non-linear and non-proportional effects of covariates. Stat Med 2016;35:3066-3084 (doi: 10.1002/sim.6881)

Booth JG, Hobert JP. Standard errors of prediction in generalized linear mixed models. J Am Stat Assoc 1998;93:262-272 (doi: 10.2307/2669622).

See Also

plot.resMexhaz, lines.resMexhaz

Examples

data(simdatn1)

## Fit of a fixed-effect hazard model, with the baseline hazard
## described by a linear B-spline with two knots at 1 and 5 year and with
## effects of age (agecr), deprivation index (depindex) and sex (IsexH)

Mod_bs1_2 <- mexhaz(formula=Surv(time=timesurv,
event=vstat)~agecr+depindex+IsexH, data=simdatn1, base="exp.bs",
degree=1, knots=c(1,5), verbose=0)

## Direct adjusted survival for the simdatn1 population
DAS_Modbs1_2 <- adjsurv(Mod_bs1_2, time.pts=seq(1,10),
data=simdatn1)

mexhaz documentation built on Oct. 31, 2022, 5:08 p.m.