Relsurv - an R package for relative survival analysis

The package includes a variety of functions used in relative survival analysis. All steps of the analysis are covered:
one can estimate the overall relative survival function, fit regression models and check their fit. Since in practice,
the most annoying part of the relative survival analysis is to put the population tables into the right format, some functions
to help with this initial step of the analysis are also provided despite the fact that the ratetable format originates
from the survival package.

Available functions

  • Overall relative survival curve: rs.surv
  • Calculates the overall relative survival curve. The syntax mimics the survfit function (survival package).

    Methods available:

    • new method by Pohar Perme et al (estimates net survival - unbiased)
    • Hakulinen (estimates relative survival ratio)
    • Ederer I (estimates relative survival ratio)
    • Ederer II (estimates net survival - biased).

    The curve can be split by covariates and plotted. Function summary gives values at specific times, std. errors and confidence intervals.

  • Additive relative survival regression model: rsadd
  • Fits the additive (proportional excess hazard) model, using several options:

    • semiparametric model (EM, no assumptions on the baseline form)
    • parametric model with max. likelihood (constant baseline within sub-intervals)
    • parametric model with glm (Poisson or binomial link, constant baseline within sub-intervals)

    For prediction: use rs.surv.rsadd(fit), available for max.lik.

  • Goodness of fit functions
  • Goodness of fit checking after fitting an additive relative survival model, a multiplicative model or a classical Cox model.

    • rs.zph for graphical inspection (mimics cox.zph, but does no formal testing, only for additive relative survival model)
    • (formal testing), can also be plotted
  • Individual relative survival: rstrans
  • Using transformation approach we can:

    • calculate the Y values (could also be done using survexp function)
    • fit the Cox model in Y
    • back transform from Y: invtime (not thoroughly checked)

    For a further presentation of the idea, see an illustration.

  • Multiplicative model: rsmul
  • Fits the Andersen et al multiplicative model.

  • Population mortality tables
  • Assistance in forming the ratetable object (defined in package survival):

    • form objects from your own tables: transrate
    • form objects from tables available on the web: transrate.hld, transrate.hmd
    • join two ratetable objects into one: joinrate

    For more information on this time consuming step, see Population tables.

    About IBMI

    Institute for Biostatistics and Medical Informatics (IBMI), formerly Institute for BioMedical Informatics (so still IBMI) was founded by the Faculty of Medicine as a result of a need for a unit which would perform, or coordinate, tasks related to data analysis and providing information, relevant for research in medicine. The programme of the institute, and its development, have been adjusting thorugh time to changes in financing and technological progress, but the basic aim remain the same: to support research in medicine. This is achieved through the following tasks:


    Institute for Biostatistics and Medical Informatics
    University of Ljubljana, Faculty of Medicine
    Vrazov trg 2, 1000 Ljubljana

    tel: +386 1 543-77-70
    fax: +386 1 543-77-71
    email: ibmi (at)