You are here: Home DFG - Research Unit Projects Multi-state


Impact of intermediate events in multi-state models

Duration: 2004-2010

Principal investigator:  Prof. Dr. Martin Schumacher

Researchers: Prof. Dr. Martin Schumacher (IMBI), Arthur Allignol (M.Sc., IMBI), Dr. Jan Beyersmann (IMBI), Dr. Martin Wolkewitz (IMBI) 



Multi-state and competing risks methodology have remained a highly complex, but also practically very relevant field in biomedical research and application. Analysis of single transition hazards is feasible, but interpretationally difficult. Also, multi-state models are usually assumed to be Markovian, buth this assumption may be questioned in practice. In this project, we will study direct regression of transition probabilities, use the relationship of multi-state models and multiple time scales to relax the nonparametric Markov assumption, and further promote multistate methodology, in particular our results from the first three years, in applications. The methodology shall be made available to the statistical community and beyond through openly available R-packages. Cooperation with statistical computing will consequently be intensive. Hospital infection data will continue to be a focus for application of our methodology. This field has recently shown an increasing interest in so-called `mathematical modeling'; our project will also work in this area, relating epidemic models from `mathematical modeling' to the actual data. The aim is to establish nonparametric data analysis based on counting processes for transmission models of hospital infection.



  • Lambert ML, Suetens C, Savey A, Palomar M, Hiesmayr M, Morales I, Agodi A, Frank U, Mertens K,
    Schumacher M, et al.. Clinical outcomes of health-care-associated infections and antimicrobial resistance in
    patients admitted to european intensive-care units: a cohort study. Lancet Infect Dis 2011; 11(1):30–8.
  • Beyersmann J, Wolkewitz M, Allignol A, Grambauer N, Schumacher M. Application of multistate models
    in hospital epidemiology: Advances and challenges. Biom J 2011; 53(2):332–50.
  • Grambauer N, Schumacher M, Beyersmann J. Proportional subdistribution hazards modeling offers a sum-
    mary analysis, even if misspecified. Statist. Med. 2010; 29(7/8):875–884.
  • Wolkewitz M, Allignol A, Schumacher M, Beyersmann J. Two Pitfalls in Survival Analyses of Time-
    Dependent Exposure: A Case Study in a Cohort of Oscar Nominees. Am Stat 2010; 64(3):205–211.
  • Allignol A, Schumacher M, Beyersmann J. A note on variance estimation of the aalen-johansen estimator of
    the cumulative incidence function in competing risks, with a view towards left-truncated data. Biom J 2010;
  • Wolkewitz M, Schumacher M. Simulating and analysing infectious disease data in a heteroge-
    neous population with migration. Computer Methods and Programs in Biomedicine 2010; doi:
  • Grambauer N, Schumacher M, Dettenkofer M, Beyersmann J. Incidence densities in a competing events
    analysis. Am J Epidemiol 2010; 172(9):1077–84.
  • Graw F, Gerds TA, Schumacher M. On pseudo-values for regression analysis in competing risks models.
    Lifetime Data Anal 2009; 15(2):241–55.
  • Beyersmann J, Latouche A, Buchholz A, Schumacher M. Simulating competing risks data in survival ana-
    lysis. Statist. Med. 2009; 28(6):956–71.
  • Wolkewitz M, Beyersmann J, Gastmeier P, Schumacher M. Efficient risk set sampling when a time-
    dependent exposure is present: matching for time to exposure versus exposure density sampling. Methods
    Inf Med 2009; 48:438–443.
  • Wolkewitz M, Beyersmann J, Gastmeier P, Schumacher M. Modeling the effect of time-dependent exposure
    on intensive care unit mortality. Intensive Care Med 2009; 35(5):826–32.
  • Beyersmann J, Wolkewitz M, Schumacher M. The impact of time-dependent bias in proportional hazards
    modelling. Statist. Med. 2008; 27(30):6439–54.
  • Beyersmann J, Gastmeier P, Grundmann H, B ̈ rwolff S, Geffers C, Behnke M, R ̈ den H, Schumacher M.
    Transmission-associated nosocomial infections: prolongation of intensive care unit stay and risk factor ana-
    lysis using multistate models. Am J Infect Control 2008; 36(2):98–103.
  • Wolkewitz M, Dettenkofer M, Bertz H, Schumacher M, Huebner J. Statistical epidemic modeling with
    hospital outbreak data. Statist. Med. 2008; 27(30):6522–31.
  • Beyersmann J, Schumacher M. Time-dependent covariates in the proportional subdistribution hazards model
    for competing risks. Biostatistics 2008; 9(4):765–76.
  • Beyersmann J, Gastmeier P, Wolkewitz M, Schumacher M. An easy mathematical proof showed that time-
    dependent bias inevitably leads to biased effect estimation. J Clin Epidemiol 2008; 61(12):1216–21.
  • Beyersmann J, Dettenkofer M, Bertz H, Schumacher M. A competing risks analysis of bloodstream infection after  stem-cell transplantation using subdistribution hazards and cause-specific hazards. Statist. Med. 2007;
  • Beyersmann J. A random time interval approach for analysing the impact of a possible intermediate event
    on a terminal event. Biom J 2007; 49(5):742–749.