Home  >>  Archives  >>  Volume 13 Number 4  >>  st0316

The Stata Journal
Volume 13 Number 4: pp. 759-775

Subscribe to the Stata Journal

Flexible parametric illness-death models

Sally R. Hinchliffe
Department of Health Sciences
University of Leicester
Leicester, UK
David A. Scott
Oxford Outcomes Ltd
Oxford, UK
Paul C. Lambert
Department of Health Sciences
University of Leicester
Leicester, UK
Department of Medical Epidemiology and Biostatistics
Karolinska Institutet
Stockholm, Sweden
Abstract.  It is usual in time-to-event data to have more than one event of interest, for example, time to death from different causes. Competing risks models can be applied in these situations where events are considered mutually exclusive absorbing states. That is, we have some initial state—for example, alive with a diagnosis of cancer—and we are interested in several different endpoints, all of which are final. However, the progression of disease will usually consist of one or more intermediary events that may alter the progression to an endpoint. These events are neither initial states nor absorbing states. Here we consider one of the simplest multistate models, the illness-death model. stpm2illd is a postestimation command used after fitting a flexible parametric survival model with stpm2 to estimate the probability of being in each of four states as a function of time. There is also the option to generate confidence intervals and transition hazard functions. The new command is illustrated through a simple example.
Buy article (PDF): $11.75 View cart
Buy entire issue (PDF): $24.75

View all articles by these authors: Sally R. Hinchliffe, David A. Scott, Paul C. Lambert

View all articles with these keywords: illdprep, stpm2illd, survival analysis, multistate models, flexible parametric models

Download citation: BibTeX  RIS

Download citation and abstract: BibTeX  RIS