Thanks to Kit Baum, a new package -multistate- is now available to download from the SSC archive. Description is below. I'll be talking about this package at the UK Stata meeting next month in London, and Paul Lambert will be talking about it at the Nordic and Baltic meeting in Oslo the week after. Various updates will follow in the next few weeks, for example the Aalen-Johansen estimator of transition probabilities, which will allow the use of the Cox model.
Description: multistate provides a set of commands for multi-state survival analysis. This includes -msset-, a data preparation tool which converts a dataset from wide (one observation per subject, multiple time and status variables) to long (one observation for each transition of which a subject is at risk), and -predictms-, which calculates transition probabilities, and many other useful measures of absolute risk, such as length of stay in each state, following the fit of a Markov or semi-Markov model using -streg-, or -stpm2-. -predictms- allows different families of models to be specified for each transition. All quantities are calculated through simulation, including confidence intervals. Differences and ratios across covariate patterns can be calculated.
Author: Michael J. Crowther, University of Leicester
Support: email [email protected]
Requires: Stata version 14.1
Thanks,
Michael
Description: multistate provides a set of commands for multi-state survival analysis. This includes -msset-, a data preparation tool which converts a dataset from wide (one observation per subject, multiple time and status variables) to long (one observation for each transition of which a subject is at risk), and -predictms-, which calculates transition probabilities, and many other useful measures of absolute risk, such as length of stay in each state, following the fit of a Markov or semi-Markov model using -streg-, or -stpm2-. -predictms- allows different families of models to be specified for each transition. All quantities are calculated through simulation, including confidence intervals. Differences and ratios across covariate patterns can be calculated.
Author: Michael J. Crowther, University of Leicester
Support: email [email protected]
Requires: Stata version 14.1
Thanks,
Michael
