I have been given some data that has date of admission, date of disposition, disposition (death or live discharge), and some biomarker levels. I would like to model this, but here is the rub:

The longest-staying patient died. All other patients left the study (were discharged from hospital) before that time period.

Is a conventional survival analysis appropriate for this?

  • 1
    $\begingroup$ why wouldn't it be? $\endgroup$ – IWS May 24 '17 at 11:58
  • $\begingroup$ In addition, this is the case where you can estimate mean life length without extrapolation using semiparametric method. $\endgroup$ – Frank Harrell May 24 '17 at 12:31
  • $\begingroup$ Their outcome for surviving patients is just censored. All that might be affected is the accuracy of the estimated survival curve. $\endgroup$ – Michael R. Chernick May 24 '17 at 16:30

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.