Recurrent event Cox PH model

[William Dudley]

Hello

I will begin working with data from a descriptive study in which the outcome is time to event (the event being change in therapy).

I need to come up to speed on the recurrent event design which takes into account the repeated measures nature of this design?

One area I am interested in is in the potential for time to event in the N event to serve as a covariate in the N+1 event.  

 

So the critical aspect of the design is that Patients will have multiple events - the nature of the disease is that over a period of 3 years (the project duration) patients might experience 7 or more events  

    Singer and Willett  refer to this as a multiple spell design in the 2003 Applied Long Regression text

   Kleinbaum and Klein in their 2012 Survival analysis text  refer to this as a recurrent event design.

I am focusing on the K&K text for now.  Although they have discussion of this design in Chapter 8 (in which they model analyses in the Bladder cancer data set) -

 I would benefit from programming examples in order to really feel competent in the analyses.

So my questions:

Can someone point me to resources that I can use to better understand how to set up the data set, run the models and interpret the results - In SPSS?

Alternatively I suppose I could model this in Mplus but I wanted to stay within SPSS if possible.

Thanks in advance,

Bill

Kleinbaum DG, & Klein M. (2012). Survival analysis: A self-learning text (3rd ed.). New York: NY:Springer.
Singer, J..D ,Willett, J.B. (2003).  Applied longitudinal data analysis: Modeling change and event occurrence, Oxford University Press, New York.

--

===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD