There is no problem here.
- Except of course the one that you must assume that being left truncated is non-informative of the incidence before the trunctaion point.
- And the problem that if you have really few persons at risk near the beginning (where you start cumulating the incidence, 0?), you will get enormus uncertainty in your cumulative incidence estimates throughout.
So briefly, unless virtually all you observations start at 0, the cumulative incidences may not be such a good idea after all.
Best regards,
Bendix Carstensen
______________________________________________
Bendix Carstensen
Senior Statistician
Steno Diabetes Center
Niels Steensens Vej 2-4
DK-2820 Gentofte
Denmark
+45 44 43 87 38 (direct)
+45 30 75 87 38 (mobile)
[log in to unmask] http://www.biostat.ku.dk/~bxc
> -----Original Message-----
> From: A UK-based worldwide e-mail broadcast system mailing
> list [mailto:[log in to unmask]] On Behalf Of Juergen Wellmann
> Sent: 16. september 2008 16:20
> To: [log in to unmask]
> Subject: Query: cumulative incidence with left truncation
>
> Dear all,
>
> I have a question regarding survival analysis with competing
> risks (different causes of death). In order to account for
> competing risks, I want to compute cumulative incidence
> functions. However, there is also a late-entry problem. How
> can one consider delayed entry when computing cumulative
> incidence functions?
>
> Kind regards
>
> Juergen Wellmann
>
|