BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:A Bayesian synthesis of evidence for estimating HIV prevalence and
  incidence - Anne Presanis\, MRC Biostatistics Unit\, Cambridge
DTSTART:20090507T153000Z
DTEND:20090507T163000Z
UID:TALK18277@talks.cam.ac.uk
CONTACT:Olivier Restif
DESCRIPTION:Disease incidence and prevalence are not always directly measu
 rable\, and increasingly are being estimated by synthesising diverse sourc
 es of evidence in a full probability model\, typically in a Bayesian frame
 work. In this talk I will describe such a model for estimating HIV inciden
 ce among men who have sex with men in England and Wales. We start with a t
 wo-stage process: first\, estimating prevalence from surveillance and othe
 r ad-hoc survey data\; then estimating incidence from the posterior preval
 ence estimates together with further data on diagnosis rates\, demographic
 s and risk behaviour change. This model is then expanded to a dynamic tran
 smission model\, and finally prevalence and incidence are simultaneously e
 stimated in a single large evidence synthesis.
LOCATION:Meeting Room 14\, Centre for Mathematical Sciences
END:VEVENT
END:VCALENDAR