SEMINAR IN LEICESTER
There will be a seminar, as part of the Nottingham and
Leicester Universities Statistical Seminars series, on
Thursday 9th November at 4.30pm (Tea 4pm) given by Dr
Kate Tilling on the estimation of incidence rates.
ALL ARE VERY WELCOME.
"Capture-recapture including covariates: estimating
the incidence of stroke."
Dr Kate Tilling
King's College, London
VENUE
The seminar will take place at 4.30pm in room G20 of the
Department of Epidemiology & Public Health, University of
Leicester, 22-28 Princess Road West, LEICESTER,
and will be preceded by tea and biscuits from 4.00pm.
Directions can be found at http://www.prw.le.ac.uk/
For further information, please contact Eve Kilbourne
tel: 0116-252-3276 or email: [log in to unmask]
ABSTRACT
Capture-recapture methods are used to estimate incidence
of a disease, using a multiple-source register. The usual
log-linear models which are used to estimate population
size assume that not all sources of notification are
dependent, and that probability of capture does not depend
on patient characteristics. The assumptions are rarely
satisfied in epidemiology.
I compare log-linear and logit models with and without
covariates, and use simulated data to compare estimates
from different models. The crude estimate of population
size is biased when the sources are not independent, whereas
analyses adjusting for covariates produce less biased
estimates. I then use capture-recapture models including
covariates to estimate incidence of stroke in South London,
together with age-standardised incidence rates, and
calculate confidence intervals for incidence which allow
for the uncertainty in estimation of population size.
Cost-effectiveness of routine registers might be increased by
using the combination of sources and covariates which most
accurately estimates population size, rather than by aiming
for 100% completeness.
Nick Taub
Lecturer in Medical Statistics
Department of Epidemiology & Public Health
University of Leicester
22-28 Princess Rd West
Leicester. LE1 6TP
Email: [log in to unmask]
Tel: 0116-252-5416
Fax: 0116-252-3272
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|