PhD Studentship in Biostatistics/Pharmacoepidemiology
at the University of Southern Denmark, Odense

Project Title: Statistical Methods for the Analysis of
Pharmacoepidemiological Databases

One of the main methodological developments in
pharmacoepidemiology is the emergence of large databases
with more or less complete capture of individual drug use
and clinical outcomes for large populations. Odense, for example, is
home of the Odense Pharmacoepidemiologic Database (OPED) where
all prescriptions delivered to pharmacies in the County of Fyn
are registered since 1991.

Until now, there have been little efforts to develop statistical
methods for the analysis of such databases. Most medical research
based on pharmacoepidemiological databases makes use of traditional
statistical methods borrowed from other fields, e.g. the analysis
of cohort and case control studies in epidemiology. However, it is
sometimes questionable, whether such methods can be used without
adaptation, because pharmacoepidemiological databases show some
specific features. For examples, some investigations are subject to
length biased sampling due to the fact, that only events in a certain
time window can be observed. Other applications require the true
treatment
or disease status to be reconstructed from the observed events. This is
today often based on rather arbitrary decision rules, which may
introduce a systematic bias.

In a previous Ph.D. project methods to estimate incidence and prevalence

based on pharmacoepidemiological databases have been investigated. The
current project should focus on the analysis of event patterns. It
should
cover both a critical review of simple ad hoc methods to evaluate
basic characteristics, for example the mean duration between events,
as well as the development of more sophisticated approaches to
describe the distribution of the time between events by modeling
the individual course as well as the heterogeneity among subjects. The
latter might be extended to include latent processes, if the events
observed
allow only a partial reconstruction of the process of interest.

The position will be  attached to the biostatistics group
(http://www.isd.sdu.dk/biostat) at the
Department of Statistics and Demography (http://www.sdu.sk/tvf/statdem)
 and to the Research Unit of General Practice
(http://www.sdu.dk/health/Research/Units/Genpract.htm). The Ph.D.
student
should participate in several applied projects
based on OPED and similar databases and conducted at the Research Unit
or at the Institute of Clinical Pharmacology. There is a possibility
to work some months prior to the enrollment on a research assistance
level in order to prepare the final application to the Faculty. The
position should be filled during the year 2001.

Applications are invited from suitable candidates with a background in
statistics. Candidates should have a degree in
statistics, biostatistics, mathematics, epidemiology or similar
subjects.
They should be interested in working on the boundary between application

and theory, should have an interest in methodology and should have good
computer skills.


Further information can by obtained from

Werner Vach
Professor of Medical Statistics
Department of Statistics and Demography
University of Southern Denmark
Sdr. Boulevard 23a
DK-5000 Odense C
email: [log in to unmask]

or

Jakob Kragstrup
Professor of General Practice
Institute of Public Health
University of Southern Denmark
Winsløwparken 19
DK-5000 Odense C
email: [log in to unmask]


Applications (including a cv) should be sent to Werner Vach.



--
Werner Vach
Institut for Statistik og Demografi (Department of Statistics and Demography)
Syddansk Universitet (University of Southern Denmark)

Mailing address: Campusvej 55, DK-5230 Odense M, Denmark
Visitors' address: Sdr. Boulevard 23A, 3rd. floor, 5000 Odense C

Phone: +45 65 50 33 83  Fax: +45 65 95 77 66
email: [log in to unmask]