Surveys and monitoring systems – design questions, analysis and presentation of results

Dates:	March 26-27, 2007
Location:	SAS-Radisson Hotel, Dipoli, Finland
Instructors:	Marcus Doherr, University of Bern, Switzerland Nils Toft, Faculty of Life Science, University of Copenhagen, Denmark
Objective:	The course objective is to make participants familiar with different software tools (Excel, @Risk, WinBugs, R) and published (tested) routines in order to analyze (risk-based) disease survey and monitoring data
Background:	The crude outcome of (risk-based) disease-monitoring systems is the number of existing (prevalent) or newly occurring (incident) cases of the outcome of interest, which for comparison purposes needs to be put into relation to the sample of animals at risk of being a case of or developing this outcome during a given period of time. The resulting proportions or rates include both sampling variability and, depending on the approach, the diagnostic tests used etc., also uncertainty, which should be acknowledged when presenting the results to decision makers and other stakeholders. Although data are similar in appearance, the design and purpose of the sampling scheme might impose very different restrictions and/or considerations when analyzing and subsequently presenting the outcome. Consider a disease that is endemic in the population. A natural first step is to assess the magnitude of this disease, i.e. establish the prevalence through a cross-sectional study. If subsequently a decision is made  to control/eradicate this disease, then there is a need to evaluate the control program by estimating the incidence or changes in prevalence across time. Finally, if the control program is successful in eradicating the disease, there is a need to demonstrate freedom from disease. The objective of this proposed workshop course is to present a practical approach to analyzing and presenting survey/monitoring data from each of the three situations described above. The course will focus on presenting a series of working solutions in terms of “templates” for analyses for the different scenarios using different statistical methods and tools. In addition, approaches how to (best) present the outcome with respect to the specific monitoring objective will be discussed. At the end of the course, participants are expected to have developed an understanding of the challenges and options in analysing such problems, including the importance of the link between design, sampling and analytic issues.
Teaching methods:	30% lectures, 70% hands-on exercises
Class size:	Maximum 25 participants
Required or suggested knowledge base for participants:	The course is designed such that participants with varied experience in epidemiological analysis can attend. Participants will be provided with a choice of exercises tailored to accommodate different levels of background knowledge and computer skills.
Course fee:	360 Euro + 120 Euro for accommodation per night, i.e. 600 Euro if arriving Sunday. It is possible to share rooms, see the registration procedure for details. There might be a 60 Euro fee for registering after January 31. The registration fee covers course materials and meals during the course.
Registration:	Registration is through the registration procedure for the SVEPM-conference. Please consult this page: http://www.svepm.org.uk/conferences.php?view=3 or if you are brave you can proceed directly to the registration, but this is really not encouraged.
Program:	A tentative program will appear here soon, but the course will essentially run from 9 to 17 on both days.