Wc-021 Researching Diagnosis And Prognosis In General Practice

Donner-Banzhoff N. (Marburg), Griffiths F. (London), Herzig L. (Lausanne), Knotterus A. (Maastricht), Buntinx F. (Leuven)
Patients present in practice with symptoms that are often ambiguous.
Aetiologies range from benign self-limiting disease to serious, life-threatening conditions. The patient’s history, the physical examination and simple point-of-care tests help to reduce diagnostic and prognostic uncertainty. However, the evidence base for these is slim, especially for items of the history. A paradigm for the cross-sectional diagnostic study has been developed over the last 20 years. But how does this apply to general practice? The aim of our workshop is to discuss and to develop the design of diagnostic studies in general practice. The results of these studies can help GPs to deal with the complexities of the patient encounter and their own uncertainty. Brief presentations will introduce pertinent problems, pitfalls and suggestions. Participants will discuss their experiences and propose new solutions. Researchers who have gathered experience with diagnostic studies in primary care or who plan to conduct such studies are invited to take part (maximum: 30).
Topic no. 1: Complex reality – crude methods. GPs collect rich data from their patients during the consultation. These include not only what is said, but also how it is said, what the patients looks like, what the GP knows from previous encounters, and what is not said. All these are potential diagnostic and prognostic indicators, they may also interact with each other in complex ways. How can descriptive and evaluative study designs capture this reality? (introduced by Frances Griffiths, University of Warwick, UK)
Topic no. 2: The right test to choose. GPs use not only conventional items of the history and the physical examination. They have also developed rules and heuristics adapted to their working environment. Some of these have been investigated in diagnostic studies (introduced by Frank Buntinx, University of Leuven and Maastricht, B/NL).
Topic no. 3: Delayed-type reference standard: a practical experience. For low prevalence conditions standard study designs are often not feasible. Instead, the delayed-type reference standard has been suggested. The experience from a large diagnostic study (n = 1200) is reported (introduced by Norbert Donner-Banzhoff, University of Marburg, D).
Researching diagnosis and prognosis in general practice