Randomised, controlled crossover pilot study on the efficacy of Intranet-based guidelines for General Internal Medicine at the Park-Klinik Weissensee in Berlin. As part of the evaluation programme for the intranet-based internal guideline system at the Park-Klinik Weissensee in Berlin, an experimental randomised, controlled trial was conducted in which 12 medical students in their final year were asked to solve 10 clinical vignettes each. A number of written and electronic sources of medical information was available. In addition, the students were allowed to use the internal guideline system for one half of the vignettes in a crossover fashion. There was a sample size-related non-significant trend in favour of the guideline system concerning the primary endpoint "time to right solution" (mean 2.8 min; 95% CI -1.6 to 7.3). A per-protocol-analysis revealed a great variability in the actual use of the guidelines available. There was an inverse "dose-response relationship" between the intensity of guideline usage and the endpoint.
