Making statistical information for medical diagnostics easier to understand

Enlarge

Statistical information plays an important role in medical diagnostics. In the intersection between medical and mathematics education, LMU researchers have now investigated to what extent a new form of frequency visualization can improve the statistical understanding of medical students compared to forms of representation based on probabilities.

As the researchers report in the journal Medical Decision Making – Policy and Practice, students were shown the so-called frequency net. The results confirmed what was already known from earlier studies: medical students understand statistical information better when it is communicated as natural frequencies. That is to say, it is preferable to speak of “80 out of 100 patients,” for example, instead of using probabilities (80%).

“Our study was able to show that students can make diagnostic judgments with greater accuracy and speed when they are shown frequency nets or frequency double-trees. We hope that these techniques will be employed in the education of medical students in the future,” says Karin Binder, Professor at the Department of Mathematics at LMU.

Publication:

Alexandra K. Kunzelmann, Karin Binder, Martin R. Fischer, Martin Reincke, Leah T. Braun, Ralf Schmidmaier: “Improving Diagnostic Efficiency with Frequency Double-Trees and Frequency Nets in Bayesian Reasoning”, In: Medical Decision Making – Policy and Practice 2022