How to Foster Citizens’ Statistical Reasoning: Implications for Genetic Counseling

Abstract

Objectives: Our aim is to provide an overview of key research findings from cognitive psychology regarding effective ways of communicating statistical information, and to point out the implications of these findings for genetic testing. Method: We review the literature on the presentation of statistical information in diagnostic test results, discuss various representations that invite misunderstandings, and propose alternative representations that foster understanding. Results: Single-event probabilities, conditional probabilities and relative risks are easily misunderstood. Specifying the class of events to which a probability refers and using natural frequency statements improve understanding. Conclusions: Cognitive psychology has identified simple and effective tools for improving statistical reasoning. They can help to improve the public’s understanding of diagnostic test results.

Nutrigenomics: linking food to genome

“…It is mostly true for physicians, who are usually hesitant to provide advice in nutrition-related matters to patients due to lack of sufficient confidence in their nutrition knowledge (Kolasa, 2005). Therefore, practitioners must know how to understand and communicate sensitive information during counseling to patients on nutrigenomics (Kurzenhauser and Hertwig, 2006). Besides, sufficient knowledge about nutrition is required to advice patients about modification in their diets as well as how to quantify risk, to design test outcomes, and to deal with false negative or positive test results (Marteau, 1999), as these are highly susceptible to false impression, even by healthcare professionals.…”

Potentiel et limites de la «médecine individualisée» (personalized medicine)

“…En principe, toute prédiction (y compris géno mique) s'exprime sous la forme d'une probabilité. La transmission de probabilités est toutefois une entre prise complexe [5,6].…”

Potential und Grenzen von «Individualisierter Medizin» (personalized medicine)

“…Generell ist festzuhalten, dass es sich bei jeder Art der Prädiktion (auch der genombasierten) um eine probabilistische Aussage handelt. Die korrekte Vermittlung von Wahrscheinlichkeiten ist jedoch ein komplexes Unterfangen [5,6].…”