Participants
300 medical students
Interventions
two characteristics of risk communication in the context of a decision regarding prenatal diagnosis for cystic fibrosis are manipulated: verbal versus numerical probabilities and the negative versus positive framing of the problem (having a child with versus without cystic fibrosis).
Main results
Independently of the manipulations, most students were in favor of prenatal diagnosis. The effect of framing was only significant in the conditions with verbal information: negative framing produced a stronger choice in favor of prenatal diagnosis than positive framing. The framing effect in the verbal conditions and its absence in the numerical conditions are explained by the dominance of the problem-occurrence orientation in health matters as well as a recoding process which is more likely to occur in the numerical (the probability "1-P" switches to its counterpart "P") than in the verbal conditions. The implications for the practice of genetic counseling are discussed.