The 1-in-X effect on the subjective assessment of medical probabilities

Abstract

Among numerical formats available to express probability, ratios are extensively used in risk communication, perhaps because of the health professional’s intuitive sense of their clarity and simplicity. Moreover, health professionals, in the attempt to make the data more meaningful, tend to prefer proportions with a numerator of 1 and shifting denominators (e.g., 1 in 200) rather than equivalent rates of disease per unit of population exposed to the threat (e.g., 5 in 1000). However, in a series of 7 experiments, it is shown that individual subjective assessments of the same probability presented through proportions rather than rates vary significantly. A 1-in-X format (e.g., 1 in 200) is subjectively perceived as bigger and more alarming than an N-in-X*N format (e.g., 5 in 1000). The 1-in-X effect generalizes to different populations, probabilities, and medical conditions. Further-more, the effect is not attenuated by a communicative intervention (verbal analogy), but it disappears with an icon array visual aid.