How Do People Perceive Health-Related Risk Ratios?

As a mathematical ratio, 1 in 10 is exactly the same as 10 in 100. Psychologically, however, the two expressions may be perceived differently, particularly when health is involved.

According to a study in Journal of Experimental Psychology: Applied (2018), people perceive health-related risks as more likely when health professionals communicate the risk as “1 in x” ratios than when they receive the same information expressed as “N in N × x” ratios.

In other words, the risk of 1 in 10 looms larger than the risk of 10 in 100. This is because people overestimate the actual probability of events when they are communicated as “1 in x” ratios.

Authors Miroslav Sirota, Marie Juanchich, and Jean-Francois Bonnefon asked participants to assess the probability of contracting a disease using different scales: subjective probability scales that use vague expressions as “very low” and “very high” probability, but also objective scales such as a probability scale and a frequency scale that use precise numerical values.

Using objective scales allowed the authors to assess how close or far people’s estimates were from the actual numerical values in different ratio conditions.

Sirota, who led the study, explains, “Based on previous research, we knew that the ‘1 in x’ ratios were subjectively perceived as conveying a higher probability than other formats, such as percentages and ‘N in N × x’ ratios. But we did not know whether this happens because the ‘1 in x’ ratios lead to an overestimation of the objective probability or because the other formats, such as the ‘N in N × x’ ratio, lead to an underestimation.”

Sirota and colleagues found that people overestimate the objective probability more when they receive the risk information in the “1 in x” format. This “1 in x” bias is limited to absolute judgments, when people separately judge how large or small the presented risk is (e.g., receiving information about a risk of a side effect), not when they are asked to compare two ratios (e.g., choosing between two treatments qualified by different ratio formats).

The “1 in x” bias matters because people might change their decisions based on this distorted risk perception. Indeed, in the study, people in the “1 in x” ratio conditions made different decisions than those assigned to the “N in N × x” ratio conditions.

To help people make informed decisions, health professionals should communicate risk information in formats that are more aligned with the objective probability.

Fortunately, changing the risk communication should be straightforward. Health professionals should avoid using “1 in x” ratios to communicate risk, so their patients can make informed decisions.

Citation

• Sirota, M., Juanchich, M., & Bonnefon, J.-F. (2018). “1-in-X” bias: “1-in-X” format causes overestimation of health-related risks. Journal of Experimental Psychology: Applied, 24(4), 431–439. http://dx.doi.org/10.1037/xap0000190

Note: This article is in the Basic / Experimental Psychology topic area. View more articles in the Basic / Experimental Psychology topic area.

Miroslav Sirota, PhD, is a reader in psychology at the University of Essex. Miroslav’s research focuses on how people perceive, reason about and communicate risk as well as how this affects their decision-making, especially in a medical domain. He is also interested in understanding sociocognitive processes underlying inappropriate antibiotic expectations and unnecessary antibiotic prescribing.

Marie Juanchich, PhD, is a senior lecturer at the University of Essex. Marie’s research aims to empower people to make better decisions for themselves, for others, and for the environment. Most of her work focuses on improving uncertainty and risk communication but she also conducts some behavioral change studies on how to encourage pro-social and pro-ecological behaviors.

Jean-Francois Bonnefon, PhD, is a research director at the French Centre National de la Recherche Scientifique, affiliated to the Toulouse School of Economics, the Toulouse School of Management, the Institute for Advanced Study in Toulouse, and the Artificial and Natural Intelligence Toulouse Institute. He conducts research on decisions which have a moral component, especially in the context of machine ethics and human-AI cooperation.