| To Register SMDM Homepage | |
|
|
|
||||
Purpose: Many clinical decisions require patients to weigh the risks and benefits of available treatments. In order to facilitate patients' ability to make good decisions about their health care, it is critical that clinicians communicate these risks and benefits so that patients can decide which alternative best fits their preferences. However, communicating detailed risk information can be challenging, because many people have difficulty understanding concepts like percentages and frequencies, a phenomenon known as innumeracy. Moreover, health care risks may be harder to think about and interpret than other types of probabilistic information. We conducted an experiment testing whether people’s numeracy is influenced by the context of the numerical quiz.
Method: This study reports data from a random subset of the Health and Retirement Study which surveys more than 22,000 Americans over the age of 50 every two years. 1010 participants answered four numeracy questions. We randomized subjects to one of 3 contexts: 1) numeric context: e.g. "What is 15% of 1000?", 2) health context: e.g. "If the chance of getting a disease is 15%, how many people out of 1000 would be expected to get the disease?", or 3) retail context: e.g. "If you were to buy a television that was discounted 15%, how much money would you save if the original cost was $1000?."
Results: Numeracy was lower in health contexts than in the other two contexts, with 10%-20% fewer people being able to answer the health questions correctly. The mean proportion of people answering the health context questions correctly ranged from 23%-71%, compared to 36%-87% and 45%-82% in the numeric and retail contexts. Accuracy was lower in respondents over 65, women, non-Caucasians, and those with a high school education or less (all p's<.0001). People who rated their health as fair or poor had significantly poorer numeracy than did those who rate their health as good or very good and this relationship remains after controlling for age, education, gender, and race (p<.001).
Conclusions: Many people have difficulty understanding risk information in health contexts, especially people in poor health. This is a potentially large barrier to successful risk communication. Future research should test whether comprehension is improved when numerical information is explained using more familiar contexts.
See more of Poster Session - Utility Theory; Health Economics; Patient & Physician Preferences; Simulation; Technology Assessment
See more of The 26th Annual Meeting of the Society for Medical Decision Making (October 17-20, 2004)