Q J Med 2003; 96: 455
© 2003 Association of Physicians
Correspondence |
Communication of risks: an analysis beyond numbers
Mayo Clinic
Rochester
USA
e-mail: ghosh.amit{at}mayo.edu
Sir,
The recent commentary by Professor McManus1 provides an excellent overview of Dr Gigerenzer’s book, and summarizes the overwhelming appeal of ‘natural frequency’ over traditional descriptions of risks as probabilities. Dr Gigerenzer has earlier demonstrated that medical students and physicians ability to understand information of risks presented as probabilities is inferior to presentation of data in the ‘natural frequency format’.2
The overall communication of risks, however, remains more complicated than mere understanding of numbers. Risks which are not undertaken by choice are perceived to be more threatening than those that the individual undertakes willingly.3 We have earlier reported that there is a lack of consensus in deciding the best technique of communication of medical risks.4 A literature search of the Medline (1980–2002), Embase(1988–2002), PsycInfo (1984–2002), ERIC(1967–2002), CINAHL (1984–2002), Web of Science and educational web-sites, using the terms ‘risk communication’, and ‘modalities’ revealed a preference for both qualitative and quantitative methods of risks communication. Among the quantitative methods, risks were presented as relative risk reduction (RRR), number needed to treat (NNT), absolute risk (ARR), pictorial display of risk, survival curves, decision aids, natural frequency, Calman Chart Risk and acceptable regret technique. Patients expressed an increased desire to follow therapy when risks were presented as RRR rather than NNT or ARR. Qualitative methods of communicating risks included focus groups using video- and audio-tape techniques, closed-circuit TV, and questionnaires. We also detected that the risk-taking behaviour of physician and patients determined the communication styles. Techniques of presentation, errors in risk estimation, educational and cultural differences also determined the overall perception of risks. Hence it is likely that perception and communication of risks could often involve complex cognitive processes, which could go beyond numbers. The presentation of risks as natural frequency, is a contribution towards the mutual understanding of a test’s results.
References
1. McManus IC. The arithmetic of risk. Q J Med 2003; 96:235–8.
2. Hoffrage U, Lindsey S, Hertwig R, Gigerenzer G. Communicating statistical information. Science 2000; 290:2261.
3. Slovic P. Perception of risk. Science 1987; 236:280–5,
4. AK Ghosh, P Erwin, K Ghosh. Effective Risk Communication: Multiple modalities, unclear consensus—A review of literature. J Invest Med 2002; 50:182.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||