Rapid Responses to:
|
|
Rapid Responses: Rapid Responses published:
-
![[Read Rapid Response]](/icons/misc/sectionDOWN.gif)
We need more evidence of effectiveness about ways to communicate risk to patients - a proposal.
- Victor M. Montori, Cheryl Carling, Holger Schünemann, Jeph Herrin, Jan Arve Dyrnes, Shaun Treweek, Doris Tove Kristoffersen, Elie Akl, Gordon Guyatt, PJ Devereaux, Victor Montori, and Andrew David Oxman (3 October 2003)
-
Relative risk reduction; Absolute risk reduction and number needed to Treat: RAT DIAGRAM
- Jaishen Rajah (4 October 2003)
|
|
|||
|
Victor M. Montori, Research Fellow McMaster University, Hamilton, Canada L8N 3Z5, Cheryl Carling, Holger Schünemann, Jeph Herrin, Jan Arve Dyrnes, Shaun Treweek, Doris Tove Kristoffersen, Elie Akl, Gordon Guyatt, PJ Devereaux, Victor Montori, and Andrew David Oxman
Send response to journal:
|
The article by Paling1 and other papers on the same issue of the BMJ provide a number of suggestions and practical tools to aid in the communication of risk. While we found the tools and suggestions attractive and compelling, we would like to draw readers’ attentions to the paucity of evidence supporting their effectiveness and the effectiveness of alternative approaches. In particular, we know little about the communication of the uncertainty associated with research estimates and of ways to communicate this information to promote decisions most consistent with patient values and preferences. That is why we have embarked on an international Internet-based series of randomized trials testing different approaches to risk communication (Health Information Project: Presentation Online or HIPPO). Our aim is to find out which ways of presenting information about the effects of healthcare best help people to make choices that are consistent with their own values. The first HIPPO study compares different ways of presenting the reduction in risk of heart disease to people who are asked to make a decision about whether to take cholesterol-lowering medication. Future HIPPO studies will compare different ways of presenting information to people making choices such as whether to have surgery for herniated disc with sciatica, about other treatments for back problems and other conditions, and about preventive interventions, such as screening tests. Healthcare professionals, patients, and general public interested in participating in the first HIPPO study should go to the study website at http://www.icru.no/hippo/cholesterol. Additional information can be found at the HIPPO website or by contacting Cheryl Carling (cheryl.carling@shdir.no) at the Directorate for Health and Social Affairs in Norway or Holger Schünemann (hjs@buffalo.edu) at the University at Buffalo, New York. The HIPPO Investigators Cheryl Carling, RN, MSc, Research Fellow1
1. Department of Health Services Research Directorate for Health and Social Affairs PO Box 8054 Dep N-0031 Oslo Norway 2. Departments of Medicine and Social and Preventive Medicine University at Buffalo 270 Farber Hall 3435 Main St Buffalo, NY 14214 USA 3. Flying Buttress Associates PO Box 2254 Charlottesville, VA 22902 USA 4. Department of Clinical Epidemiology and Biostatistics McMaster University Room 2C12 Hamilton, Ontario, L8N 3Z5 Canada 1 Pailing, J. Strategies to help patients understand risk. BMJ 2003;327:745-748 Competing interests: The authors are investigators of a study assessing how to communicate medical information, including risk, to patients. |
|||
|
|
|||
|
Jaishen Rajah, Consultant Pediatrician Sheikh Khalifa Medical Center, Abu Dhabi, 51900, UAE
Send response to journal:
|
A picture is not woth a thousand words , but a million words, if the words are couched in terms of conditional probabilites, relative risk and other misleading statistics. Paling (1) and Gigerenzer (2)lead us away from statistical innumeracy through simple visual aides and a heuristic model of thinking respectively. Imagine a messenger from heaven sent to deliver a news item. There is only one problem: the angel is tone deaf. Upon hearing the distorted message, you shoot the messenger. The analogy is that many doctors, when transmitting the evidence to themselves and worse still to their patients, deliver such messages. We could not be blamed for thinking that the angel were in fact the devil. What are our patient to think of us? While on the subject of devils, may I play devils advocate and suggest (somewhat tongue in cheek!)the use of a diagram called the RAT diagram (or I smell a RAT diagram. Relative risk reduction, Absolute risk reduction and number needed to Treat). The pervasive use of selectively using RRR (relative risk reduction) to exaggerate benefit and ARR (absolute risk reduction) to minimise harm could be reduced if authors were forced to depict such a comprehensive figure. The Y axis would be the RRR (0;25;50;75;100%). On either side would be 2 X axes. The bottom X axis would be NNT (number need to Treat) (1;25;50;100) An arrow below would depict 1 to infinity from L to R. The top X axis would be ARR. Since NNT and ARR are reciprocals they could easily be depicted opposite each other. The arrow above the ARR (from R to left) would run opposite to that of NNT ie. from R to L(1% ARR : NNT 100; 1.3% ARR :NNT 75; 2%ARR:NNT 50; 4%ARR:NNT 25 and so forth). The plot would be a single dot linking a 3 points. 1. Paling J. Strategies to help patients understand risks. BMJ 2003;327: 745-8 2. Gigerenzer G, Edwards A. Simple tools for understanding risks: from innumeracy to insight. BMJ 2003;327: 741-4.[ Competing interests: None declared |
|||