Patients’ Preferences for Ways to Communicate Benefits of Cardiovascular Medication

This study is a landmark study: In the field of risk communication research, many data are based on focus groups with 30 to 40 students, physicians or patients. Goodyear-Smith et al studied almost one thousand patients and their preferences in risk communication.

Patients clearly prefer relative risk formats, independent on their literacy or numeracy (as pointed out in the letter of Hashim MJ). Relative risks meet our inborn need to compare with our neighbour: Are we better off or worse? Visual formats have the potential to enhance this comparison and improve understanding, and therefore we should use them routinely - here a disagree to the comment of Hashim.

For example, bar charts visualizing a person with and without risk allow the comparison to a reference point and are subjectively better understandable compared to numerical or verbal formats, as we showed in a recent study (Neuner-Jehle S et al, BMC Family Practice 2011,12:15). Bar charts could even be titled as “visual translation” of relative and absolute risk formats.

As Goodyear-Smith et al points out in the final discussion section, using relative risk formats is persuasive and may be manipulative. Adding absolute risk formats, though not the “first choice” of our patients, may prevent us from this danger. In summary, I believe that a mixture of different formats, presented as simple as possible, will be the first choice in future and should be investigated, in order to develop valid tools for risk communication in family medicine practice.

Competing interests:   None declared

I found the article by Goodyear-Smith et al to be interesting and relevant to the current debate on risk communication with patients.

This well-conducted study with about 1000 patients underlines the difficulty in resolving the issue of finding a single best mode of communicating risk estimates with patients. I agree with the authors’ cautionary statement: “Patients, however, still may have difficulty understanding descriptions of treatment benefit regardless of presentation, or simply prefer clinical opinion.”

Perhaps the underlying issue has more to do with our aversion to consciously estimate future risk in a tangible manner. Patients would rather remain in the dark than think of issues such as own death and disability.

I would like to highlight some interesting findings for designers of health education materials:

1. Pictures/graphs do not seem to add much value over numbers.

2. A simple bar graph is as good as an intricate people chart.

3. Positively framed statements are somewhat more preferred

4. Relative risk reduction is preferred (but as the authors indicate is susceptible to misleading statements). NNT seems to be the least preferred.

That education, numeracy score and socio-economic status did not affect the preferred mode of risk communication is, I think, the most startling finding of this study. For years, we have been taught that patients with limited literacy had a different approach to risk perception compared to more educated persons. We need to re-evaluate health education strategies based on these findings.

Competing interests:   None declared

Wouldn't it be nice to have a single yardstick when explaining risk reductions to patients? In the present study patients found relative risk reduction more helpful and encouraging than other effect formats when considering a risk reducing drug therapy, but Goodyear-Smith et al rightly reject the idea of using this or any other effect format as a single yardstick. In the recent past there were claims that the number needed to treat (NNT) might be easier to understand, but several empirical studies seriously contested these claims. Some ways of communicating statistical information may be more transparent than others, but even if a patient truly understands the concepts of baseline risk and relative risk reduction, she might still find it difficult to judge whether a relative risk reduction of 33% is good or bad. This may require knowledge, experience or well formed preferences that many patients simply do no have. Therefore it is perhaps not surprising that more than half of the patients in the present study preferred the doctor's opinion to statistical information. Straight talk about numbers may be a good way to involve patients in the decision making process, but if we insist that the numbers should speak for themselves, there is a risk that patients might feel abandoned. The task of providing an honest opinion along with transparent information without being manipulative is, however, far from trivial.

Competing interests:   None declared

We agree with Dr Hahn that it is important to inform rather than persuade patients.

The question being asked, if we understand correctly, is what eventual decision respondents might come to in relation to each manner of expressing the same risks. We think this is addressed in our Figure 1, which shows remarkably similar response patterns to questions about 'encouraging' to take medication and 'making a decision' to take medication. We interpret this to show that respondents did not distinguish between these concepts, i.e. that information that encouraged them also helped make a decision, and this decision would presumably be to take the medication. This is perhaps all the more concerning to practitioners who wish to deliver information that informs without persuading.

Competing interests:   None declared

Goodyear-Smith et al present findings from New Zealand in which they present different methods of communicating CVD risk and ask patients which ones would encourage them to take medications daily. The patient's answers only explain < 10% of the variance and thus do not provide a convincing argument for any preferred communication strategy, which the authors acknowledge. We studied this question using focus groups and found that a cardiovascular age metaphor was the most intuitive and promising amongst US participants.(Ann Fam Med 2006;4:205-12) We then tested this in a pilot study using a quasi-experimental design and demonstrated benefit in the number of patient's at cholesterol goals after one year using a PDA decision support tool that incorporated this "heartage" metaphor along with pictures that allowed for relative risk and absolute risk reduction to be discussed.(J Med Pract Manage 2009;24(4):224-30). A RCT testing this strategy in southeastern New England has been completed and a manuscript has been submitted. Using a similar strategy of cardiovascular age, Grover et al in the CHECK_UP study, demonstrated improved LDL lowering in dyslipidemic patients in Canada. (Arch Intern Med. 2007;167(21):2296-2303). So while I agree that not single risk communication strategy works for all patients, use of a cardiovascular age metaphor appears promising and is worthy of further study.

Competing interests:   None declared