Quite often patients ask me to predict the future outcome of their illness: “How long can I expect to live with this illness, doctor, or how dangerous is the procedure that you are recommending?” A facile answer involves quoting data taken from clinical or epidemiologic studies, but this kind of information is rarely useful for the average patient who has not had a great deal of experience with statistics in any form, and especially the kind of biostatistics that healthcare workers encounter on a daily basis. In addition, many individuals are skeptical about statistics as noted in the quote that heads this editorial. “Lies, damned lies, and statistics” was popularized by Mark Twain in the United States, but the quotation has a much longer history and was not created by Twain.1 Of course, it implies that one can hide the truth by quoting statistics.
Statistical percentages quoted to patients can easily be misinterpreted and lead to considerable patient anxiety. For example, once a surgeon told one of my patients that the planned operation involved a 5% risk of death. When I talked with this same patient later that day, I discovered this information had made the patient extremely fearful about the upcoming surgery. When I told this individual that, in fact, the odds were in his favor at a rate of 20:1, he was greatly relieved and did well at surgery the next day. Thus, the same statistic presented differently resulted in a markedly different psychological outcome.
Unfortunately, the vast majority of patients have never studied statistics and do not use them in their daily work or home lives. Healthcare workers, on the other hand, are deluged with statistics every day involving hospital admissions and census, mortality and morbidity rates associated with a particular illness or procedure, epidemiologic reports, and so forth. How we communicate this information to our patients requires, in my opinion, careful translation into “lay speak.” The example above involving the surgeon is not a rare occurrence in daily hospital rounds. It is essential that we help patients understand the concept of risk and that we explainit in a manner that is easily grasped. Over the years I have developed a simple explanation for various statistical data that enables patients to understand risk and outcome percentages despite the fact that they have never had formal training or education in statistical methods.
My approach involves what I call “the horse track analogy.” Even if one has minimal understanding of statistics, just about every American understands the odds involved in horse racing. The moment I tell patients about the 100:1 long shot versus the 2:1 favorite, I can see that we are standing on common ground. In the surgical scenario quoted above, it would have worked best if the surgeon had stated that the patient was like a favorite horse at the local track with a 20:1 chance in favor of winning. I use the same analogy when talking with patients about taking evidence-based medication. In an attempt to increase patient adherence to a specific medical regimen, I tell them the horse track analogy and urge them to bet their well-being on the “favored horse” and not on the 100:1 long shot. I often continue this line of reasoning by telling patients that it is their money and their life we are talking about, so “please bet on the favorite horse” and take the prescribed medications.
I have not done a careful statistical analysis of whether the horse track analogy works by increasing patient adherence to a specific medical program, but patients do tell me that they understand the message tied to this comparison. I am sure that other analogies could be applied in the same manner, for example, the odds involved in a variety of sports events in which one team is the overwhelming favorite to win a particular game. I hope that others will be convinced by this essay to use the horse track analogy in their daily clinical activities.
As always, I welcome communications about this editorial on our blog at amjmed.org.
To read this article in its entirety please visit our website.
-Joseph S. Alpert
This article originally appeared in the November 2017 issue of The American Journal of Medicine.
