Tag Archives: availability heuristic

The Resta Paradox: On The Perception of The Likelihood of Rare Events

Genetic counselors – and their clients – tend to make a big to-do about numbers. We expend a disproportionate amount of energy  calculating, reciting, and explicating numbers such as the complication rate of amniocentesis, the likelihood of Down syndrome, bayesian probabilities, or empirical recurrence risks. Sometimes we even pull them out of thin air. Those statistics pass through patients’ psychological, cultural, social, and educational filters and out of the other side comes a figure that more often than not bears little resemblance to the number that went in.

Counseling strategies attempt to reduce anxiety by downplaying large numbers when it comes to undesirable outcomes (“A one in four recurrence risk means a 75% chance that it will not happen again”) and reframing small numbers for patients who perceive a low likelihood outcome, such as 1/500, to be a very “high” number (“One in five hundred means there is over a 99% chance this will not happen. If you were to score a 99+% on a difficult test in school, you would be very happy. That’s an A+!”).

But reframing can sometimes backfire. For many patients, rare events are paradoxically perceived to be more likely to occur than high likelihood events. The rarer the event, the more some patients are convinced that it will happen to them. This phenomenon, which I have immodestly labeled The Resta Paradox,* is a sort of corollary to Murphy’s Law.

Just about every genetic counselor has heard some variation of this statement “I know that one in 50,000 is pretty unlikely, but if it’s going to happen to someone, it’s going to happen to me.” To demonstrate how he or she is one of those people whom the Gods have chosen to be the object of their malefic whims, patients will then cite a litany of rare catastrophes that have previously befallen them:

  • “My surgeon said she never had a patient develop a complication after surgery, but I got an infection.”
  • “One time, a piece of an airplane broke off, fell put out of the sky and crushed my car.”
  • “I took lisinopril and developed a never before reported side effect and I was in the hospital for a week. They even wrote an article about me” (The patient will then look at you, hoping that you will say that you were familiar with the article).

The seemingly contradictory availability heuristic for these patients is “Rare events have happened to me before. Therefore, I am likely to experience rare events in the future.”

In my favorite example of how reframing can sometimes spectacularly go awry, a colleague once described to me a genetic counseling session in which she told the patient that the risk of recurrence of the particular disorder was as likely as getting struck by lightning. The patient replied “As a matter of fact, I have been struck by lightning.”

Such magical thinking is surprisingly common. It does not necessarily mean that a patient suffers from innumeracy. Most people are perfectly capable of balancing their checking accounts, completing income tax forms, and accurately summing long columns of numbers. Complications arise, though, when patients try to assign an emotional value to a number or a risk, especially when trying to make a complex medical decision.

I have also encountered The Resta Paradox among patients who make their living off of understanding numbers, such as statisticians, engineers, and epidemiologists. Even these professionals have a hard time comprehending rare events and very large/small numbers in psychologically and personally meaningful ways.  Life is complicated and highly unpredictable, and we all try to make sense of it however we can, sometimes in seemingly inconsistent ways.

And you can call me Mr. Guilty too. At my institution’s Tumor Boards, I am infamous for frequently stating that I worry most about my low risk patients who undergo BRCA testing, since they always seem to test positive for a mutation while my highest risk patients almost never carry a mutation. Interestingly, in support of my paradoxical thinking, BRCA risk carrier prediction models tend to under-predict carrier status in low risk patients and over-predict carrier status in high risk patients, so perhaps my thought process is not entirely magical. Or maybe computers engage in magical thinking too.MR900283911

On a deeper psychological level, believing oneself to be a statistical outlier may be part of the general impulse to view oneself as special, a way of saying “I have an unusual trait that makes me different from others in an interesting way.” It may also be a defense mechanism to psychologically prepare for a bad outcome so that, should it occur, the individual is better prepared to deal with the stress.

The Resta Paradox serves to remind us of a lesson we seem to need to be reminded of repeatedly: Numbers, though a critical component of many genetic counseling sessions, are not the endpoint by which to measure the effectiveness of genetic counseling, but rather are the point at which genetic counseling begins. It’s not the number that matters; what is important is how and why that number matters to the patient.

* – Hey, it’s a lot better having an insightful paradox named after me than a medical syndrome.


Filed under Robert Resta