Being Human Means Living with Risk
You weigh risk and reward when you buy a stock, choose a life insurance policy, gamble at a casino, drive through a yellow light or take a pill. Both decisive action and inertia have consequences.
However, few choices are more daunting than being told that you or a loved one needs surgery. Much of the communication associated with potential surgery involves brief conversations with doctors, sometimes unexpectedly, followed by anxious, online searching of medical terminology. Increasingly, the decisions involved are not just surgery vs. medication but choosing among several alternative procedures and different health care providers.
I maintain that most of us overestimate our ability to understand risks. Even with formal education, the intrusion of emotion and the number of factors to be considered confound us. There is a pressing need to study the best methods for presenting risks in an unbiased and useful manner. A few studies I have found provide some direction, but for a topic of such individual and social magnitude, it is amazing that best practices have not been established and adopted.
Among the underlying issues that need to be evaluated:
- Absolute versus Relative Risk
- Number Needed to Treat (NNT) statistics
- Visual versus verbal presentation of information
- “Denominator neglect”
- Average Risk vs Individual Risk
- Terminology and its affect on perception of danger or pain
If a surgeon says you have a 3 percent chance of dying during a surgery how do you translate that into a useful nugget for decision-making? Do you picture that there are 100 people in the hospital having your surgery and that 3 of you die? 97 live? Or, do you picture that your chance of being killed in a motor vehicle crash is 1 in 113 so you are only slightly more likely to die in the surgery than in the car on the way to the hospital? Would seeing an “icon array” of 100 people-like figures with 3 of them in black affect your perception of your own risk?
People also need to understand Absolute Risk (your chance of developing a disease over a specified period of time) versus Relative Risk (comparing risk in two different groups of people). If I have a 2 in 20 risk of a disease without treatment (absolute risk) and treatment reduces Relative Risk by 50%, it sounds very significant. Is it as compelling when the Absolute Risk drops from 2 in 20 to 1 in 20? Similarly, the popularly used, “Number Needed to Treat,” is a difficult concept to grasp (the number of people who need to take a treatment for 1 person to benefit).
As humans, studies show that we are quite fallible. We are most persuaded by the last statistic we hear in a conversation (if the doctor describes risks and then benefits, the benefits stick with us). We fall prey to “denominator neglect,” we think 1200 out of 10,000 people having a complication is worse than 24 out of 100 people, simply because we don’t process the total sample size.
The terminology chosen to describe procedures can also have quite a significant effect on perception of discomfort and expected length of recovery. For example, the ubiquitous term “minimally-invasive procedure” suggests to many--no cutting, no bleeding, no complications. In reality, it is surgery, but the size of the incision is limited. Similarly, the procedure referred to as “virtual colonoscopy” (CT Colonography) sounds like it takes place on a screen and does not communicate pumping air into the colon via a tube in the rectum.
Information designers, plain language writers and ethnographic researchers need to team up with medical professionals to develop and test informational methods for communicating risk. In this case, perhaps there is no risk, only reward.