Risk-Illiteracy: An Expensive Shortcoming in the Health Sector
Health care doesn’t need more money — it needs better knowledge. Such is the sobering message in a new report by German psychology professor and director of the Harding Center for Risk Literacy, Gerd Gigerenzer.
In the report, Better Doctors, Better Patients, Better Decisions: Envisioning Health Care 2020, more than 40 authors from different fields of research and professions argue, that doctors and patients are “statistically-illiterate.” Doctors and patients (and journalists and politicians…) are unable to interpret the meaning of numbers and statistics, the authors say, and this causes serious wasted spending in health care — public as well as private.
“Wasted spending includes panic reactions that happen when patients are frightened by things they do not understand. A well documented example is the contraceptive pill-scare in the U.K. when women were told (by the Committee on Safety of Medicines, ed.), that if they were to take the new generation of contraceptive pills they would increase the risque of potentially lethal thrombosis with 100 percent. The women panicked and stopped eating the pill which led to an increased number of unwanted pregnancies and abortions. They never learned that 100 percent in this case meant the risque had increased from 1 in 7000 to 2 in 7000. This is just a very simple example of a basic statistical conflict. You have a relative risque (big numbers) which scares the women and an absolute risque (small numbers) that is never known by large parts of the public and this leads to unnecessary harms to the women and increased spending in the health care.”
The following year, the total number of abortions in England and Wales rose by 13,000 while expenses in public health care went up by 4-6 mill. due to the pill scare, says Gigerenzer.
PSA-screening and mammography are good examples
According to Gigerenzer, risk-illiteracy is a huge problem in the health sector. It affects the medical guidance that physicians give to their patients, the information found in health pamphlets about potentially life changing treatments, the way medical information is covered in the media and it has a big say in the way public spending is prioritized in the health care, he says.
Take statistics associated with breast cancer screening as an example.
Recently, a group of Danish scientists claimed that mammography (breast cancer screening) had had no effect on reducing the number of women who die from breast cancer in Denmark (link in Da.) This was troubling, they said, since the Danish public had been informed that the technique would reduce the number of deaths with a full 20-30 percent.
Such findings are important. But the interesting part about Gigerenzer’s report is that it explains how both of these seemingly contradictory statements — no effect and 20-30 less deaths — can, in fact, both be correct at the same time.
Another way to explain the effect of mammography is to say that 5 out of 1000 women die from breast cancer in regions without a screening policy. And that 4 out of a 1000 women die from breast cancer when screening is implemented on a large scale.
With a better statistical understanding with journalists, the public and with doctors, in other words, Danes could have known all along what those 20-30 percent actually meant.
Pros and cons
Saving 1 out of a 1000 patients is still saving a patient’s life. And you can’t put a price on that. But what are the potential harms associated with mammography?
That is another topic that is left out in most pamphlets and the media, says Gigerenzer.
According to Gerd Gigerenzer’s interpretation of the available evidence, 50-200 out of 1000 women are subjected to unnecessary biopsies when screening is implemented on a large scale. These are painful and causes severe psychological stress to the women and their families because both doctors and patients generally underestimate the number of false-positives associated with the technique.
Moreover, he says, it is estimated that 2-10 out of 1000 women have their breast removed partially or completely due to the discovery of cellular changes in the tissue that would have never made a difference to the patients’ health had the changes gone unnoticed.
Benign tumors detected at this early stage, to reiterate, are seldom given the benefit of the doubt but are treated as potentially malignant tumors.
- If you’d like to read more about Gerd Gigerenzer’s findings on risk-illiteracy — or if you would like to read about it in Danish — you can check out this article I wrote for Dagens Medicin (Daily Medicin.) This blog post is my personal take on the issue and should not be confused with Dagens Medicin in any way.
- If you wish to continue on your own or do not speak Danish, I would advice you to check out this site and download the first chapter of the book by Prof. Gigerenzer and J.A. Muir Grey (also available in Ge.) The topic of risk-illiteracy reaches much farther than panic-reactions and mammography (think about how much public spending is wasted in a modern administrative society that doesn’t understand statistics!)
- You should also check out the web page of the Harding Center for Risk Literacy (Harding-Zentrum für Risikokompetenz) at the Max Planck Institute in Berlin. There, or here, you can learn more about mammography and see how the Harding Center would present statistics on mammography to the public in a way it argues is transparent.
Before you go anywhere, however, there is something else you have to see. Please watch American Greg Powell land the world’s first Special Flip!
There can be no misunderstandings there.