Student Presentations

A simple test to determine if a recommendation is meaningful or just bullshit is to see if offers a real choice. Would you actually consider doing the opposite? Consider:

• get on the "learning curve"
• build capacities

• get off the "learning curve"
• destroy capacities

That aside, I want to make a point about cancer. There's a serious problem in how cancer statistics are interpreted by scientists, doctors, the media, and the public. I think Gird Gigerenzer's book was my first exposure to this paradox. It's this: cancer screening is not a purely positive thing, and actually may be a bad thing in some cases. Let's look at how this could be true:

Take a simple country with 1000 people, who live to age 70, and only one kind of cancer. If there is no cancer screening or cancer treatment, 10 will die of cancer, and all at age 60. Since there's no cancer screening, these ten cases are not discovered until they have severe symptoms, let's say at age 58. So the average survival duration after cancer detection is 2 years.

Now restart our clocks and add cancer screening, every two years starting at age 40. This time, 10 cases of cancer are discovered, all at age 54. Everybody gets treatment. They all die at age 60. The average survival duration after detection has risen to 6 years. But all that really happened was that ten people each spent an extra four years dealing with cancer. They didn't actually have longer, better lives.

Now, let's go one more time around, adding super-sensitive cancer screening. This time, 20 cases of cancer are discovered, all at age 50. Everybody gets treatment. Many go into remission, but ten still die of cancer at age 60. The rest die of other causes at age 70. The average survival duration after detection has risen from 2 to fifteen years! But in fact, nobody lived any longer than they would have without screening, and twenty people lived as cancer survivors for years or decades, having paid in money and blood and tears for treatment that didn't actually help.

Think this model is absurd?

AN Australian researcher says there's little evidence that prostate cancer screening saves men's lives.

And Professor Simon Chapman of the University of Sydney said a study of Australian newspaper and television stories about such screening for prostate cancer found most of them promoted it aggressively, ignoring the almost complete absence of evidence that it would save lives. — The Australian

The point is that not all cancers will kill you, at least not before something else will. We can detect cancers that we can't effectively treat, and we can't always differentiate between cancers that will kill you and cancers that won't. And it's a fallacy to say it's always better to be safe than sorry, because it doesn't work that way. False positive results, being told you have cancer when in fact you aren't slated to die from cancer, can lead to more than a little sorrow, especially if you undergo expensive and painful unnecessary treatment. The National Cancer Institute in the US says the same thing, but in a much more convoluted way:

At least two requirements must be met for screening to be efficacious:

1. A test or procedure must be available to detect cancers earlier than if the cancer were detected as a result of the development of symptoms.
2. Evidence must be available that treatment initiated earlier as a consequence of screening results in an improved outcome.

These requirements are necessary but not sufficient to prove the efficacy of screening, which requires a decrease in cause-specific mortality. For example, these two criteria are met in the case of screening for childhood neuroblastoma by assessment of urinary catecholamine metabolites. On the basis of these criteria, a mass screening program was conducted in Saitama Prefecture, Japan, between 1981 and 1992 for 6-month-old infants.[3] Over that 12-year period, the annual incidence of neuroblastoma in children younger than 1 year increased from about 28 per million to 260 per million but without a significant reduction in incidence in children older than 1 year. Because there also was no reduction in mortality for the disease, this experience provided strong evidence of overdiagnosis—diagnosis of some neuroblastomas detectable by screening, which would not have been clinically diagnosed later. Similar experiences have been reported elsewhere in Japan [4] and in the Quebec Neuroblastoma Screening Project (QNSP) in Canada.[5] The history of screening for neuroblastoma also provides a useful illustration of the benefit of undertaking well-designed evaluations of emerging screening technologies before implementing screening programs. Although such studies are very costly, it has been shown that the QNSP itself averted unnecessary morbidity for thousands of children and did so while returning a yield plausibly estimated at a cost savings 64.5 times the investment in the study.[6] —NCI

Cancer screening is a bad idea unless there's a test that finds cancers early, and treating these cancers early actually helps. Even then, screening may not be a good idea. They did a twelve-year test in Japan where they found way more brain cancer in infants under age 1, but cancer detection rates in older children didn't change and on average nobody lived any longer. So screening infants for brain cancer (at least, with that kind of screening and that kind of brain cancer) was a big waste. That 64.5 times savings they mention is, if you read carefully (and I had to check the abstract of footnote six to be sure I had it right), is the savings from scrapping unnecessary cancer screening programs, not the savings from performing screening. Bury the lede much?

Remember, these are general points. This is not a diatribe against all screening, or in favor of cancer. But it is clear that screening is not an unmitigated positive, and it's a big mistake to think it is. This is a tough point to make in the face of powerful individual appeals from survivors, but the underlying issue is the same as other kinds of medicine: individual testimonials are not data. If you have a mental picture of someone dying unnecessarily from a late diagnosis, you need to balance it with a mental picture of someone dying unnecessarily from treatment for a cancer that they don't actually have. Then you can put all of this emotion to the side and get back to evidence-based medicine. Put another way, humans are not wired to think accurately about statistics, and we need to remind ourselves of this weakness constantly.