Epidemiologist Steven Goodman—Doctoring the Numbers

Monday, July 01, 2002
RELATED TAGS: CANCER

dialogue
Photograph by Rebecca Greenfield
The cancer news can make your head spin: Charred meat is dangerous, and this or that gene kills. Most recently, some studies have shown that mammograms may do more harm than good—even as some doctors promote the idea of whole-body screening to get an early jump on cancer. How to make sense of it all? Steven Goodman, a professor of oncology, biostatistics, and epidemiology at the Oncology Center at Johns Hopkins School of Medicine, has spent his career decoding the meaning of such medical statistics. He spoke with Discover reporter Jocelyn Selim.

What do you make of the media obsession with genetic and environmental risk factors?
Almost every disease involves both genetics and environment. So what does it mean if you have a gene that makes you susceptible to some environmental insult, and then you get the insult? What was responsible for your cancer? It's often said that cancer is 100 percent genetic and 100 percent environment. That's not necessarily a contradiction, because the cause is never a simple either/or situation. It's the interplay of the two factors that matters.

So should we just ignore ominous reports about burned meat or the BRCA1 gene?
I have a slide I use for presentations that lists supposedly legitimate, published cancer risks: being short, being tall, fluorescent lights, broken arms (but only in women), and my personal favorite, owning a refrigerator. When you take into account the design and the size of the studies, there's a lot less real information than people think. A lot of the claims are ludicrous unless you appreciate for the tenuousness of the numbers.

Cancer screening seems an obvious way to save lives. What is the controversy here?
Previously, we looked at the world in terms of people who were sick and people who were not. Screening technologies reveal a health state that is neither sick nor well but that could go either way. Let's say during a mammogram you find out you've got a bit of abnormal tissue. You get it biopsied and it looks cancerous, so you get treated. The problem is that we have no sure way of saying that the disease ever would have progressed into anything life threatening.

Still, haven't you reduced your risk by having the suspicious bit of tissue removed?
Probably, but it's uncertain by how much—and early diagnosis and treatment come at a cost. In the media you hear about this movie actress or that senator's wife who was "saved" by early detection; yet some proportion of those women actually had a form of cancer that never would have caused any trouble in their lifetime. For them the subsequent mastectomies or lumpectomies would not have been performed if they hadn't had a mammogram. This makes the balance of harm versus benefit a lot closer than most people think.

How do you measure that trade-off between harm and benefit?
That's best judged in large clinical trials in which we can compare death and surgery rates among women who were or were not screened with mammograms. For every 10,000 women aged 40 to 70, mammograms seem to prevent 0 to 10 breast cancer deaths—but they instigate 40 extra surgeries. Most people have heard about the debate over the lifesaving benefits from cancer screening. Almost nobody appreciates the surgery part, which changes the equation from "can't hurt, might help" to "might hurt, might help." That makes it a tough call, something that's been obscured by the drumbeat of unequivocally positive early-detection messages.

Where do we go from here?
I don't think the issues are so complex. We can make the statistics accessible and the issues clear. In fact, we have to. Then we can intelligently evaluate the true costs of different types of screening. Whole-body screening, for instance, is likely to cause a huge amount of harm even it it does have some benefit. This is snake oil for the 21st century.
Comment on this article
ADVERTISEMENT

Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

ADVERTISEMENT
ADVERTISEMENT
Collapse bottom bar
DSCJulyAugCover
+

Log in to your account

X
Email address:
Password:
Remember me
Forgot your password?
No problem. Click here to have it emailed to you.

Not registered yet?

Register now for FREE. It takes only a few seconds to complete. Register now »