Shots - Health Blog
3:22 pm
Mon May 28, 2012

With PSA Testing, The Power Of Anecdote Often Trumps Statistics

Originally published on Tue May 29, 2012 9:46 am

Millions of men and their doctors are trying to understand a federal task force's recommendation against routine use of a prostate cancer test called the PSA.

The guidance, which came out last week, raises basic questions about how to interpret medical evidence. And what role expert panels should play in how doctors practice.

About 70 percent of men over 50 have gotten a PSA blood test. Some are convinced it was a lifesaver.

Tom Fouts of Florida is one of them. He and his doctor had been watching his PSA (prostate-specific antigen) creep up for almost two years. Fouts was losing sleep over it, wondering if it meant a silent killer was incubating in his prostate gland.

Finally, he decided to act. After three painful biopsies, doctors discovered a moderate-grade cancer and Fouts had surgery to remove it.

Today he's fine. "I'm a firm believer the PSA test has saved my life," he says. And he doesn't think much of the U.S. Preventive Services Task Force, the government-appointed expert panel that advised against routine PSA testing after analyzing reams of statistics.

"My theory on statistics," Fouts says, "is anybody can look at the same stats and come up with their own opinion. Government does it; each political party does it. Whatever you want it to come up to read, you can fine-tune it and make it come up to that."

Hal Arkes, a psychology professor at Ohio State University, says Fouts' way of thinking is nearly universal. The power of the anecdote almost always overwhelms statistical analysis, he says.

"Statistics are dry and they're boring and they're hard to understand," Arkes tells Shots. "They don't have the impact of someone standing in front of you telling their heart-rending story. I think this is common to just about everybody."

Arkes says anecdotal thinking "contributes to the widespread gross over-estimation of the benefits of PSA screening." He suggests people do a mental exercise to understand what the numbers are saying about PSA:

Imagine an auditorium filled with 1,000 men who had PSA screening tests and another auditorium with 1,000 men who didn't. That represents the kind of studies the federal task force was relying on.

"Take a look at the men in the two auditoriums, the men in the screened and the men in the not-screened auditorium," Arkes says. "There's just as many men who died of prostate cancer in each auditorium, which leads us to think in the aggregate it didn't do any good."

Arkes breaks it down in the journal Psychological Science.

In each auditorium, there would be eight men who died of prostate cancer. But among the thousand who got PSA tests, there would also be 20 men who were treated for prostate cancers that would never have grown and caused symptoms. And five of these needlessly treated men would have lifelong complications, such as impotence and incontinence.

Dr. Ian Thompson of the University of Texas Health Science Center at San Antonio says Arkes "is exactly correct – but only according to the current clinical trials of PSA screening."

"If you leave them the way they are, that article is smack-on correct," says Thompson, a urologist. "But the trials had problems."

Thompson says the chief issue is that men in the best study to date, from Europe, have been followed for a maximum of 13 years — and that's not long enough.

"When you analyze those trials very early, what you pick up on are the harms of testing," Thompson says. "And it really takes many, many years to see the benefits."

He doesn't know how much longer the European men would need to be watched, but thinks it would eventually become clear that PSA testing saved many more from a prostate cancer death. He doesn't think the Preventive Services Task Force should have taken a stand against testing at this time.

Dr. Michael Barry, head of the Informed Medical Decisions Foundation in Boston, thinks Thompson has a point.

"I'm reluctant myself to make a decision for someone else about PSA screening," Barry says. "And as a result, I'm also reluctant for expert panels to take that position of telling men what to do here."

Barry says his way out of the controversy is to take it "one man at a time." That is, doctors need to lay out the evidence as clearly as they can, which he says indicates there's very little, if any, benefit to PSA testing.

"I think many men won't want the test in that circumstance," he says. "But some will, and I'm comfortable with that."

Up to now, Barry says, those discussions haven't been happening nearly enough.

Copyright 2013 NPR. To see more, visit http://www.npr.org/.

Transcript

MELISSA BLOCK, HOST:

This is ALL THINGS CONSIDERED from NPR News. I'm Melissa Block.

Millions of men and their doctors are trying to understand new recommendations by a federal task. They advise against routine use of a blood test for prostate cancer test, the PSA test. The guideline raises some bigger questions: How should we interpret medical statistics and what role should expert panels should play in determining how doctors practice?

NPR's Richard Knox explains why it's so hard to answer those questions.

RICHARD KNOX, BYLINE: Tom Fouts and his doctor had been watching his PSA level creep up for almost two years, and he was losing sleep over it.

TOM FOUTS: My fear is in the middle of the night that this, you know, this is a silent killer. What if? What if? What if?

KNOX: Finally, he decided to act. Doctors did three painful biopsies and discovered a medium-grade cancer. He had surgery to remove it. Today, Tom Fouts is fine and he's convinced the PSA test was a lifesaver.

FOUTS: I'm a firm believer the PSA assay saved my life.

KNOX: And he doesn't think much of the U.S. Preventive Services Task Force, which last week recommended against routine PSA testing after analyzing reams of statistics.

FOUTS: My theory on statistics is anybody can look at the same stats and come up with their own opinion on it. Government does it. Each political party does it. Whatever you want it to come up to read, you can fine-tune it and make it come up to that.

KNOX: Hal Arkes says Fouts' way of thinking is nearly universal. Arkes is a professor of psychology at the Ohio State University.

HAL ARKES: Statistics are dry and they're boring and they're hard to understand, and they don't have the impact of someone standing in front of you telling their heartrending story. So I think this is common to just about everybody.

KNOX: The power of the anecdote almost always overwhelms statistical analysis, he says. Arkes suggests people do a mental exercise to understand what the numbers are saying about PSA.

Imagine a thousand men who had PSA screening tests in one auditorium, and a thousand who didn't in another. That represents the kind of studies the federal task force was relying on.

ARKES: Take a look at the men in the two auditoriums, the men in the screened and the men in the not-screened auditorium, there are just as many men who died from prostate cancer in each auditorium - which leads us to think that in the aggregate it didn't do any good.

KNOX: Arkes spells it out in the journal Psychological Science. In each auditorium, there would be eight men who'd died of prostate cancer. But among the thousand who got PSA tests, there would also be 20 men who got treated for prostate cancers that would never have grown and caused symptoms. And five of these needlessly treated men would have lifelong complications, such as impotence and incontinence.

DR. IAN THOMPSON: That article is exactly correct.

KNOX: That's Dr. Ian Thompson, a urologist at UT, San Antonio medical center. He says Arkes's numbers are right, according to the current clinical trials of PSA screening.

THOMPSON: If you leave them the way they are, that article is smack-on correct. But the trials had problems.

KNOX: Thompson says the chief issue is that men in the biggest and best study, from Europe, have been followed for a maximum of 13 years and that's not long enough.

THOMPSON: When you analyze those trials very early, what you pick up on are the harms of testing. And it really takes many, many years to see the benefits.

KNOX: He doesn't know how much longer the European men would need to be watched before it's clear that PSA testing did save many more from a prostate cancer death. But he doesn't think the Preventive Services Task Force should have taken a stand against testing at this time.

Dr. Michael Barry thinks Thompson has a point.

DR. MICHAEL BARRY: I'm reluctant myself to make a decision for someone else about PSA screening. And as a result, I'm also reluctant for expert panels to take that position of telling men what to do here.

KNOX: Barry is a Boston expert on medical decision-making.

BARRY: For me, the way out of this controversy is one man at a time.

KNOX: That is, doctors need to lay out the evidence as clearly as they can, which currently says there's very little, if any, benefit to PSA testing.

BARRY: I think many men won't want the test in that circumstance. But some still will and I'm comfortable with that.

KNOX: Up to now, Barry says, those discussions haven't been happening nearly enough.

Richard Knox, NPR News, Boston. Transcript provided by NPR, Copyright NPR.

Related program: