For people with no symptoms routine screening can have a downside.
A little over a week ago, a group of cancer gurus, having pondered 40 studies on the potential value of screening men for prostate cancer, ended up much like a hung jury.
The preponderance of the evidence, they concluded, simply did not show, as many had hoped, that the benefits of testing every man over 50 for prostate cancer outweighed the risks.
There are possible, though unproven, benefits to catching cancer early. But these potential benefits don’t outweigh a more clear and present problem – that worrisome results on screening tests set men up for yet more testing, often followed by surgery and radiation that can lead to impotence and incontinence.
If these risks scare you more than cancer – which might not kill you for years – you may want to skip this test, the doctors suggest.
In January, a panel convened by the National Institutes of Health slogged through 100 studies and came to a similarly dreary conclusion about mammograms for women between 40 and 49.
Mammograms clearly save lives for women over 50. But 98.5 percent of women who get annual mammograms in their 40s get no benefit, says biostatistician Donald Berry of Duke University, a member of the panel. The others do benefit some – gaining on average 200 extra days of life, he says.
Like the men who go through invasive testing after prostate screening, many women with suspicious mammograms find out they’re okay only after more tests and often, surgery.
Now, if you happen to be the one whose life is saved by screening, you’re undoubtedly a true believer. And some medical groups, including the American Cancer Society, agree, reasoning that it’s often better to know what’s going on sooner rather than later. This may be especially true for people at high risk of a disease.
But for people at normal risk, faith in testing is coming under attack as evidence for the downside of screening grows.
“The problem with screening is that it can put you on a slippery slope. And if you know enough about that slippery slope in advance, you may not want to walk out on it,” says Dr. Harold Sox, chairman of the department of medicine at Dartmouth Medical School and president-elect of the American College of Physicians.
“Every screening test will have its downside, not only because of the potential risks of the diagnostic workup plus the therapy, but because it will flag many people as abnormal who don’t have cancer,” adds Dr. Barry Kramer, deputy director of cancer prevention and control at the National Cancer Institute.
The dilemma in screening huge numbers of asymptomatic people boils down to a deceptively simple question – is it worth it?
If the screening test is for a disease for which there is a treatment that clearly works without causing too many side effects, it probably is. If there is no treatment that prolongs life significantly, or if the treatment causes harm, it may not be.
Complicating this are other knotty issues – like whether catching something early really leads to more years of life or just more years of knowing you have the disease – the so-called “lead time bias” question.
Often, the decision to be screened “depends on subjective responses that vary enormously from one person to another,” says Dr. Albert Mulley, chief of general medicine at Massachusetts General Hospital.
Many women in their 40s swear by mammograms for the reassurance value – even though in this age group they miss up to 25 percent of invasive cancers. (Among older women, mammograms miss only about 10 percent.)
On the other hand, many men – half, in Mulley’s experience – respond with a firm “No thanks” when told about the tests and treatments that can follow screening with a digital rectal exam or a blood test for PSA (prostate specific antigen).
But it’s not just personal factors like family history or fears of a particular disease that make these decisions so hard.
It’s that, with depressing regularity, different panels of experts look at the same data and come to opposite conclusions.
Sometimes, of course, the data do finally line up one way or the other. Though cholesterol screening has been controversial, many people are tested regularly, believing that if their cholesterol is too high and they lower it, they can reduce their heart attack risk.
The data that clinched the case, the so-called West of Scotland study, came only about a year ago, says Dr. Daniel Levy, director of the Framingham Heart Study. The study showed that, at least among middle- aged men, lowering cholesterol with medication reduces the risk of a first heart attack. In this case, screening clearly pays off.
But many screening decisions aren’t so simple when you look at the evidence on treatments, the quality of the studies, and how much doctors’ groups vary in the criteria they say a test must meet to be worthwhile.
For instance:
Prostate cancer. The American College of Physicians, after examining 40 studies, concluded the decision to screen should be up to each man and his doctor. It’s a tough call because many tumors grow so slowly that men die of something else and because of the harm that can follow a suspicious screening test.
Younger men (between 50 and 69) may benefit more than older men because they have more years of life at stake. But even for them, screening yields on average only two extra weeks of life.
But the American Urological Association does recommend annual PSA and digital exams beginning at 50 for most men, and at 40 for African Americans and men with a family history of prostate cancer. The American Cancer Society recommends an annual digital exam and PSA test beginning at 50 for men with at least a 10-year life expectancy and for younger men at high risk. But the society is considering changes in this recommendation.
And a government-funded panel, the US Preventive Services Task Force, which painstakingly reviewed the data on dozens of screening tests about a year ago, gives a “D” to both these tests, meaning there is fair evidence to recommend against them.
Colorectal cancer. In general, the evidence for screening is good and growing. One test, which examines stool specimens for blood, seems to reduce the risk of death from colorectal cancer by a third, says Dr. David Atkins, senior health policy analyst at the Agency for Health Care Policy and Research.
Another, called flexible sigmoidoscopy, in which a doctor uses a viewing tube to examine the rectum and lower colon, reduces the death risk by 60 percent. Many doctors think it’s better to do both tests, though this is unproved.
And the consensus is in flux. Last week, the cancer society issued new guidelines, saying that at 50 everyone should have either a stool test plus a sigmoidoscopy, with repeat tests every year and five years respectively, or a total colon exam.
The total colon exams means either a colonoscopy – a viewing tube exam of the whole colon – or a barium enema, with these tests repeated every 10 years or 5 to 10 years, respectively.
Breast cancer. For women between 40 and 49, the whole issue of screening is a “mess,” says Dr. Jerome Kassirer, editor in chief of the New England Journal of Medicine.
In January, the NIH panel concluded it could not recommend regular mammograms for younger women. Last week, the American Cancer Society went the other way, strengthening its previous guidelines to recommend annual mammograms for women in their 40s. Yet another panel, the National Cancer Advisory Board to the National Cancer Institute, is still considering the issue.
The value of breast self-exams is debatable, too. The task force gives them a “C.” And a recent study of more than 267,000 Chinese women failed to find the exams reduced the death rate, though the women were followed for only five years.
Cervical cancer. Pap tests – at least every three years – rank among the best screening tools. Every year, Pap tests find serious, precancerous conditions in 300,000 women who need treatment. But the tests also pick up mild abnormalities in 2 to 3 million others, for whom the next step in not clear.
The National Cancer Institute is assessing the options: minor surgery, repeat Pap tests in six months or testing for human papillomavirus, which causes cervical cancer.
For any test, especially those about which experts disagree, the best course “is to engage your doctor in a discussion of the facts, particularly the facts of your own history,” says Dr. Larry Gottlieb, deputy medical director at Harvard Pilgrim Health Care.
This, unfortunately, is getting ever tougher, as managed care chips away at the amount of time doctors have to talk.
But it is “high time,” says Dr. David Rosenthal, president-elect of the American Cancer Society, “to rely less on blanket screening and put decision-making back where it belongs, into the hands of the doctor-patient team.”
Medical screening tests: Which are worth doing?
An analysis of the major screening tests by the US Preventive Services Task Force, a government-funded panel, found little scientific justification for routine use of many procedures. The rankings apply to healthy, asymptomatic people at normal risk of disease. Tests were graded A to E.
A – Good evidence to recommend the test
B – Fair evidence to recommend the test
C – Insufficient evidence either way
D – Fair evidence against screening
E – Good evidence against *