|
|
Endless Screenings Don’t Bring Everlasting Health
By Lisa M. Schwartz, The New
York Times
April 16, 2012
Picture
Credit: Carl Wiens
This month, nine major
medical specialty groups published a list of 45 tests
and procedures that often have no clear benefit for
patients and can cause harm — CT scans for simple
headaches, for example, and X-rays for routine lower
back pain. You don’t often hear calls from doctors for
fewer tests and procedures.
And that’s too bad. Many of them have been oversold,
their benefits exaggerated and their harms ignored.
Consider cancer screening. For decades, it has been
nearly impossible to watch television, read popular
magazines or ride public transportation without seeing
advertisements urging regular mammograms,
colonoscopies or P.S.A. blood tests. These messages
have had a profound effect: the public is now
extremely enthusiastic about the notion that we should
routinely screen people without symptoms for cancer.
In one national survey, most Americans said that
cancer screening is almost always a good idea and that
finding cancer early saves lives most of the time.
Certainly, the rationale behind screening seems
obvious. The earlier cancers are diagnosed, the more
often lives will be saved, right? With enough
screening, we might even stop cancer.
If only. Finding cancer early isn’t enough. To reduce
cancer deaths, treatment must work, yet it doesn’t
always. Second, it must work better when started
earlier. But for some cancers, later treatment works
as well. (That’s why there is no big push for
testicular cancer screening — it is usually curable at
any stage.)
And some of the worst cancers aren’t detected by
screening. They appear suddenly, between regular
screenings, and are difficult to treat because they
are so aggressive.
So how can we be confident that getting a screening
test regularly is a good idea? The only way to be sure
is to look at the results of randomized trials
comparing cancer deaths in screened and unscreened
people. Even when screening “works” in such trials,
the size of the benefit observed is surprisingly low:
Generally, regular screening reduces fatalities from
various cancers between 15 percent and 25 percent.
What does that mean? Think about a “20 percent off”
sale at a store. Whether you save a lot or a little
depends on the item’s regular price. You’ll get huge
savings on a diamond ring, pennies on a pack of gum.
The benefit of screening is like a sale, only you
don’t save money — you “save” on your chance of dying.
Whether you save a lot or a little depends on the
“regular price”: your chance of dying without
screening.
For most of us, the chance of dying of cancer in a
given 10-year period is small: less than 1 percent. So
regular screening with a proven test may bring a 20
percent reduction in a 1 percent risk over a decade.
Put another way, two deaths would be prevented for
every 1,000 people screened during that period.
And what of the other 998 whose fate was not changed
by screening? Some of them will have been harmed.
The most familiar harm is a false alarm: The screening
test is abnormal, but in the end there is no cancer.
False alarms matter because the follow-up tests needed
to rule out cancer can be painful, dangerous and
scary.
But overdiagnosis — the detection of cancers never
destined to cause problems — is arguably the most
important harm of screening. Some cancers grow so
slowly that they would never cause symptoms or death.
When screening finds these cancers, it turns people
into patients unnecessarily.
Since there is no reliable way to know whether a
screening-detected cancer represents overdiagnosis,
most people seek treatment. People on the receiving
end of overdiagnosis can only be harmed — sometimes
seriously — by unnecessary surgery, radiation and
chemotherapy. While it’s hard to precisely estimate
the amount of overdiagnosis that occurs, most experts
agree that it’s an inevitable consequence of
screening.
The bottom line is that while screening may help some
people avoid a cancer death, it will harm many others.
We struggle personally, and as doctors, with these
trade-offs. We all want to avoid dying of cancer, but
no one wants to become a cancer patient unnecessarily.
Increasingly, patient and professional organizations
are taking these trade-offs into account. The National
Breast Cancer Coalition, for example, has said “there
is insufficient evidence to recommend for or against
universal mammography in any age group of women.”
Similarly, the United States Preventive Services Task
Force judged that harms outweighed benefits in P.S.A.
screening for prostate cancer, and recommended against
its routine use.
In the absence of pronouncements like these, it’s
often difficult for patients to decide whether they
want to be tested. For years, our colleagues at
Dartmouth and elsewhere have been gathering evidence
on many tests, but the facts need to be more
accessible. Here’s how we think four of the most
common, and most rigorously evaluated, stack up for
patients.
As noted by the task force, the harms of P.S.A.
screening probably outweigh the benefits, which
actually are uncertain. None were observed in an
American trial, and just a small benefit was found in
a European trial. The test results in a disturbing
amount of overdiagnosis.
For women at average risk, mammography is a close
call. Trials suggest that there are about one to three
fewer deaths for every 1,000 women screened over 10
years (the benefit increases with age). But there is
substantial overdiagnosis — as many as 10 unnecessary
diagnoses per death avoided. Rather than persuading
women to be screened, we should help them be informed.
For colon cancer screening, the benefit of the fecal
occult blood test outweighs the harms. (Surprisingly,
colonoscopy hasn’t been evaluated yet in a randomized
trial.) There are about three fewer colon cancer
deaths per 1,000 people screened over 13 years,
without a lot of false alarms requiring invasive
procedures. There has been no evidence of cancer
overdiagnosis with this test.
For people at high risk of lung cancer because of
heavy smoking, the benefits of CT screening outweigh
harms if it is done at highly skilled centers. In a
recent trial, repeat scans for false alarms were
common, but biopsies were infrequent. The reduction in
cancer-specific deaths in the screened group exceeded
that of other cancer screening tests — indeed, there
were fewer deaths period, a first for any cancer
screening. Data on overdiagnosis, however, have not
yet been published.
That is our take. But the most important thing is that
you have a take of your own. Talk with your doctor
about your risks; ask about the potential benefits and
harms in regular testing. Sometimes you’ll want to say
yes to screening. But it is increasingly clear that
often it is O.K. to say no.
More Information on US Health
Issues
|
|
