A few days ago, Merrill Goozner at Gooznews posted a great commentary on a recent New York Times article that reveled in the so-called “revolution in medical prognostication.”

This time, the amazing innovation is a DNA test that helps to predict mens’ risk of getting prostate cancer at the low, low price of $300. But as the article points out, this test “cannot predict which men will get aggressive cancers” and thus “could lead to more screening and unnecessary surgery and complications.” In other words, it gauges the risk—not the inevitability, and not the severity—of prostate cancer. Defenders of the test say that if all goes according to plan, men “may want to get the new genetic test when they are young” so that they can get a jump on prostate cancer treatment.

But here’s the problem: prostate cancer treatment is, more often than not, less useful than you might think. Merrill points out that it’s “already overdiagnosed and overtreated with horrendous side effects for thousands of late middle-aged men.” Here at Health Beat, Maggie has also noted that while the risk of prostate cancer is often hyped, almost every authoritative medical body on the matter agrees that the benefits of screening and treatment of early-stage prostate cancer are highly uncertain. There is no evidence that early treatment prolongs life by a single day.  We excel at finding out if there might be the need to do something, but we know very little about the impact of what it is we actually do.

What we do know, however, is that prostate cancer treatment has some
unpleasant side-effects, such as incontinence and impotence. Merrill
gives us a poignant image of how these side-effects can burden the men
who undergo prostate cancer treatment. His own father “was diagnosed
with prostate cancer around age 70, when he was already several years
into his Alzheimer’s decline. His urologist operated, “leaving my mom
with the added duties of cleaning up after an incontinent patient in
addition to looking after her increasingly vacant husband.”

Sadly, this whole obsession with being able to forecast risk—while
being powerless to actually address the potential danger effectively—is
also at play with regards to Alzheimer’s. As Health Beat has noted
in the past, recently a procedure was created to gauge individuals’
risk of contracting Alzheimer’s (although not with 100 percent
accuracy), even though we still have no way of effectively combating
the disease.

To its credit, the NYT piece notes that panic over prostate cancer
treatment is counter-productive because “most prostate cancers grow so
slowly that they would have been harmless if left alone.” The voice of
reason in the article is Dr. Peter C. Albertsen, a surgery professor
and prostate cancer specialist at the University of Connecticut, who
views prostate prediction as a nightmare that feeds off the “cancer
phobia.”

Underlying this whole issue is a substantive medical question: how much
faith should we be putting in biomarkers, i.e. genes, biochemical
molecules that indicate the presence of a disease, but aren’t the
disease itself? As Merrill says, a lot of folks want to make sure that
we are committed to biomarkers, since they are the foundation of “a $40
billion a year industry.” The credibility of biomarkers is key to
‘innovations’ like the prostate cancer and Alzheimer’s risk
assessment—and to more traditional medical problems like heart disease.

In the case of heart disease, Merrill rightly calls cholesterol the
“king of all biomarkers” because it’s widely accepted that more
cholesterol suggests a higher risk of heart disease. But while many
doctors think that reducing cholesterol is the end-all-be-all of heart
health, a recent trial of the drug Vyotrin has called even this gospel
into question. 

In the trial Vyotrin, which lowers bad cholesterol, had no effect
on plaque growth in arteries, a condition closely correlated with heart
attacks and strokes. Before that, Pfizer found that one of their drugs
that supposedly lowered bad cholesterol actually caused heart attacks
and strokes. The bottom line is we aren’t 100 percent sure that you can
just lower cholesterol and guarantee better cardiac health—although
this is something that we’ve treated as a certainty for years.

Yet rather than try and look deeper into the relationship between biomarkers and heart disease, a 2004 JAMA article
(membership required) makes it clear that the modus operandi for
medical professionals is to find more biomarkers. We don’t want to
understand how things are related, just find out as many relationships
as we can. The problem is that, as the JAMA piece suggests, this
approach readily confuses “predictive value” with “background noise.”

Yet Merrill is right when he points out that a more nuanced
understanding of biomarkers is important to becoming smart about
predicting health hazards. He brings up an alternative approach to the
strategy of amassing as many potential biomarkers as possible: thinking
more closely about the nature of biomarkers. In this vein, he notes
that some think that the key relationship between cholesterol and heart
disease is the anti-inflammatory effect of certain drugs that work much
like aspirin. If we drill down a little deeper into the question, there
may be a nontraditional—but ultimately simple—answer. As Merrill puts
it, we should ask if anyone has “ever conducted a comparative trial
testing Lipitor [a cholesterol-lowering drug] to aspirin for people at
risk of heart disease?”

This is the sort of outside-the-box thinking that gets lost in the
madness over messianic medicine that constantly promises absolute
certainty in health care through the miracle of innovation. We’ve seen
it before—for example, when drug companies play down lifestyle changes that might make their product obsolete or create new medicines for diseases that we know simple $1-a-day antibiotics
can cure. There’s little profit in thinking small yet effective; the
money is in thinking big and vague. Common sense is thrown out the
window.

Unfortunately, this sort of directionless overkill can have real repercussions. A more recent NYT column
notes that when the new doubt surrounding cholesterol-lowering hit the
airwaves, the news “set off a panic among patients using cholesterol
drugs. One doctor said he assigned a nurse full time to take the calls
‘and convince patients not to stop their medicine.’ ”

We need to remember that, no matter what we are promised, uncertainty
is ubiquitous in medicine.  Patients don’t always want to hear this.
They trust in medical professionals. They trust in procedures. They
trust in the capacity of innovation to make their lives better. And
they should—but for our health care system to work, those in whom we
trust must be committed to efficacy rather than innovation for its own
sake. The focus needs to be on what should be done, rather than what
can be done. The two are not the same, no matter what medical
prognosticators would have you believe.