Dangers in Early Detection
By H. Gilbert Welch
Thursday, July 1, 2004; Page A23
You feel well. You're only 60. Your PSA -- the blood test for prostate cancer -- is normal. Much to your chagrin, you learn of recent research that suggests you still might have prostate cancer. But the only way researchers know this is because they performed a more aggressive test -- placing a probe through the rectum of normal men and inserting a biopsy needle six, maybe 12 times to search for cancer in various parts of their prostate. Should this procedure be performed on you? Should it be performed on all healthy men?
This is American medical care today -- care increasingly directed toward the well. Ironically, the primary service we offer them is relentless testing to establish whether they are, in fact, sick. We screen for early forms of diabetes, heart disease, osteoporosis, hepatitis, vascular disease and, of course, cancer. The conventional wisdom is that early detection improves health. But this assumption may be wrong.
Why? Because early disease detection means more people become patients. Inevitably some will be treated needlessly and suffer as a result.
To understand this, you need to understand that each of us harbors early forms of disease. Even in middle age, many of us who feel well have evidence of diabetes, heart disease, osteoporosis, hepatitis, vascular disease and cancer. Just because we harbor these early forms of disease doesn't mean that they will ever affect our health. Some diseases progress so slowly that people die of other causes long before the diseases generate symptoms. Other diseases may not progress at all. Unless we were tested, we'd never have known we were sick.
Prostate cancer is the classic example. Among men age 60, around half have microscopic evidence of prostate cancer if we look hard enough. Yet only four in 1,000 will die from prostate cancer in the next 10 years. How can this be? Because prostate cancer isn't just one disease: It's a spectrum of disorders. Some forms of prostate cancer grow very rapidly and kill men. Some grow slowly and men die of something else before the cancer ever causes symptoms. And others look like cancer under the microscope but never grow at all.
A little over a decade ago, doctors started looking hard for prostate cancer using the PSA and lots of needle biopsies. And we found a great deal: Roughly 2 million cases were diagnosed in this period -- almost a million more than would have been without the test.
Did prostate cancer screening help men? To be honest, we aren't sure about the net effect. There has been a small decline in the death rate from prostate cancer, but this may simply reflect that our treatments are better. While screening probably has helped a few men live longer, it has also clearly hurt others. Millions have been biopsied who otherwise wouldn't have been. Many with nonprogressive disease have been turned into cancer patients unnecessarily. Most have been treated, and many have suffered ill effects. A few have even had their lives shortened by treatment.
This is the reality of early detection. A few may be helped, because their disease is destined to cause problems and because early treatment is able to solve those problems in a way that later treatment cannot. But many simply are told earlier that they have a disease and gain nothing, because their disease could have been treated just as well later, when symptoms appeared. And others are hurt by treatment for a disease that would have otherwise never affected their health.
What's next? Consider CAT scans of the chest to look for lung cancer. During mass screenings in one region of Japan, CAT scans found 10 times as many patients with lung cancer as had been found a few years earlier using chest X-rays. Incredibly, nonsmokers were almost as likely to have lung cancer as smokers. Is smoking getting safer? Of course not. Everyone agrees that smoking is far and away the most important cause of lung cancer. The CAT scans were simply labeling some people as lung cancer patients who otherwise would never be affected by a few abnormal cells.
Why not treat these patients -- just to be safe? Because some people die from treatment. In the Mayo Clinic study comparing lung cancer screening (using chest X-rays) to standard care, more people in the screening group were told that they had lung cancer. It didn't help them live longer; in fact, slightly more people in that group died.
And some think we should scan the whole body. But the harder we look, the more we find. CAT scans of the chest lead more people to be told they have lung cancer, and there are even more abnormalities to find in the abdomen. As one radiologist who has read thousands of these scans put it, "With this level of information, I have yet to see a normal patient."
Millions of healthy Americans are being told that they are sick (or "at risk"). More are undergoing invasive evaluations with needles, flexible scopes and catheters. And more are taking drugs for early forms of diabetes, heart disease, osteoporosis, hepatitis, vascular disease and cancer.
We need to start asking hard questions about whose interests are served by the relentless pursuit of disease in people who are well. Clearly it's good business -- for test manufacturers, hospitals, pharmaceutical companies. And it's good for some doctors.
But is it in society's interest? Many suggest that it saves money by lowering the cost per patient. But the savings per patient (if they exist) are overwhelmed by the increased expense of having so many more to treat. Is it in the interest of sick patients? Absolutely not, as caring for the well increasingly distracts doctors from caring for the truly sick. And what about the well? Is it in their interest? Only they can decide -- after they have been informed that early detection is a double-edged sword.
The writer is a professor of medicine in the Department of Veterans Affairs and Dartmouth Medical School. He is the author of "Should I Be Tested for Cancer? Maybe Not and Here's Why."
© 2004 The Washington Post Company