Don’t make yourself sick fretting over medical tests

Does medical screening save lives? Is it worth the accompanying risks?

It’s been a controversial month for news about diagnostic screening procedures. Newspaper headlines have proclaimed that radiation from CT exams can increase your risk of cancer, or even cause cancer. A task force in the United States recommended that doctors stop using the PSA blood test for prostate cancer screening, stating that the procedure “has no benefit or that the harms outweigh the benefits.”

Articles in both the British Medical Journal and the Journal of the American Medical Association raised the issue of overuse and over-diagnosis caused by advanced imaging techniques such as CT scans, leading one commentator to declare that a “plague of over-diagnosis” exists in medicine. In other words, tests are now finding problems or cancers that — left alone — would not pose any risk to the individual or require treatment.

Ontario, desperate to rein in a $15-billion deficit, announced that it would cut fees for diagnostic exams by 50 per cent for doctors who refer patients for testing in their own clinics.

Finally, a book and numerous articles by Alan Cassels, a drug policy

The degree of risk is all dependent on the context — and people need to understand that as they view the headlines.

researcher at the University of Victoria, have overwhelmed Canadians with accounts of the negative and risky elements of various screening tests. He contends that high numbers of false positives, radiation exposure and the pain and suffering of resulting investigations from false positives often outweigh the benefit of the test. In some cases, he’s right. But it doesn’t mean we pass on every screening or testing procedure.

All of the above have no doubt planted seeds of worry in many health care consumers — particularly those concerned about cancer — leaving them rightly confused about who to believe or what study to trust and what procedures they should agree to undertake.

We’ve heard conflicting stories about the value of mammograms for years. Mammograms were the poster child for cancer screening; women were consistently admonished to get a yearly mammogram and do a monthly breast self-examination. The latter is no longer deemed relevant, and the mammogram is now considered so unreliable because of an excessive number of false positives, that most U.S. states now require a mammogram to be accompanied by at least one other test, such as ultrasound. They are no longer recommended on a yearly basis beginning at age 40. The new regime calls for screening every two or three years beginning at age 50.

The story is much the same for PSA testing. PSA levels are unreliable and can be elevated from benign factors, enlarged prostates or infections. They result in far too many false positives that require further procedures that can cause more harm than good.

So what are we to do?

First of all, we must remember that medical screening is different than medical testing. Screening involves searching for early signs of disease in a healthy person, while diagnostic testing is done when there is clinical justification — i.e., the patient has symptoms and there is a legitimate need to understand the source and extent of the problem.

The risks and benefits must be evaluated for each patient — we can’t make broad-based claims of the value of testing for entire populations. For someone who is sick, has symptoms, risk factors or a strong genetic history of a particular disease, the benefits of undergoing testing — or even routine screening — probably outweigh the risks.

Second, it is likely that most of the public receives their information on the latest health care research from the media, where studies are rarely reported within the context of the whole body of research on the subject. Context is particularly problematic when a study assigns a risk to a medical procedure.

For example, a 2007 study (that is often quoted) calculated the risk of cancer death resulting from the radiation of a whole body CT scan as one in 2,000. But phrase it another way, and the same statistic translates into 99.95 per cent chance that you won’t get cancer.

Similarly, it was reported last week that radiation exposure from two or three CT scans in childhood could triple the risk of developing brain cancer, while five to 10 scans triple the risk of leukemia. Researchers estimate that for every 10,000 head CT scans performed on children 10 and younger, there is one more case of leukemia and one more brain tumour case that would typically be expected.

Triple the risk? It sounds horrific — especially when it is the headline for the story. But when you do the math, it means that there is a 99.98 per cent chance that they will be just fine. If a child had a head injury that was bleeding or a known cancer that needed to be monitored, and his or her parents were told there was a 99.98 per cent chance that there would be no side-effects, it’s essentially a certainty that the parents would give the go-ahead.

The degree of risk is all dependent on the context — and people need to understand that as they view the headlines.