I'm busy working on my blog posts. Watch this space!
How BRCTs can help doctors and patients discuss screening tests: Part One of a series.
November 7, 2016
Often I am asked by patients if I plan to do an annual EKG to make sure their heart is ok. To many that is part of a thorough physical, one that involves enough of an exam and enough tests and x-rays to assure that the patient is in good health. Done under these circumstances, and EKG is called a screening test, a test done on someone without symptoms or suspicion of disease in the hope of finding silent heart disease that would have been otherwise undetected. Our medical culture has embraced many screening tests that are performed routinely and usually reimbursed by insurance. These include mammograms, colonoscopies, PSA blood tests for prostate cancer, cholesterol lab tests, heart stress tests, bone density tests, listening to the neck for carotid disease, and many more. There are even commercial outfits that promise, for $100 or so, to make sure your heart is ok, your carotid arteries are clean, your bones are strong, and that you have good circulation all by screening you with a test called an ultrasound.
If screening tests do sometimes reveal a disease that is silently festering in someone’s body, they also have a darker side, one less evident to the people who are being screened. This can be both confusing and counter-intuitive to patients and even difficult to explain. BRCTs are very effective in demonstrating both the merits and pitfalls of certain screening tests. I use them in my medical practice to good effect, and we have provided information about many screening tests in our book and this blog. Due to their familiar setting and their numberless depiction of actual risk and benefit, BRCTs are easy to understand and enable patients and doctors to have a sensible conversation about whether a certain screening test makes sense for each individual patient. An excellent discussion of how BRCTs help with screening can be found in Jay Hancock’s Kaiser Health News article, which also contains a video that crisply explains why screening tests can be problematic and how BRCTs enable people to understand their individual risks and benefits.
What could be wrong with screening people for disease? What is the potential dark side? There are three perils of performing tests on people who have a low likelihood of having the disease being looked for.
First, due to what is labeled as a false negative, often the test is not sensitive enough to find the disease, and thus a patient is told they are normal when in fact they are not. The test gives them false reassurance. Some tests only detect disease when such disease is fairly advanced and no longer amenable to treatment, which is typical of prostate blood tests, blood tests for ovarian cancer, and even mammograms.
Second, due to what is labeled as false positive, often the test says there is a disease present when in fact there is not. Even tests that are accurate 99% of the time are prone to being falsely positive, especially when the chance of having the disease is low (low pre-test probability of disease). The less likely a person is to have a disease, the more there is a chance that any abnormal results will be false positive results. In fact, with most screening tests, an abnormal result is much more likely to be a false positive than a real positive. Why is this important? First, it frightens patients unnecessarily and causes stress. But second, it can lead to harm. A false positive test will trigger further tests, and even unnecessary treatments, that can hurt an otherwise healthy person unnecessarily. For instance, when we find a lung nodule on a chest x-ray of a healthy patient, it is more likely that nodule is a harmless blip than a cancer, but to find that out we have to do a biopsy, which has a chance of popping the lung or even killing the patient. Thus, we can harm healthy patients by doing a test due to the high prevalence of false positives.
Third, many diseases exist in the body without causing problems, and fixing those diseases is more harmful than leaving them alone. Many cancers, such as prostate, are harmless in the clear majority of cases, while surgery for prostate cancer can cause significant harm. Thus it may be more dangerous to find and fix a prostate cancer than to not find it at all. In fact, 20% of cancers regress spontaneously or do not grow; the body takes care of the cancer without our intervention. When we detect these cancers and treat them, we have not helped someone, but rather exposed them to harm from unnecessary medicines or surgery. Blocked arteries in the heart or carotids may already have been bypassed by our bodies, and yet when we find them and “fix” them with surgery or stents we can cause healthy patients to have heart attacks, strokes, kidney disease or even to die from our intervention. Many tests can detect problems in our bodies, but what the tests do not tell us is whether these problems need to be fixed, or even if the fix is more dangerous than the disease. The more we dig into the asymptomatic body with screening tests and exams, the more we may find problems that are best left alone.
EKGs are a good example of how screening tests can be detrimental. As noted, many patients expect an EKG at their annual exam to make sure their heart is doing well. Some want an EKG before they start an exercise program to make sure they are safe to exercise. Surgeons often want EKGs before surgery to assure there will be no risk of heart attacks during the operation. But EKGs do not help answer those questions. The BRCT on the left has no blackened seats, because out of 1000 people who get an EKG, none will avert a heart attack beyond what the doctor can learn from a good history and evaluation of risk factors for coronary disease. That is why the US Preventive Services Task Force states that EKGs are ineffective in picking up silent heart disease. In fact, as the BRCT shows at the top of this article, of 1000 people who have a heart attack, 800 of them would have had a normal EKG immediately prior. Thus a normal EKG does not exclude serious heart disease. Can an EKG cause harm? Yes, due to its high false positive rate (800 out of 1000 abnormal EKGs occur in people without serious heart disease), abnormal EKGs can lead to unnecessary testing and treatment in healthy people, and some of that testing/treatment (such as catheterizations, stents, medicines) is potentially dangerous.
In fact, out of 1000 people who have EKG’s, 20 will suffer serious harm due to false positive tests, including heart attacks, kidney damage, stroke, and even death, as shown in the BRCT to the left.
In the next few blogs we are going to use BRCTs to assess several of the most common screening tests ordered by doctors and embraced by patients. Some clearly have benefit, and virtually all have potential harm due to high rates of false positives. By portraying those benefits and risks in a 1000 seat theater, patients can have a better sense of whether the test is appropriate for them, and if they are comfortable with the level of uncertainty that such tests universally engender. Ultimately a discussion between doctor and patient is the best way to assure that screening tests are ordered appropriately and that their results are interpreted with the caution that is demanded by the inaccurate nature of the tests. For that to happen, it is vital that doctors and patients start with accurate information that both can understand. BRCTs are the ideal tool for such a discussion.