Posts Tagged false positive
DON’T GET AN ANNUAL EXAM. The data are clear — see the recent article in the New England Journal of Medicine and the op-ed in the New York Times — perhaps you missed this counter-intuitive health advice?
Mechanical devices need preventative maintenance. The aircraft mechanic in the illustration prevents engine failure by checking and replacing parts before they go bad. He knows the MTBF (mean time between failures) for the various engine components. You would think this is how the human body works but THAT’S NOT TRUE. You don’t take out an appendix like a spark plug just because they sometimes go bad — you fix it only when needed because surgery hurts and has complications.
One third of the US adult population get annual physical exams and primary care doctors spend 10% of office visits doing those exams. Sound research shows the annual physical is not needed and worse yet, may be harmful because of false positives (tests that say something is wrong but later are proven wrong). It’s the very essence of a false positive — an abnormal test in a healthy person! You know where that leads: “we need to do some additional tests or a biopsy” — just hope it’s not a brain biopsy.
The US healthcare system needs the wasted 10% of primary care time elsewhere. It’s totally crazy — doctors doing unnecessary annual exams that clog up the appointment calendar and make it hard for people with actual problems to get an appointment. And, a large number of people have health problems who don’t see health care providers when they should (but that’s another story)!
Doctors like to do annual physicals — it’s nice to visit with patients and not have to make any hard decisions. And, they make a lot of money doing the exams under the guise of “maintaining a relationship”. But, the exams are not needed.
A proactive patient would make health care appointments as needed for the following:
- Annual flu shot
- Tetanus vaccination every 10 years.
- Cholesterol test every 5 years
- For women over 40 a pap smear every 3 years and a mammogram every 2 years.
Do you really need to have a health care provider tell you the following things, or is this list enough?
- DO keep weight in normal range (BMI below 25)
- DO walk 30 minutes every day
- DO wear seat belts
- Don’t use drugs or alcohol
- Don’t smoke
- DO Check blood pressure every year (automated checks are just fine)
- DO see a health care provider if you have a health problem.
Keep in mind this discussion is about an exam for nothing in particular — just a “check-up” — which you don’t need. On the other hand, a patient needs visits with a health care provider to treat and monitor abnormal conditions. You need routine visits to adjust blood pressure medications, to treat diabetes, to treat acne and to evaluate arthritis.
An x-ray, biopsy or other medical test can have a number of possible results. The uninformed patient or the unwary doctor can be fooled by some of the possibilities. How could a little test hurt you — it’s just a harmless x-ray or small biopsy or just a few drops of blood? There is some underlying truth about a condition we want to know. Do we have a problem or not, yes or no, cancer or no cancer, pregnant or not pregnant, AIDS or no AIDS and many other questions.
Most people understand if the weatherman predicts rain tomorrow, but no rain comes, we laugh because we know prediction of weather is not always accurate. Yet we don’t apply the same common sense to medical tests. If a test predicts a person has cancer we believe it — the sad fact is medical tests are not always accurate. A test can predict a condition when none is present which is called a false positive. A test can predict a condition is absent when it really is present which is called a false negative. A test can predict a condition which turns out to be correct which is a true positive.
So how can a little test hurt? If the test is a false positive which leads to surgery which has a complication then a person could die. This is not just theoretical, it happens. The problem is made worse by a tendency of health care professionals to over-diagnose disease. For example, if a pathologist is not certain if a biopsy shows cancer the safe thing to say is “possibly cancer” rather than “I don’t know”. The surgeon says if it is “possibly cancer” the best thing is to “take it out”. Later, when the patient is missing some body part and no cancer is found the surgeon says “you were lucky”. Really? Perhaps the patient should not have had the test in the first place.
In certain circumstances the chance of getting a false positive is higher than the chance of a true positive. A good example is finding a small spot on a chest x-ray. Most spots on chest x-rays are not cancers. Because of the unreliable nature of the chest x-ray for cancer screening, routine chest x-rays are not advised.
People who study the statistics of medical testing can figure out which tests have the best chances of true results. This is important information for patients and doctors. Recently, the PSA test has come under criticism. It seems the test has a high rate of false positive results and to make it worse, positive results may lead to unnecessary surgery. In view of the statistics, experts now do not recommend PSA tests for routine screening for prostate cancer. Hopefully, this will lead to better tests while the old PSA test heads to the history books.
Another important question is whether the result of a test will change treatment. If not, then don’t do the test. For example, surgery in the very elderly is dangerous. So if no surgery would be recommended for a 100 year old person then don’t do a CT scan of the brain. Is that mean and uncaring? No, it is taking care not to do tests that lead to harmful procedures.
So, more information is not always better. It is smart to avoid tests with a high false positive rate. National guidelines do exist for many tests so search for them on the Internet and as they say “ask your doctor”.