What is the harm of overdiagnosis?

When I was an adolescent, I loved the cartoon Rocky and Bullwinkle. Along with the humor, it had a very adult satirical point of view.  One villain, Boris Badenof, relieved his frustration by acting on his motto: "Do something to somebody quick!!"  Unfortunately, doctors often subconsciously agree with Mr. Badenof when dealing with patients they cannot actually make better. Thus, they treat patients with treatment that they should know will not help. Two articles in the November issue of Pediatrics illustrate this all too common situation.

The first study talks about a rarely mentioned subject, how treatment of an accurate diagnoses does not always improve outcome. The detection of the abnormality does not benefit the patient. The article also suggests that overdiagnosis may affect commonly diagnosed pediatric conditions such as attention-deficit/hyperactivity disorder, bacteremia (bacteria in the blood), food allergy, hyperbilirubinemia (jaundice), obstructive sleep apnea, and urinary tract infection.

For example, the age of children in a first grade class can range from 5.5 years to almost 8 years. All of us parents know that the maturity and the ability to sit still changes immensely during this time period. So if one looks at the demographics of the diagnosis of ADHD, the younger child is being compared to his/her classroom peers. Then this child is more likely to receive the diagnosis of ADHD and more likely to be treated with medication. Do they have ADHD or are they simply young for their grade? Being young for your grade should not lead to being medicated.

For adults, cancer screening done on all people, rather than just high risk people, leads to overdiagnosis. Screening all women for breast cancer results in finding more breast cancer and treating more breast cancer. However in a 25 year study of Canadian women, the added diagnoses, with often very invasive treatment, did not change the deaths from cancer at all and markedly increased the costs, pain and side effects in that cohort of women. Identical results have been found in asymptomatic young children screened for neuroblastoma and in asymptomatic men screened for prostate cancer: The mortality and morbidity actually increased from all the needless procedures.

There are many other situations where overdiagnosis can lead to needless treatment. An old joke says that since every time God looked down in the Hebrew Bible things were bad, maybe if he looked down less, things would get better. The easy accessibility of all our new imaging modalities (MRI, CAT scan, lower radiation X-rays) and decreasing amounts of blood needed for lab tests have resulted in practitioners ordering more studies.  If you order more studies, you will find more problems. The National Heart, Lung and Blood Institute of the National Institutes of Health has recommended blood lipid (cholesterol and other fats ) screening, and if cholesterol is even a little high, treatment of all children 9 to 11 year olds with life-long anti-cholesterol drugs.  Unfortunately, they did this without any data on the effects of these drugs on pre-pubertal children or the effects of these drugs if taken for 60 years or more.

In the same issue of Pediatrics, there was updated guidance on how medical professionals should diagnose, manage and prevent bronchiolitis. Bronchiolitis is a lung manifestation in young children of a viral infection by RSV (Respiratory Syncytial Virus). It is a common cause of breathing problems leading to hospitalization in infants in the United States from around September to April.

RSV bronchiolitis is so serious, especially in very small premature babies after they are discharged home, that these very high-risk prematures receive a preventive medication, Synagis (Palivizumab) monthly during the five worse months and it costs up to $2,500 per dose. Currently, I am supervising research which preliminarily seems to show that palivizumab does not help prematures long term in preventing lung disease, but may help while they are young and vulnerable. Therefore, it should only be given when they are premature or if a baby has a chronic condition like lung disease.

The revised guidelines emphasize many "do not's" and very few "do's" in treating bronchiolitis. Why so many "do not's?" Because of the easy availability of diagnostic testing and treatment options, these sick babies get a lot of tests and many treatments, all of which do not change outcome and many of which can actually cause harm. So why do they get the tests and treatments?  The parents are upset and want the medical provider to "do something quick."  Often the medical provider in the acute care facility, especially if inexperienced with very young babies, also feels more comfortable "doing something quick," than just observing and supporting the baby.

Diagnosis is by examining the infant, not by x-rays and cultures and blood tests. Treatment should be by examining the infant and seeing if the child is "dry" and needs oral or intravenous fluids, or seeing if the child is in respiratory distress and needs oxygen, not by just looking at the pulse oximetry meter which is often wrong because the child moves around so much. And, finally, most medication just does not work. Albuterol, steroids and epinephrine (asthma medications) do not work and can make the child sicker with their side effects. Antibiotics do not affect viruses, do not improve bronchiolitis, and can cause yeast infections and very bad diarrhea.  Hypertonic saline inhalation may help a little, but are only recommended if the child is hospitalized.

In January 2014, I had an editorial in the Cochrane Review that said bronchodilators should not be used routinely in the management of bronchiolitis, but even my residents will on occasion get unnecessary tests or give unnecessary medication. Why? People are more comfortable doing than watching sick children.

Every day, a parent asks me when should they give tylenol (acetaminophen) or ibuprofen for fever. I try to tell them that in general fever is good for illness and makes it go away faster and that febrile seizures are very rare unless the parents also had them. I then say, "If the child feels ill, treat the pain or fever by giving the child the acetaminophen or ibuprofen, and if you are uncomfortable with the child's fever, maybe you should take the medicine for your headache.  Treat the child not your anxieties."

But people like to "Do something to somebody quick," even if it does not help.

Have a question for the Healthy Kids panel? Ask it here. Read more from the Healthy Kids blog »