The New York Times had an excellent article on newborn screening tests.
- Panel to Advise Testing Babies for 29 Diseases. Kolata G. The New York Times, February 21, 2005.
Unfortunately, this article is no longer available online. But it discusses a recent push to standardize and expand the screening tests for newborns to include 29 different diseases. It seems like such an obvious thing to do: let's screen for these conditions, because the more we know, the better we are able to care for these children.
Proponents say that the diseases are terrible and that an early diagnosis can be lifesaving. When testing is not done, parents often end up in a medical odyssey to find out what is wrong with their child. By the time the answer is in, it may be too late for treatment to do much good.
Opponents, however, point out that false positive results may present more problems.
But opponents say that for all but about five or six of the conditions, it is not known whether the treatments help or how often a baby will test positive but never show signs of serious disease. There is a danger, they say, of children with mild versions of illnesses being treated needlessly and aggressively for more serious forms and suffering dire health consequences.
The article also offers a historical perspective.
The history of newborn screening, they say, is filled with cautionary tales. %22The majority of newborn screening tests have failed,%22 said Dr. Norman Fost, a professor of pediatrics and director of the program in medical ethics at the University of Wisconsin. Over the years, Dr. Fost said, %22thousands of normal kids have been killed or gotten brain damage by screening tests and treatments that turned out to be ineffective and very dangerous.%22
and cites phenylketonuria (PKU) testing as an example.An infant with PKU cannot metabolize phenylalanine, and the build up of this amino acid can lead to serious neurological damage. The treatment, a diet low in phenyalanine, is very effective, but only if the condition is diagnosed early. The PKU testing done today is very good, but tests performed 45 years ago had problems.
Back then, any infant who tested positive would be put on this special diet. When phenylalanine is withdrawn from the diet of a healthy infant, that infant suffers from even more serious neurological problems and can even die. Many infants who falsely tested positive were put on this diet and their harms outweighed the benefits of PKU screening. As researchers learned more, they were able to refine the test to prevent most false positives, but the damage had already been done.
An additional article about Universal Newborn Hearing Screening (UNHS),
- The false-positive in universal newborn hearing screening. Clemens CJ, Davis SA, Bailey AR. Pediatrics 2000: 106(1); E7. [Abstract] [Full text] [PDF]
also discusses the problems with false positives:
However, support for UNHS is not universal. One of the most concerning issues raised is the high rate of false-positive results. The literature reports false-positive rates between 3% and 8%. This has caused a number of critics to decline to recommend UNHS until the false-positive rate can be decreased and/or there is further knowledge of the emotional effect this false-positive labeling has on families. A number of studies from other newborn-screening tests have shown that false-positive results can engender lasting anxiety and adversely affect the parent-child relationship. In addition, deUzcategiu and Yoshinga-Itano surveyed mothers immediately after their children had failed the newborn hearing screen and found that 20% to 50% of mothers reported feelings, such as anger, confusion, depression, frustration, shock, and sadness. However, it is still unknown how persistent or detrimental these feelings are. pediatrics.aappublications.org/cgi/content/full/106/1/e7
This article ultimately concludes that with a reduction in the false positive rate, that the benefits of UNHS outweigh the costs.
I'm trying to develop a good set of web pages on diagnostic testing, but there is a lot of work that I need to do. I also offer a couple of training classes that discuss diagnostic tests:
- Stats #21: What do all these numbers mean? Sensitivity and specificity
- Stats #24: What do all these numbers mean? Likelihood Ratios
You can find an earlier version of this page on my original website.