Saturday, October 1, 2016

Count Down

I am blessed with three healthy children, but one of them was (incorrectly) diagnosed with down syndrome. We declined to abort him (of course) and are so glad he is part of our life. In our case it was both a blood test and an ultrasound that "confirmed" the blood test. We did not go for the more accurate (and more expensive, and invasive, and dangerous to the baby) amniocentesis test, mostly because we would not have changed our decision to keep our baby no matter what the results were. And while I have your attention, please check out Reese's Rainbow.

Today I was reading this article that mentioned that 92% of all babies diagnosed with Down Syndrome are aborted. I know, it's horrible, but I want to talk about something a bit different.

I wondered how many mothers went through all the extra tests, and the accompanying cost and pain and worry and stress of not knowing, and how many said "blood and ultrasound are good enough" and aborted right then. Which got me thinking about false positives.

It's hard to find numbers on false positives (unless maybe you have access to medical documents I don't have). But I did find this article in the Seattle Times which talks about how much more accurate the more extensive, late trimester tests are as compared to the standard first trimester tests (like we had).
cfDNA, tests have a detection rate of 99 percent for Down syndrome, with a false-positive rate of as low as 0.1 percent...That compares with a detection rate of about 79 percent through standard first-trimester screening, and a false-positive rate of 5.4 percent, the study found.
According to the CDC, one in 700 babies is born with Down Syndrome. Taking into account that 92% are aborted that means the rate of Down Syndrome is about 12/700 or 1.7%. That means if you are pregnant and the test comes back positive for Down Syndrome, the chance of the baby actually having Down Syndrome is 20%.**

Here's the math. There are 2 cases where the test will result in a positive:

1. The baby has DS (1.7%) and the test detected it (79%) = 1.7 x 79% = 1.35%

2. The baby does not have DS (98.7%) and the test had a false positive (5.4%) = 5.31%

So a positive will occur 1.35 + 5.31 = 6.66% of the time (I know, right?) but the baby will only actually have DS 1.35% of the time. 1.35 / 6.66 = 20% That means 80% of the time the baby is perfectly healthy and it's the test that's wrong.

This is reassuring if you are a mathematician, but if you are an expectant mother... I wonder how many mothers are frightened or bullied into aborting their "normal" children because of these tests? I would hope that every doctor would counsel a mother to wait for further testing, and even then help her to accept that all babies are imperfect and hers is just as worthy of love as any other baby, regardless of her baby's health or abilities. Sadly, I doubt many doctors do. Ours did not.

But let's say a mother rides out the fear and has the cfDNA test done. Things must be pretty definite then, right? Let's make sure:

1. The baby has DS (1.7%) and the test detects it (99%) = 1.7 x 99% = 1.68%

2. The baby does not have DS (98.7%) and the test has a false positive (0.1%) = 98.7 x 0.1% = 0.1%

A positive will occur 1.69% but the baby will have DS 1.68% of the time. The chance of the baby actually having DS is 99%. Pretty accurate.

Prenatal testing is one of those morally questionable technologies. On the one hand, it's good to allow parents to be prepared for health issues they may have to deal with down the line, and in some cases prenatal testing detects problems which can be treated. However, it tempts parents to treat the problem by eliminating the child rather than the disease.

[** Note to mathematicians out there. I realize that if you take into account the false positives then that 12/700 number is more like 2.4/700, which also affects the final probability, etc. Since I don't have any information on how many mothers abort after the first test and how many have a more accurate test down before aborting, I stuck with the most conservative numbers I could.]


Post a Comment