The Rest of the Story

When I encountered this article in the British Medical Journal, I was hoping to find some reliable report of the accuracy of the test they mention.  All they do is repeat the manufacturer's claim of 99.9% accuracy.  Then I tried to figure out why this product exists.

BMJ  2005;331:69 (9 July), doi:10.1136/bmj.331.7508.69-c

Home test shows sex of fetus at five weeks of pregnancy
Janice Hopkins Tanne

A finger prick test for pregnant women that can tell them the sex of their child has aroused huge public interest since it was featured on a US television show. The test, which can allegedly show the sex of the fetus at only five weeks of pregnancy, is claimed to be 99.9% accurate.

"We've had more than a thousand inquiries in just three weeks. Our phone is ringing off the hook," said Sherry Bonelli, president of the Pregnancy Store, the company that sells the test kit, the Baby Gender Mentor test (www.pregnancystore.com).

She said the company would promote the test in pregnancy and parenting magazines. "We're always on a hunt for new products," she added.

She said that Acu-Gen, a biotech company in Massachusetts, performs the new test. Woman can order the $25 (£14; {euro}21) kit from the company's website (pictured below), take a finger prick blood sample, and send the dried sample to Acu-Gen for analysis, which costs $250. The laboratory analyses fetal DNA in the woman's blood sample, looking for a Y chromosome, indicating that the fetus is male. Two to three days later women use their assigned code to learn the sex of the fetus over the internet. Users are offered a refund if the result is later found to be wrong. The cost is not covered by health insurance but is similar to the cost of ultrasonography ordered by a doctor or commercial ultrasonography used by parents who want pictures of their fetus ( BMJ 2004;328: 853[Free Full Text]).

Ms Bonelli said that the test gave results at an earlier stage of pregnancy than chorionic villus sampling, which gives results at 11 weeks, or amniocentesis, which gives results at 18 weeks or later and carries some small risks. The Baby Gender Mentor test had no risks, she said.

She said that expectant parents were eager to learn the sex of their fetus. It "personalises the child," she said, meaning that the parents can name the child and prepare nursery rooms.

Concern about using sex selection and abortion to balance the sexes in a family with two or three children of one sex "has come up," she said, but she felt that parents were glad to accept any child and she thought this would not be a problem in the United States.

She emphasised that the test was not intended for diagnostic purposes, such as to detect sex linked disorders, and was not marketed to doctors.

So the test kit costs $25, and the test itself is another $250.  You find out quickly if the fetus is a boy or a girl.  The fundamental methodology is sound, since there is only one possible reason for a Y chromosome to be circulating in a pregnant woman's blood.  I suppose that if you get a big enough sample, and examine it carefully, the test should be accurate.  However, since it is not being marketed as a medical test, it has not undergone the rigorous testing that the FDA would require.  

The company does provide the abstracts of some studies to back up their claims, but I note that none of the studies actually validates the test as they provide it.   That is, if you do the test on a limited basis, with every step of the process conducted by trained personnel, monitored and governed by a research protocol, you get good results.  We do not see any evidence that an effort has been made to see if the results in the lab correlate to the results one gets with the commercially-available test, as performed by ordinary folks.  

So there is one question that remains unanswered: how accurate is the test when performed in an uncontrolled setting?  That, however, is not what I am trying to figure out.  Since there is no way to figure it out with the information provided, there is no point in trying.  What I am trying to figure out, is why this thing should be so popular.  The company estimates that 50-70% of parents-to-be want to know if they’re having a baby boy or girl.  

Come on, people!  Life is full of uncertainties.  That's what makes life interesting.

I don't want to pull a Cruise, and pass judgment on how others choose to handle their pregnancies; if someone else wants to know the baby's gender ahead of time, that is fine with me.  I just can't figure out why they would want to.