Tag Archives: down syndrome epidemiology

Is Down Syndrome Disappearing? Well, Not Exactly But….

Iceland has given the world the Eddas, Sigur Rós, Björk, and some magnificent geology.  A more ambiguous achievement, though, is suggested by a recent CBS News story that claimed that Down syndrome is disappearing from Iceland as a result of prenatal testing. The claim has been bouncing around the Internet for a few years. The earliest reference I could find was a November, 2015 letter sent to the Office of The United Nations High Commissioner for Human Rights authored by Downpride, an international advocacy group for people with Down syndrome. The letter is an “[a]ppeal to the United Nations to stop discriminatory use of prenatal genetic screening aimed at eradication of people with Down syndrome and other groups.” It was, in my view, an understandable and justifiable reaction to largely non-critical widespread adoption of Noninvasive Prenatal Testing (NIPT) from a community that has good reason to be concerned. Needless to say, it generated a lot of heated reaction. Just Google “Iceland Down syndrome” and you will see what I mean.

Delving into the story was like getting lost in a hall of mirrors; many sites simply referenced each other. But the claim that Down syndrome is disappearing from Iceland and that 100% of pregnancies with Down syndrome in Iceland are terminated turns out to be not quite so straight-forward. While Iceland represents a microcosm of the larger concerns of people with disabilities, their families, and their supporters, it is not necessarily an accurate reflection of the macrocosm of the larger population dynamics of Down syndrome in other countries, particularly the United States.

The ultimate source of the data, according to the Downpride letter, was testimony presented to The Althing, the Icelandic parliament that is the world’s longest existing legislative body. I tried unsuccessfully to find that testimony. I then searched PubMed but found only limited help. So I decided to do my own back-of-the-napkin calculations. I obtained the birth distribution by maternal age in Iceland for 2016, and grouped the ages by quinquennia. The expected frequency of Down syndrome was based on data from 1976, prior to the advent of widespread prenatal diagnosis.

Age Group # of Births Exp. Frequency of Down S. Exp. # of births with Down S.
15-19 72 1/1667 0
20-24 592 1/1587 0.37
25-29 1305 1/1087 1.2
30-34 1218 1/763 1.6
35-39 672 1/248 2.7
40-44 165 1/79 2
45+ 10 1/24 0.4
Total 4034  1/488 8.27


Thus, in Iceland in 2016, there were 4034 births. In the absence of prenatal diagnosis and selective termination, 8 or 9 babies with Down syndrome would be born, for a frequency of ~1/450-500 births. I then made the following assumptions, acknowledging that each has some potential error:

  • Based on a 2016 publication, about 80% of pregnant Icelandic women will choose to undergo prenatal screening
  • According to Dr. Hulda Hjartardóttir, chief of obstetrics at Iceland’s National University Hospital, among Icelandic woman who have a positive screen, about 25% decline diagnostic testing and continue the pregnancy. Thus, roughly 1/3 of Icelandic pregnant women either do not undergo screening to begin with or decide to continue the pregnancy and not proceed to diagnostic testing if a screening test is positive. The impact of these percentages on Down syndrome frequency depends on the age distribution of those who declined screening or diagnostic testing, but for argument’s sake, I assumed an equal distribution across maternal ages.
  • 100% of women whose pregnancies are diagnosed with Down syndrome will choose to terminate. I could not verify this claim, but I decided to go with the most extreme scenario. This has not been the experience in many countries, where termination rates have been high but not typically 100%.
  • The CBS News story mentions the Combined Screen, so I assumed this was the standard screening test in Iceland when the claims were made in The Althing. I therefore set the detection rate for Down syndrome to 90%, that is, of all women undergoing screening, about 10% of pregnancies with Down syndrome will be screen normal and would not proceed to termination (some studies suggest that the Combined Screen may have a sensitivity somewhat less than 90% but because about 21% of pregnancies in Iceland occur in women 35 and older, a higher sensitivity – and false positive – rate is expected).

Based on these assumptions and the above table, of the potential 8-9 babies born with Down syndrome, about 2-3 would actually be born because their mothers did not undergo either prenatal screening or diagnostic testing, and another baby with Down syndrome would be born because the Combined Screen would be expected to miss about one case. In other words, the total number of newborns with Down syndrome in Iceland would be expected to drop from 8-9 every year to about 3, maybe 4, per year. These numbers could increase or decrease with many factors, such as changes in fertility rates, maternal age distribution, the sensitivity of screening tests, social trends that influence the choice of abortion, and random fluctuations that occur with any demographic trend especially with the small number of births in Iceland (about that many babies were born last year in the hospital where I work in Seattle). If readers know of empirical data from Iceland to support or refute my estimates, please share it.

Of course, for advocates, every loss of a pregnancy with Down syndrome is serious, no matter how small the number. But these estimates put the concerns in some perspective. Among other things, it is fair to say that most, but not 100%, of pregnancies with Down syndrome are terminated in Iceland, and the birth prevalence of Down syndrome in Iceland is falling considerably but not likely, in my view, to disappear entirely.

I think a more realistic picture of the impact of prenatal screening on Down syndrome, in the US at least, is provided by Brian Skotko and his colleagues Frank Buckley, Jennifer Dever, and Gert de Graaf in a recent publication in the American Journal of Medical Genetics. Over the last few years, they have consistently provided some of the most reliable estimates of the demographics of Down syndrome and the effects of prenatal screening.

According to the de Graaf et al. paper, a detailed look at changes over time in the demographics of Down syndrome in 9 states, the number of people living with Down syndrome has steadily increased since 1950. The two major factors driving that growth have been longer survival due to better medical care along with the unrelenting trend of the last 35-40 years of delayed childbearing. This growth, however, has been partially offset by a loss of births with Down syndrome due to prenatal screening. The loss varies with geographic region, but overall, the prevalence of Down syndrome is roughly 70% of what it would be if prenatal screening were not available. Interestingly, the most growth in the Down syndrome population occurred among Hispanics and Native Americans. So, unlike the near elimination of Tay-Sachs disease in many Ashkenazi Jewish communities, the prevalence of Down syndrome is dropping, but not close to disappearing, at least in the US.

Other factors may affect the Down syndrome birth frequency, such as changes in maternal age distribution, availability of abortion, and access to health insurance. For example, in the highly unlikely event that every woman 35 and older refrained from pregnancy, the birth frequency of Down syndrome in the US and many Western European countries would be reduced by more than 50%. On the other hand, if abortion were to become illegal (not highly unlikely), then presumably the birth frequency of Down syndrome would increase. Limiting access to good medical care (unfortunately also not highly unlikely in the US) could lower the overall prevalence of Down syndrome because of reduced survival.

Current trends suggest that, for the immediate future, prenatal screening will continue to reduce the birth prevalence of Down syndrome. It is becoming increasingly easier for women to undergo prenatal screening and more difficult to just say no. This is due to aggressive marketing by commercial labs of “newer, better, bigger, cheaper” screening tests like NIPT; the dearth of time and resources devoted to unbiased education about Down syndrome and the pros and cons of screening tests; inequitable social distribution of medical resources and social support; and the rarity of long, difficult discussions between pregnant women/couples and their providers about whether they should even enter the prenatal screening cascade to begin with. It also does not help matters that the current US President lacks any moral decency and takes pleasure in mocking people with disabilities.

Although I am a strong supporter of women’s reproductive rights and well-informed, gut-wrenching decisions to terminate a pregnancy, it is becoming increasingly difficult to provide ethical justification for further expansion of prenatal screening, or expanded carrier screening for that matter. This is something that society needs to address but particularly genetic counselors because we are in the thick of it.

As I have previously argued, almost no research has been conducted that has tried to demonstrate whether prenatal screening can improve the medical, social, and emotional lives of people with disabilities and their families. Some women undergo prenatal screening because they think it will prepare them for raising a child with Down syndrome, but we really can’t tell them if screening does help or if it is worth their emotional and psychological investment. Carrying out such research is critical. If we can demonstrate broader benefit of prenatal screening, then we can open up a dialogue with the disability community rather than continue the shouting matches, and offer greater and more equitable justification for NIPT and other screening technologies.

Or we can continue shouting at or dismissing one another.



Filed under Robert Resta