Richard Dawkins made quite a stir on Twitter recently by answering a question about whether he would recommend abortion a fetus that was likely to be born with Down Syndrome. One of the fascinating (at least to me) issues surrounding Dawkins tweets was a) what percentage of fetuses with Down Syndrome are aborted (Dawkins had suggested it was up to 90 percent), and b) as tests become more sophisticated, are the number of people born with Down Syndrome declining?
Time magazine aptly captured the conventional wisdom in 2012 with an article by Bonnie Rochman titled Why Down Syndrome Is On the Decline. According to Rochman,
The number of babies born with Down syndrome has been shrinking for at least two decades, and new, non-invasive prenatal blood tests that can be administered as early as 10 weeks — long before a woman even looks pregnant — have the potential to diminish their ranks further.
Unfortunately, Rochman doesn’t provide any evidence for this claim. What evidence there is for this is conflicting.
It is import to understand that statistics on Down Syndrome births do not appear to be systematically collected and centrally reported. The National Birth Defects Prevention Network collects statistics on prevalence of birth defects, however, and a number of researchers have used it to estimate prevalence of Down Syndrome.
In 2009, Pediatrics published an analysis of Down Syndrome prevalence rates from 10 US regions in order to derive an estimate of both the overall prevalence and current trend of Down Syndrome in the United States. The researchers concluded that the number of Down Syndrome births appeared to be increasing through 2003,
From 1979 through 2003, the prevalence of DS at birth increased by 31.1%, from 9.0 to 11.8 per 10000 live births in 10 US regions. In 2002, the prevalence among children and adolescents (0–19 years old) was 10.3 per 10000. The prevalence of DS among children in a given age group consistently increased over time but decreased with age within a given birth cohort. The pooled prevalence of DS among children and adolescents was lower among non-Hispanic black individuals and other racial/ethnic groups compared with non-Hispanic white individuals; it was also lower among females than males.
In 2013, The Journal of Pediatrics published a study, Current Estimate of Down Syndrome Population Prevalence in the United States, that attempted to estimate the total number of Americans with Down Syndrome. Among other things, that study estimated,
Because birth defects surveillance systems are fairly new and limited to specific US regions, we adopted the approach of deGraaf et al for estimating annual births of infants with Down syndrome by applying maternal age-specific birth prevalence of Down syndrome to annual numbers of live US births. Data on live births by maternal age group were available for 1940-2008, and data on total US live births were available for 1909-1994, from which we estimated the number of live US births per maternal age group for 1909-1939. We estimated the total number of births of infants with Down syndrome using birth rates per 1000 live births by 5-year maternal age groups reported by Carothers et al. Figure 1, A shows that the number of live births with Down syndrome, which parallels the total number of US births, increased in the 1950s-1960s, followed by a decline in the 1970s, and then a steady increase since the 1980s.
The source of the claim that Down Syndrome births are declining appears to be from a 2008 update to an earlier article by James Egan, et al, in The American Journal of Obstetrics and Gynecology, An update on trends in down syndrome live births in the U. S. from 1989 to 2005. Egan and the other authors looked at birth certificate data from the National Center for Health Statistics from 1989 to 2005.
A major problem with attempting to count the number of Down Syndrome births this way is that previous studies have shown that Down Syndrome is significantly underreported on birth certificates. Egan, et al, adjusted the reported Down Syndrome live births to try to compensate for this known underreporting. They found,
There were 4,040,958 LB in 1989. They declined to 3,887,069 LB in 1997 and increased to 4,131,210 LB in 2005. The number of actual DS LB declined from 4,801 in 1989 to 3,437 in 1998 and has increased to 4,104 in 2005. The reported DS LB rate (DS LB /10,000 LB) decreased from 5.86 in 1989 to 4.37 in 1998 and increased to 4.90 in 2005. The estimated DS LB rate (DS LB /10,000 LB) has increased steadily from 13.94 in 1989 to 18.71 in 2005. The percent of DS LB declined steadily from the expected number through 1998 before rising slightly.
So are the number of Down Syndrome births increasing or decreasing? The best answer seems to be that we don’t really have a good way to reliably answering that question. According to the CDC, only 41 states currently have systems in place for tracking birth defects, and even within those 41 states there are different methods used to track and report birth defects in live births.
Maybe I’m missing something here, but surely we can do better as a nation to track something as important as birth defects.