What Percentage of Spina Bifida Kids Are Aborted?

Because of the sophistication of ultrasound and other prenatal diagnostic techniques, doctors can diagnose certain disabling conditions prior to birth with a fair amount of accuracy.  Spina bifida is one of those conditions.  In a recent newsletter, I relayed the testimony of one young man with spina bifida at the University of Texas at San Antonio.  I made the claim in that newsletter that more than half of children diagnosed in the womb with spina bifida are killed through legal abortion. (I meant, in the U.S.)

Is this claim accurate?

After looking at a number of studies, I’ve concluded we don’t really know.  (So, you’ll notice I changed the figure in my newsletter.)  I had based my “more than 50%” claim on the Mansfield study listed below, and I felt my estimate was safe.  From that study, we can know that during the period from 1980-1991, two U.S. studies counted 15 fetuses prenatally diagnosed with spina bifida and 15 of those were killed by abortion (100%).  But the sample size here is too small to help us make conclusions about the general population then, and much less today.  From the other studies that I cite below, we can know that in the late 80’s, there was a 30% termination rate in California, and that in the late 80’s to mid 90’s a 48% termination rate was recorded in Hawaii.  By my reading of the data, we can also know that in 1995 in Houston, there was at least a 37% termination rate of spina bifida children after prenatal diagnosis.  And we have some data for six states for the period from 1985-1994 (see below).  We have a few studies from Europe that show 66%, 72%, and 78% termination rates (see below), but we can’t say much confidently about U.S. practice from European studies.

There also seems to be some evidence that the termination rates in the U.S. vary by geographic location, timing of diagnosis, and other factors.  Another concern is the age of the data in these studies.  Only one United States study has any data from the 2000’s (Cragen, et.al 2009), but the termination rate from that study is unclear (it may be 34% and it may be higher, but we can’t know from the published data).  Only two from Europe have any data from the 2000’s (Aguilera, et.al. and Amari, et.al).

In many scientific studies (of risk factors for breast cancer, for example), the age of the study doesn’t affect it’s relevance to the discussion, precisely because the question being considered isn’t time-dependent.  But studies of how many spina bifida children are aborted after prenatal diagnosis is really a sociological question rather than a biological or epidemiological question.  In other words, if we’re asking “How many parents abort a child that is diagnosed in utero with spina bifida?” we are asking a question about current behaviors of a certain community of human beings, and older studies are only helpful if we can also demonstrate that relevant social and medical factors haven’t changed.  Examples of social and medical factors that should be factored in when attempting to extrapolate from older studies on this question include…

• “Have attitudes towards certain kinds of abortion changed?”  This sociological question is important because people are less in favor of late-term abortion now than in the past, and many times a spina bifida diagnosis comes later in the pregnancy.  If people are against late-term abortion and late-term abortion is the only option other than giving birth to their spina bifida child, then they may very well choose to give birth.
• “Are we more able now to diagnose spina bifida earlier in pregnancy?” This medical question is important to the question at hand because people are generally more in favor of abortion if it’s done earlier.  In other words, if a spina bifida diagnosis can be reliable early in pregnancy when people feel more comfortable with abortion, they are probably more likely now to abort spina bifida children.
• “Are there now more successful treatments for spina bifida than there were even ten years ago?”  This medical question is important to the question at hand because people are sometimes influenced by quality-of-life considerations when deciding about abortion.  So, if they perceive the quality of life of their spina bifida child can be relatively good, it may decrease the willingness to abort in spina bifida cases.  In other words, if there are more successful treatments now than there were, it’s reasonable to assume more people would opt for giving medical treatments to their unborn child rather than killing her.

So, it’s difficult to conclude much about what percentage of unborn children diagnosed in utero with spina bifida are killed by abortion (in the U.S.).  Sure, we can guess at how many parents elect to have their spina bifida kids killed, but at this point, my view is that we don’t have much more than guessing.

Following are some of the studies I reviewed, along with my comments.

Studies Referring to Data from the United States and Europe (1):

• Mansfield C, et. al., Termination rates after prenatal diagnosis of Down syndrome, spina bifida, anencephaly, and Turner and Klinefelter syndromes: a systematic literature review.  Prenatal Diagnosis, Sept. 1999; 19(9):808-12.  This oft-quoted 1999 study by Mansfield, et. al. reviewed four European studies and two United States studies.  Four of the studies, on my reading, have sample sizes too small to make very strong generalizations about the question at hand (Belgium, Italy, and two from the U.S.).  The two studies with more significant sample sizes, UK and France, reported termination rates of 61% and 63% respectively.  Note the years of the spina bifida studies, though: 1987, 1991, and 1995 (Note: this is the year of the study; both the 1991 and 1987 studies refer to data from the 80’s; the 1995 study refers to data from 1990-1991).  I’m not sure much we can accurately conclude about current practice from these studies.  See an excerpt from Table 1 below.

Mansfield C, et. al., Termination rates after prenatal diagnosis of Down syndrome, spina bifida, anencephaly, and Turner and Klinefelter syndromes: a systematic literature review.  Prenatal Diagnosis, Sept. 1999; 19(9):808-12; Table 1 (excerpt).

Studies Referring to Data from the United States (5):

• Ellen M. Velie and Gary M. Shaw, Impact of Prenatal Diagnosis and Elective Termination on Prevalence and Risk Estimates of Neural Tube Defects in California, 1989–1991,  Am J Epidemiol 1996;144:473-9.  This study from California shows about 30% termination rate for unborn children diagnosed with spina bifida.  The data is about 20 years old, though.

• Mathias B. Forrester, Ruth D. Merz, and Paula W. Yoon.  Impact of Prenatal Diagnosis and Elective Termination on the Prevalence of Selected Birth Defects in Hawaii.  Am. J. Epidemiol. (1998) 148(12): 1206-1211.  This study from Hawaii shows a 48% termination rate for spina bifida diagnosed unborn children.  All of the data used in the study, though, is at least 15 years old; some of it is 24 years old (1987-1996).

• Cragan, et. al.  Surveillance for anencephaly and spina bifida and the impact of prenatal diagnosis–United States, 1985-1994.  MMWR CDC Surveill Summ. 1995 Aug 25;44(4):1-13.  See Table 2.  This study summarizes data on termination rates for spina bifida from 6 states or areas in the United States: California (29%); Iowa (19%); Arkansas (3%); Atlanta, Georgia (26%); South Carolina (25%); and Hawaii (23%)  This data is also at least fifteen years old.  View the abstract for a summary of the study here.

• Cragen, et.al.  Including Prenatal Diagnosis in Birth Defects Monitoring: Experience of the Metropolitan Atlanta Congenital Defects Program.  Birth Defects Research (Part A) 85:20-29 (2009)   It’s unclear from the published data in this study what the termination rate of spina bifida kids after prenatal diagnosis is, but it’s likely that is was at least 34% and was probably higher. [Notes on how I arrived at the 34% figure: Although Fletcher, et.al. concluded that this study showed a 35% termination rate, the table from which they appear to have drawn that conclusion is misleading.  See Table 2 on Page 25.  It includes live births and still births that are not necessarily following a prenatal diagnosis.  The study is attempting to show how including termination numbers (not rates) increases the overall prevalence numbers of various deformities.  So, without more interaction with the authors about unpublished data they may or may not have, we can’t conclude for certain what the termination-after-spina-bifida-prenatal-diagnosis rate is from this study either.  The dates of the data of this study (1995-2004) gave me the most promise of any study I looked at that we might get some help with the question at hand, but all we can really conclude regarding spina bifida children from this study is that during the 9 years of the study, 64 spina bifida children were killed by elective abortion, 102 were born, 14 were stillborn, and 7 had unknown outcomes (although the authors guess that termination in an outpatient facility was the likely outcome).  It’s likely that many of the abortions followed a prenatal diagnosis, but we can’t know for sure.  If that’s the case, the number might be 34% (64/187).  If we use that as a baseline and include the “unknown outcomes,” and if we also consider the fact that the number of live-born and stillborn kids prenatally diagnosed is likely lower than the total number of those kids, then the percentage of terminations after prenatal diagnosis would be higher (64 aborted / 187-n prenatally diagnosed).]

• Waller DK, et. al.  Frequency of prenatal diagnosis of birth defects in Houston, Galveston and the Lower Rio Grande Valley, Texas 1995.  Fetal Diagn Ther. 2000 Nov-Dec;15(6):348-54.  This study shows at least a 37% termination rate for spina bifida after prenatal diagnosis.  Keep in mind, though, that the data for this study is now more than 15 years old, so we should be cautious of conclusions we draw from the study about current practice today.  Also, keep in mind that in 8 of the 10 cases of abortion for spina bifida in this study, the child also had other serious abnormalities, so we can’t really conclude that the spina bifida part of the diagnosis was the decisive factor for the couples who elected for abortion.  [Notes on how I arrived at 37%: Although Fletcher, et.al. (page 2) cited this study as reporting a low 7-10% termination rate for spina bifida, that’s a misread of the abstract, in which the authors say that between 7 and 10% of the cases of elective termination were not included.  Since this study is not really attempting to determine termination rates after prenatal diagnosis, Table 2, which lists a 16% number, along with an 8% number, is confusing for our purposes.  Unfortunately, my attempts to get clarification from the author of the study were unsuccessful.  So, I went back to the study and read it more carefully.    As I said, Table 2 appeared to report a 16% termination rate.  Upon a more careful reading of that Table 2 and the study, I realized that the percentage in the table refers to fetuses terminated out of the entire number of fetuses in the study.  The study explicitly says that only a third of the fetuses included in the study were prenatally diagnosed.  (In fact, that’s the purpose of this study, to determine how many babies with deformities were prenatally diagnosed.  It’s not intending to tell us what people did with their children after prenatal diagnosis.  So, our work here is an extrapolation from the data.)  In other words, 10 of the 64 children (16%) with spina bifida in this study were killed by abortion.  But 64 is the total number of spina bifida children in this study.  If you look in Table 1 to find how many of those children were prenatally diagnosed, the number is 27 (22 with only spina bifida + 5 with spina bifida and other deformities).  While we can’t be positive from this data that all of the children listed as being killed by abortion in Table 2 (10) were killed following a prenatal diagnosis, it’s likely.  If that’s the case, the percentage of parents who decided to kill their spina bifida child after prenatal diagnosis is at least 37% (10/27).  Remember that some of the terminations were excluded, so even this termination rate is probably low.]

Studies Referring to Data from Europe (3):

• Aguilera, et. al.  Prognosis of spina bifida in the era of prenatal diagnosis and termination of pregnancy.  Fetal Diagn Ther. 2009;26(2):68-74. Epub 2009 Sep 11.  This Bristol, UK study had a sample size of 74.  72% of the unborn children diagnosed with spina bifida were killed by abortion.  The study refers to data from 1999-2007.

• Amari, et. al.   Prenatal course and outcome in 103 cases of fetal spina bifida: a single center experience.  October 2010, Vol. 89, No. 10 , Pages 1276-1283.  This study, which refers to data from Germany (1993-2008), reports a 66% termination rate.  A total of 103 children were prenatally diagnosed with spina bifida, and 68 were killed by abortion.

• E. Garne et. al.  Prenatal diagnosis of severe structural congenital malformations in Europe.  Ultrasound Obstet Gynecol 2005; 25: 6–11.  This study, which refers to data from 1995-1999 from 17 European-based registries of congenital malformations, reports a termination rate of 78%.  405 children were prenatally diagnosed with spina bifida.  314 were killed by abortion.

Two Documents from the 1980’s
As I was about to publish this piece, David Lee pointed me to two other notable documents from the 1980’s.

• Guide to Clinical Preventive Services: Report of the U.S. Preventive Services Task Force. 2nd edition. US Preventive Services Task Force.  Washington (DC): US Department of Health and Human Services; 1996.  Section 1, Chapter 42: “Effectiveness of Early Detection”.  This report claims that a number of studies confirm that “Induced abortion is sought by the majority of women who choose to be screened and whose screening tests reveal neural tube defects, thus leading to a decreased birth prevalence of affected infants among screened women.”  It lists numerous studies.  One of them, from South Wales, follows.  I may take all of these up in a subsequent post, but the dates of the studies are 1984, 1982, 1983, 1985, 1983, 1979, and 1991, so even if the studies do show that a majority seek abortion in the case of spina bifida, the data would be almost 30 years old, and some of them are from Europe, so then we would have to speculate on the relevance of that data to the situation in the United States today.
  • Roberts CJ, et.al.  The efficacy of a serum screening service for neural-tube defects: the South Wales experience.  Lancet. 1983 Jun 11;1(8337):1315-8.  A quick look at the abstract of this study reveals a 40.7% termination rate for spina bifida.  Note: If one isn’t careful, he’s apt to read the 56.1% termination rate for “open neural tube defects” as the termination rate for spina bifida.  But 56.1% is the termination rate in this study for both spina bifida (40.7%) and anencephaly (66.6%) combined.  As far as I can tell, it is uniformly true in studies like this that anencephalic kids are more likely to be aborted than spina bifida kids.  So, we must take that into account when we’re attempting to understand spina bifida in isolation.

• Faden, R.R., et.alPrenatal Screening and Pregnant Women’s Attitudes toward the Abortion of Defective Fetuses.  Am J Public Health, 1987; 77:288-290.  This study analyzed the attitudes of 490 Maryland women in a prenatal screening program in 1980-82.  Figure 1 seems to indicate that as the certainty about the diagnosis increases, more people are willing to abort.  If I’m reading that Figure correctly, each of the women in the study only got to pick one of the options of the certainty level of the diagnosis at which they would choose abortion (25%, 50%, 75%, 95%, 100%).  Keep in mind that the data in the this study is almost 30 years old.