How has survival improved for children with Down syndrome?

One measure of the remarkable improvements in the quality of life experienced by people with Down syndrome is childhood survival rates:

  • In the 1920s, only 20% of children with Down syndrome survived to 10 years of age
  • By 2010, over 90% of children with Down syndrome (in more developed countries) were living beyond the age of 10 years

survival-rates-children-down-syndrome-1917-2003

[Larger image]

Source: de Graaf G., Buckley F., Skotko B. G. (2016). Estimation of the number of people with Down syndrome in the United States. Genetics in Medicine, advance online publication. https://doi.org/10.1038/gim.2016.127

How many babies would be born with Down syndrome each year in the UK in the absence of prenatal screening?

Given we can quite accurately estimate the number of babies expected to be born given the ages of their mothers[1,2], and as we have a good idea how many babies are actually being born, it is fairly straightforward to calculate how many births are being prevented as a result of prenatal screening.

Here are estimates for the UK, based on live births data from the Office for National Statistics for England and Wales, and from National Records of Scotland and the Northern Ireland Statistics and Research Agency:

figure-births-babies-down-syndrome-uk-prenatal-screening-terminations-1991-2015

[Larger image]

This shows the number of live births of children with Down syndrome in each year (green bars) and the 5-year moving average live births of children with Down syndrome (red line). It also shows the additional number of live births we would expect each year had there been no prenatal diagnoses and subsequent terminations (blue bars) and the 5-year moving average for total UK births in the absence of screening (orange line). The reduction rate (the proportion of babies not born as a % of all babies expected to be born), averaged over 5 years, is also provided (yellow dashed line).

Year Live births of babies with Down syndrome 5-year UK average births Additional births in the absence of selective terminations Total births in the absence of selective terminations 5-year UK average births in the absence of selective terminations 5-yr UK average reduction rate (%)
1991 841 326 1167
1992 753 430 1183
1993 704 485 1189
1994 727 485 1211
1995 675 740 546 1222 1194 38%
1996 691 710 569 1260 1213 41%
1997 779 715 518 1297 1236 42%
1998 716 718 605 1321 1262 43%
1999 688 710 641 1329 1286 45%
2000 678 710 663 1341 1310 46%
2001 654 703 710 1365 1331 47%
2002 645 676 757 1402 1352 50%
2003 687 671 804 1492 1386 52%
2004 755 684 813 1567 1433 52%
2005 835 715 773 1608 1487 52%
2006 842 753 843 1685 1551 51%
2007 817 787 943 1760 1622 51%
2008 841 818 965 1806 1685 51%
2009 895 846 907 1803 1732 51%
2010 815 842 1030 1844 1780 53%
2011 844 842 1034 1879 1818 54%
2012 867 853 1027 1894 1845 54%
2013 814 847 1023 1837 1851 54%
2014 848 838 1006 1854 1862 55%
2015 861 847 1022 1884 1870 55%

Some observations:

  • Between 1991 and 2015, an estimated 18,925 fewer babies with Down syndrome were born in the United Kingdom as a result of prenatal screening and diagnosis and subsequent decisions to terminate pregnancies
  • In the absence of prenatal screening, we would currently expect an average 1,870 babies to be born each year with Down syndrome in the UK (compared to the 847 actually born)
  • The average reduction rate has risen from 38% in the early 1990s to 55% today (the rate at which the expected live birth rate is reduced as a result of screening and pregnancy decisions)

A common question:

  • Given registry data suggests that around 23,600 pregnancies diagnosed with Down syndrome were terminated between 1991 and 2015, how could only 18,925 fewer babies have been born? This is because a relatively high number of pregnancies with Down syndrome result in miscarriage or stillbirths. In other words, many of the terminated pregnancies would not have resulted in a live birth if not terminated.

Once again, I must acknowledge helpful discussions about the data with my colleague Gert de Graaf. However, any errors are all mine!

References

  1. Morris JK, Mutton DE, Alberman E. (2002) Revised estimates of the maternal age specific live birth prevalence of Down’s syndrome. Journal of Medical Screening, 9, 2-6. http://msc.sagepub.com/content/9/1/2.full.pdf
  2. Wu J, Morris JK. (2013) Trends in maternal age distribution and the live birth prevalence of Down’s syndrome in England and Wales: 1938–2010. European Journal of Human Genetics, 21, 1016–1019. https://doi.org/10.1038/ejhg.2012.294

How many babies with Down syndrome are currently born each year in the United Kingdom?

I am not sure why, but the figure most often quoted for the number of babies with Down syndrome born each year recently in the UK is 750. I am guessing this figure is drawn on data published the National Down Syndrome Cytogentic Register.  However, this is does not cover the whole of the UK – only England and Wales.

England and Wales is fortunate to have a registry recording the births of babies with Down syndrome. It has reported live births, prenatal diagnoses and pregnancy outcomes since 1989 with a high degree of accuracy. In fact, it offers probably one of the most useful datasets available anywhere in the world.

Rather irritatingly, the work of NDSCR has recently been transferred the National Congenital Anomaly and Rare Disease Registration Service (NCARDRS) and the Congenital Anomaly Register and Information Service (CARIS) for Wales. The move around means no data for 2014 or 2015 has been published for England. I am told it will be next year before this appears. CARIS only appears to be reporting total cases, which is unhelpful. Hopefully, the NCARDRS and CARIS will pool their data to continue to publish the England and Wales data series.

Scotland reports live births back to 1991, though apparently not prenatal diagnoses and pregnancy outcomes. Northern Ireland does not seem to publish data on live births of babies with Down syndrome.

Drawing on these data sources, we can compile a total for UK live births going back to 1991:

figure-births-babies-down-syndrome-uk-england-wales-scotland-northern-ireland-1991-2015

As Northern Ireland does not seem to publish a count of live births of babies with Down syndrome, I have estimated them by taking published data on all live births broken down by age of mother and multiplying these by the formula expressing the maternal-age-related chance of having a baby with Down syndrome (to get the expected number of ‘natural’ live births in the absence of selective terminations). I have then reduced this number at the same rate as ‘natural’ live births have been reduced by terminations in each year in England and Wales. To estimate births in 2014 and 2015 for England and Wales, and for 2012-15 for Scotland I have again calculated the expected ‘natural’ live births and reduced by the recent average reduction rates in each area. (Of course, the reduction rates may have risen in the past few years with the private availability of NIPT, but we do not have the data yet to see this.)

Given these assumptions, we can calculate that the average annual live births of babies with Down syndrome in the United Kingdom over the past 5 years is 847.

Year England and Wales Scotland Northern Ireland UK 5-year UK average
1991 737 75 29 841
1992 663 65 25 753
1993 621 60 23 704
1994 639 64 24 727
1995 581 73 21 675 740
1996 607 62 22 691 710
1997 667 88 24 779 715
1998 633 60 23 716 718
1999 604 63 21 688 710
2000 594 64 20 678 710
2001 580 54 20 654 703
2002 589 36 20 645 676
2003 616 51 20 687 671
2004 665 68 22 755 684
2005 747 64 24 835 715
2006 764 53 25 842 753
2007 723 70 24 817 787
2008 754 61 26 841 818
2009 793 76 26 895 846
2010 737 53 25 815 842
2011 758 61 25 844 842
2012 779 63 26 867 853
2013 728 62 24 814 847
2014 759 64 25 848 838
2015 774 62 26 861 847

As ever, I must acknowledge helpful discussions about the data with my colleague Gert de Graaf. As soon as we can get the most recent data I hope we will publish a larger historical births and populations dataset. However, any errors are all mine!

Selective terminations of babies with Down syndrome have not risen by 35% in three years

piece in the UK Mail on Sunday claims Department of Health statistics show “women choosing to abort babies with Down’s syndrome and other serious disabilities soars 34% in three years” and that “the biggest proportion was linked to Down’s syndrome, with 693 terminations last year compared with 512 in 2011”. Data collected by the National Down Syndrome Cytogenetic Register (NDSCR) shows this is not the case.

UPDATE (15 June): Amended to correct dates. Mail article refers to 2014 Department of Health data, while NDSCR has only reported to 2013. It remains the case that (a) NDSCR data reports 2% drop between 2011 and 2013 – which is not consistent with 35% rise over 3 years, and (b) DoH abortion statistics since 2011 consistently differ from NDSCR data. More here.

There is a remarkable headline in The Mail on Sunday today claiming terminations of babies with Down syndrome have “soared 34% in three years”. It quotes “an investigation into figures published by the Department of Health”.

In England and Wales, there is a national registry collecting data on prenatal and postnatal diagnoses. Their data (collected since 1989) is widely cited in academic studies and is 94% complete. I’m not sure where the Mail on Sunday (and perhaps more worryingly, the Department of Health) are sourcing their information, but it is way off.

The article claims there were 512 terminations of pregnancies diagnosed with Down syndrome in 2011 and 693 in 2014 – a rise of 181 (35%).

NDSCR reports 1,134 terminations of pregnancies diagnosed with Down syndrome in 2011 and 1,018 in 2013 – a fall of 2% over two years. NDSCR has not yet reported on 2014.

This graph shows the number of terminations of pregnancies diagnosed with Down syndrome (columns) and terminations as a percentage of all diagnosed cases (prenatal and postnatal).

down-syndrome-terminations-england-wales-2004-2013

(Note: The % terminated statistic is not the reduction rate – i.e. the rate at which live births are reduced as a result of screening and selective terminations. This is because approximately 23% of the terminated pregnancies would not have resulted in a live birth.)

Terminations of babies with Down syndrome in England

Last weekend, the UK Sunday Times reported that many terminations of babies with Down syndrome are not being correctly recorded by the Department of Health. In many reports that have since appeared online it has been stated that in England 92% of babies with Down syndrome are terminated. This is incorrect.

You would think from some of the reports and comment that have appeared online in the past week that the births of virtually all babies with Down syndrome in England are prevented through prenatal screening and selective terminations. This is not the case:

It is true that 92% of diagnosed pregnancies are terminated. However, this is very different from 92% of all pregnancies. Many people choose not to screen. Many who are screened and receive a ‘high risk’ result choose not to have invasive diagnostic testing as it carries a risk of miscarriage. Some who are screened and do not receive a ‘high risk’ result will, in fact, go on to have a baby with Down syndrome.

Overall, the rate at which live births are reduced due to prenatal diagnosis and selective terminations in England and Wales in recent years was around 50%-55%.

The costs of prenatal screening services in England

In response to question from Fiona Bruce MP, the Department of Health has published estimates of the costs of prenatal screening in England.

The Chair of the UK Parliamentary Inquiry into Abortion on the Grounds of Disability, Fiona Bruce MP, recently tabled a question requesting the costs of prenatal screening for Down syndrome. In reply, the Department of Health provided estimated costs of the National Health Service Foetal Anomaly Screening Programme – which includes screening for Down syndrome. These offer a base, lower and upper scenarios at 2009/10 and 2010/11 prices for England:

Base case Lower Upper
2009-10 prices 2010-11 uplift 2009-10 prices 2010-11 uplift 2009-10 prices 2010-11 uplift
Estimates of birth used 671,058 650,000 700,000
Central coordination 1,895,735 2,000,000 1,895,735 2,000,000 1,895,735 2,000,000
Total screening strategy from the offer to the screening results 17,080,673 18,020,110 16,544,677 17,454,634 17,817,344 18,797,298
Invasive diagnostic procedure (CVS / amniocentesis) 7,802,120 8,231,237 7,557,287 7,972,938 8,138,617 8,586,241
Cost of healthy foetal loss 93,600 98,748 90,663 95,649 97,637 103,007
Cost of those T+ who terminate 486,739 513,510 471,461 497,391 507,732 535,657
Total cost of T21 screening population live births 983,759 1,037,866 952,889 1,005,298 1,026,188 1,082,628
Total screening pathway cost 28,342,626 29,901,470 27,512,712 29,025,911 29,483,253 31,104,831
Total cost, excluding continued pregnancies 27,358,867 28,863,604 26,559,823 28,020,613 28,457,065 30,022,203

In 2010/11, approximately 850 babies who would have been born with Down syndrome were not born as a consequence of the screening programme. I therefore estimate the cost of each live birth ‘prevented’ to be around £34,000.