Why do counts of terminations of pregnancies diagnosed with Down syndrome in England and Wales differ?

Following yesterday’s Mail on Sunday article and my observation that the data quoted was wrong, many people have asked how the Department of Health abortion statistics can be so different from the terminations recorded by the National Down Syndrome Cytogenetic Register.

The Mail on Sunday reported 693 abortions “linked to Down syndrome” in 2014. This figure comes from figures for England and Wales compiled by the Department for Health (DH). These differ from those reported by the National Down Syndrome Cytogenetic Register (NDSCR).

The DH data comes from forms completed by registered practitioners terminating pregnancies. NDSCR collects data from the cytogenetic laboratories that analyse samples obtained through amniocentesis or chorionic villus sampling (CVS) to diagnose (or not) Down syndrome in pregnancies.

The DH data consistently underreports Down syndrome diagnoses:

2011 2012 2013 2014
National Down Syndrome Cytogenetic Register
Terminations of pregnancies diagnosed with Down syndrome 1134 1131 1108
Department of Health
Abortions by principal medical condition 512 544 590 662
Mentions by principal medical condition 570 613 693
Difference
Number not counted in DH data 622 587 518
Number not counted in DH data as % NDSCR count 45% 48% 53%

These differences were investigated in data matching exercise undertaken by the DH and a review undertaken by the Royal College of Obstetricians and Gynaecologists – published last year.

The NDSCR has recently been merged into a new National Congenital Anomaly and Rare Disease Registration Service (NCARDRS). NDSCR data for 2014 is yet to be published. However, my understanding is that not substantial changes have been detected in the data collected by NDSCR for 2014 and that increases in the abortions recorded as due to a diagnosis of Down syndrome in the DH data are a result of efforts to improve reporting by practitioners terminating pregnancies.