Evidence based screening for Down's syndrome

An important lesson in all medicine, but particularly illustrated in screening programmes, is the continued need to review and audit. Serum screening for Down's syndrome, introduced by many health authorities in the past decade,^1–3 is a good example. The original demonstration projects compared the detection rate when Down's syndrome was identified after serum screening with earlier data derived from screening targeted towards pregnancies in older women.² Howe et al from Southampton now challenge some of the assumptions (see p 606).⁴ They found that the Down's syndrome detection rate in one Southampton maternity hospital averaged 68% (and at least 41% in the pregnancies of women aged less than 35), without using serum screening. The higher detection rate without serum studies undermines the cost benefit arguments for such screening and raises questions about what to do next.

One reason for this higher than expected detection rate is a change in the age distribution of pregnant women. In the Southampton study 10% of pregnancies occurred in mothers older than 35, compared with 6% a decade ago.¹ Because of this, the proportion of conceptions with Down's syndrome would increase, as would the detection rate. There has also been an increase in the proportion of fetuses with Down's syndrome, or other trisomies, detected using ultrasound markers of chromosome anomaly.^4–7 In our Nottingham genetic service ultrasound abnormalities are increasingly the trigger for placental biopsy or other intervention.

So, how do we go forward? Should health authorities cease serum screening in favour of more targeted ultrasound facilities? Should serum screening be restricted more, perhaps to pregnancies in women under a certain age? A sensitive question would be whether couples whose screening for chromosome anomalies at the local antenatal clinic was provided by the NHS based on the most relevant evidence should be able to purchase additional testing. Though I would not favour such an option, there does need to be a greater involvement in decision making by pregnant women and their partners.⁸ Let us begin by ensuring that women whose tests (by whatever technique) are “screen negative” are not left with the impression that there is no risk at all. Conversely, those who are screen positive must know that their risk is increased, based on a threshold, but the baby is still likely to be normal.

To the couple who plan, or have embarked on, a much wanted pregnancy the things that matter are any initial risks, the ambience of the antenatal clinic, the availability of the information needed to decide on any tests, and, if uncertainties crop up, an easily accessible account of the options available as well as the gestational age at which a clear diagnosis can be established. The couple's decisions must be informed. In this context it was alarming that nine years ago 12% of antenatal records for pregnancies where the subsequent diagnosis was Down's syndrome did not document whether counselling had been given about risks, prenatal tests, or the available options.⁹

Couples whose pregnancy is shown by screening to be at greater risk, and the few in whom serious fetal abnormalities are confirmed, must be given the information in an appropriate setting and in such a way that they can make the decision that is best for their family. Doctors often allow insufficient time to tackle the sensitive disclosure of possible or confirmed bad news. Since this is partly a training issue, we have developed seminars for senior medical students in Nottingham on breaking bad news (Raeburn JA, Walker D, Raeburn AR, unpublished), but information already exists which every obstetrician, fetomaternal medicine expert, and geneticist should study.¹⁰ In general clinicians are not good at providing patients with opportunities to take informed decisions, especially when the concepts or procedures are complex.¹¹

Those planning a pregnancy and the professionals who help them all need to ensure that relevant risks are addressed using evidence based methods. A forthcoming report of the National Screening Committee will make recommendations about screening in pregnancy for conditions such as Down's syndrome. Also, several comparative studies of serum screening and nuchal thickening as discriminators for pregnancies at higher risk will shortly report their results, as well as studies on serum markers such as PAPP-A,¹² which are valid earlier in pregnancy.

In the meantime Howe et al say that serum screening for an increased risk of Down's syndrome is so firmly established “that it is unlikely that it will ever be tested properly.” Initially I found myself agreeing that a randomised controlled trial was unlikely. Yet what a bad precedent against evidence based medicine that sets. If relevant, change is essential, backed by evidence. The changing nature of the population who want screening and the relentless development of new approaches make audit—if not trials—vital. If adjacent health districts decide on different policies this provides opportunities for continuing comparative audit, involving clinicians, managers, and scientists who passionately believe in their own system.

Papers p 606