Abstract
The majority of women in the UK choose to be screened antenatally for the genetic condition Down's syndrome. However, the majority of these women are not given the test for Down's that has been recommended by the NHS. The recommended test is the most accurate and has other benefits, and should be readily available on the NHS to all women. Providing tests that are less accurate regrettably leads to more babies miscarrying and the risk of Down's being missed. This article aims to raise awareness of the problem and to consider the medicolegal issues.
Testing for Down's syndrome
About 60,000 people in the UK have the genetic condition Down's syndrome, and about 1 in every 1000 babies is born with it. 1 It is the largest cause of learning disability in the UK. Every pregnant woman in the UK should be offered a test for Down's, and 59% proceed with it. The test is to assess the risk of Down's; it is not diagnostic and a further invasive procedure will be required to ascertain whether the baby does have Down's. Since April 2009, a chance greater than 1:150 of Down's is deemed high risk.
There are four main tests for Down's offered throughout the NHS, 2 and maternal factors such as age and previous pregnancies are also taken into account:
1–2 Triple or quadruple blood test: this is done in the second trimester at 15–20 weeks. The blood is tested for three (triple) or four (quadruple) hormone levels that are indicative of Down's, depending on whether the Trust offers triple or quadruple testing;
3 Ultrasound scan: a scan is taken at 11–14 weeks to measure the nuchal translucency, a fluid pocket at the back of the baby's neck. All babies have nuchal translucency, and the higher the measurement, the higher the chance of Down's;
4 Combined test: this is taken at 11–14 weeks of pregnancy and involves the mother having a blood test and an ultrasound scan.
Unlike the first three tests, the combined test gives a double assessment of risk – blood and scan – in the first trimester. There is a regularly updated map of the country provided by the UK National Screening Committee which shows the tests given in each area (
The recommended test and failure to implement it
The antenatal test for Down's syndrome that has been recommended by the UK National Screening Committee (which advises the NHS), the NHS Fetal Anomaly Screening Programme and NICE, 4 to be offered to all pregnant women is known as the combined test. 5 The only pregnancies for which the combined test is not recommended is in the 15% of women who present to hospital late in the pregnancy.
The combined test is the most accurate and the least likely to give a false positive, i.e. to wrongly assess a baby as high risk. But, despite the combined test having been identified for national rollout in 2003, 70% of NHS Trusts in England (and the vast majority of Trusts nationally) still only offer blood or scan tests, which are more likely to give a false-positive or false-negative result. The map referred to above [JH1]shows that the majority of women nationwide are getting one of the blood tests at 15–20 weeks.
The accuracy of tests for Down's is assessed by the detection rate and false positive rate. The accuracy targets set by the NHS and NICE were for national detection rates to be greater than 75% and false-positive rates to be less than 3%, by 2007. 6 These should have been achieved by now but it was reported recently that the false-positive rate is still 5% and it has apparently taken two years to get it down from 5.9%. 7 It is anticipated that if combined testing were introduced nationwide, the NHS and NICE requirements would be met. 8
A mother who is told that she is at high risk of having a baby with Down's will usually undergo further invasive investigation – amniocentesis or chorionic villus sampling (CVS) depending on gestation – to confirm whether the baby does have Down's. Amniocentesis involves having a needle inserted into the sac containing the baby's amniotic fluid, and CVS involves having a needle inserted into the placenta. There is a 0.5–1% chance of amniocentesis causing miscarriage and a 2% chance of miscarriage due to CVS (although many people think the risk is 1% for both).
The more women who are told wrongly that their baby is at high risk, the more amniocentesis and CVS procedures will be carried out and the more babies will miscarry as a result. It is estimated that 146 babies – normal babies without Down's – die each year as a result of false positives. 7 Sadly, the proper introduction of the combined test would not eliminate these deaths entirely, as there would be some false positives even with the combined test. But, it would reduce them significantly to about 88, thereby saving about 58 babies each year (i.e. a reduction of two-fifths).
In addition to better accuracy, the combined test has other benefits. Parents are given the test results in the early stage of pregnancy, so termination can be considered before the pregnancy is too advanced, rather than at 4–5 months if only a blood test is given. Also, it is convenient as the nuchal translucency scan and bloods can be taken at one hospital visit, and the nuchal translucency test can be done at the same time as the dating scan depending upon when the woman presents.
The 30% of trusts that have introduced the combined test are based in some parts of London, and parts of the south-east, south-west and the Midlands. Those women living in other areas and who can afford to do so may have the combined test privately at a cost of around £270, assuming that they are made aware of it.
The reasons for failing to implement the combined test
The usual reason for a trust's failure to provide particular treatment is lack of funds. However, it appears that the costs of doing the combined test are not dissimilar to the costs of doing the blood test or scan. In Screening for Down's Syndrome: UK National Screening Committee Policy Recommendations 2007–2010: Model of Best Practice, it is stated that money can be transferred from other tests into implementing the combined test and that the ability to calculate the risk of Down's is already up and running, although some changes to the computer software will be necessary. The equipment for CVS is said to be the same as for amniocentesis. Indeed, it is recognized by the NHS that fewer false positives will save the NHS money in the long term – fewer women will need counselling after having wrongly been given a false-positive result and CVS and amniocentesis procedures will reduce. 9 Also, NICE will have considered the cost-effectiveness of the combined test before recommending it.
The NHS and UK National Screening Committee has stated that the primary problem is a lack of trained sonographers to carry out the ultrasound part of the combined test to measure nuchal translucency. 10 Sonography is also not given sufficient prestige, making it an unappealing career choice. Both these problems are resolvable, and although training costs may be high at the outset, eventually the expertise will be passed in-house from senior team members to junior ones, and the costs will level out.
Steps are being taken by the UK National Screening Committee in relation to training and a number of NHS papers have been produced to try to introduce standards and policies; steps are also being taken to make sonography more prestigious.
Action plan
It is now planned that the combined test will be available nationally on the NHS by April 2010, according to the NHS Fetal Anomaly Screening Programme. 9 However, Dr Anne Mackie, head of NHS screening programmes, has admitted that the deadline will not be met, 7 but the combined test should be available everywhere by the end of 2010.
Legal remedies
The failure to introduce combined testing raises interesting questions about possible actions that could be brought, both in clinical negligence and public law.
Clinical negligence
Where the baby dies
The question is whether a woman could bring a claim where she has been told, wrongly, that she was at high risk of having a Down's baby after having been given a blood test or scan (rather than the recommended combined test), and then loses her baby as a result of the invasive testing, amniocentesis or CVS. Such an action would be very difficult in both negligence and causation. In relation to negligence, the allegation would be that she should have been given the combined test. However, the unavailability of the combined test would no doubt be defensible under the Bolam test given that 70% of trusts in England and the vast majority of trusts nationally do not offer this. In addition, it would not be possible for the mother to point to the recommendations of the NHS, the UK National Screening Committee and NICE to prove that a trust has negligently failed to give her the proper test, as guidelines and recommendations do not, in themselves, determine the standard of care expected under Bolam. A challenge would have to be made under Bolitho that the failure to implement the combined test did not stand up to logical analysis and, although the arguments would be compelling, it may be that the political and practical ramifications of such a decision may make a judge reluctant to uphold this.
In relation to causation, the woman would have to show that the results of the combined test would have indicated that she was at low risk of having a Down's baby, and therefore would not have undergone invasive testing and lost her baby. The first problem is that the information required to carry out the combined test to ascertain whether she would, in fact, have been low risk if the combined test had been done would not be available – only the blood test or scan would have been done rather than both. In the absence of such evidence, it would be difficult to argue that, on the balance of probabilities, if the combined test had been done it would have indicated that she was at low risk, bearing in mind that combined testing only reduces the number of miscarriages due to false-positive results from 146 to 88, i.e. a reduction of two-fifths, which is less than 51%. However, when the combined test does become available nationally, there may well be developments and improvements in the testing and software which will result in the balance of probabilities test being able to be established.
There may, of course, be other claims for negligence as a result of failings in testing for Down's syndrome, whichever test is given, for example failing to obtain the relevant maternal information or administrative errors which result in a woman being told wrongly that she is high risk and subsequently losing her baby after invasive testing. There may also be failings in the performance of the invasive testing, which cause the baby to miscarry, although in these cases it will be difficult to assess whether the miscarriage was due to fault or known risk.
When Down's is missed
The question is whether a woman could bring a claim, which would be of high value, in negligence based upon the argument that if she had been offered the more accurate combined test, the risk of Down's would have been picked up and confirmed by further investigation. Again, the data required to ascertain what the risk would have been if the combined test had been given – blood test and scan results – would not be readily available to prove negligence. It would be difficult to argue that the fact that the baby had Down's necessarily proves negligence, because there is a failure in detection even in the combined test. However, this argument may become more feasible as improvements are made in the combined test over time. In addition, for now, similar problems arise in relation to causation as above.
There may, of course, be a claim for negligence if the possibility of Down's is missed for some reason other than the failure to implement the combined test, and the baby is born with the disability, for example administrative errors or failures in the interpretation of the scan. In these cases, causation will be established if the mother would have had a termination had the tests been done properly and had appropriate advice been given. The basis of the action is for wrongful birth, in that the mother was denied the chance to terminate the pregnancy. Ninety-two percent of women terminate pregnancies after having received a diagnosis of Down's, so causation is unlikely to be a problem. It is well-established that where a disabled child is born as a result of the lost opportunity to terminate a pregnancy, only the costs incurred as a direct result of the disability can be claimed. See, for example, Rand v East Dorset Health Authority, 11 and interesting articles on wrongful birth claims by Ben Troke, and Anne Winyard and Darren Conway listed in the footnotes. 12–14
Consent
The fact that the majority of trusts do not offer the combined test raises questions of consent, and the possibility of a clinical negligence action. In these trusts, doctors and sonographers offering the blood test or scan can be expected to know the following: that it is less accurate than the combined test, more likely to result in the loss of the baby as a result of amniocentesis or CVS, more likely to fail to detect the risk of Down's, and is not the test recommended by the NHS. They should advise patients of these factors, and that the recommended combined test is available privately. It is difficult to see how consent could ever be informed without this information.
The NHS has produced a leaflet, ‘Screening tests for you and your baby’, for use in consultations relating to Down's syndrome. The leaflet includes reference to blood testing which it says is usually done between ‘10 to 18 weeks’. But, we also know from the NHS Fetal Anomaly Screening Programme Model of Best Practice that blood tests alone taken before 13 weeks do not meet the detection rate of at least 75% and false-positive rate of less than 3%. So, the information recommended in the leaflet to be given to the patient refers to a test which the NHS knows fails in accuracy. In addition, there is no mention in the leaflet that the combined test is the most accurate, has been recommended by the NHS, and that it is available privately. This leaflet, and the advice being given to patients, is in urgent need of revision and is putting practitioners at risk of complaint and litigation.
There are clear indications that the Courts are willing to include an obligation to advise patients of the availability of private treatment within the expected standard of care. In EA and AA v King's Lynn & Wisbech Hospitals NHS Trust 2007, 15 the mother was carrying twins, which meant that the blood test available at the trust was not appropriate for testing for Down's syndrome, but there was no discussion about an alternative test. Both twins were born with Down's. The trust was held to have been negligent in failing to advise the Claimant of alternative options, whether funded privately or on the NHS. An award of £1.5 m was made. Although the facts were rather unusual, the principle could easily and reasonably be extended to cover circumstances where the recommended combined test is not offered.
In addition, it is also clear that in some circumstances the duty to advise of the risks of a procedure includes providing details of the comparative risks of alternative treatments. In Birch v UCL Hospital NHS Foundation Trust 2008 16 the claim succeeded on the basis that the patient had been advised of the risks associated with catheter angiogram, a slightly invasive procedure with a 1% risk of stroke, but had not been given the comparative risks associated with an MRI, which carries no risk of stroke. There seems to be no logical reason why this could not extend to the Down's situation, where the risk of error in one test is greater than in another test, and especially where one test and not the other has been recommended by the NHS and NICE.
In relation to causation, it would not be possible for a woman to establish that ‘but for’ the failure to provide the combined test, she would not have either lost her baby or had a baby where the risk of Down's had been missed, for the reasons mentioned above. However, this may change as technology develops. In addition, it might be possible to apply the principles in Chester v Afshar 2004 17 where a claim succeeded on the basis that a patient had not been told of the 1–2% risk associated with a back procedure. In Chester, the argument was that if the patient had been warned of the risk, the operation would not have taken place at that time or with that surgeon, and because the risk was so low the outcome may have been different. There were public policy reasons for giving a remedy where a doctor had failed in his or her duty to give an appropriate warning. Although the full ambit of Chester is yet to be determined, by analogy it could be argued that a woman who is not advised of the lower risk associated with combined testing, who but for the negligence would have had the testing done on a different day with a different possible outcome, should succeed in her claim.
Novel arguments could also be made for a possible infringement of articles 8 and 12 of the European Convention on Human Rights, the right to respect for private and family life and right to marry and have a family, although these are unchartered territories. An argument for infringement of the babies' ‘human’ rights could also be made, although these have been resisted by the Courts so far – see below – but this is an area ripe for development. These claims would probably only be for nominal damages, and would not provide full compensation.
Public law
The question is whether judicial review could be deployed to secure combined testing either for an individual or on the basis of a pressure group seeking timely implementation of the recommended test.
There are reasonably good reasons for thinking that it could. Before examining the merits of such a challenge it is necessary to understand the context in which such a claim would have to be made.
The National Health Service Act 2006 imposes a statutory duty upon the Secretary of State for Health to provide, in essence, a National Health Service. The core duty to ensure the provision of a health service is laid upon the Secretary of State in extremely broad terms in sections 1 and 3 of the 2006 Act. Section 1 provides that:
‘It is the Secretary of State's duty to continue the promotion in England and Wales of a comprehensive health service designed to secure the improvement (a) in the physical and mental health of the people of those countries, and (b) in the prevention, diagnosis and treatment of illness.’
Section 3(1) provides that:
‘It is the Secretary of State's duty to provide throughout England and Wales, to such extent as he considers necessary to meet all reasonable requirements:
…; …; medical, dental, nursing and ambulance services; such other facilities for the care of expectant and nursing mothers and young children as he considers are appropriate as part of the health service; …; such other services as are required for the diagnosis and treatment of illness.’
These sections give the Secretary of State a broad discretion as how best to secure the ends aimed at by the Act. Sections 1 and 3 of the 2006 Act do not impose an absolute duty to provide any specific services. The effect of these provisions have been considered in numerous cases and often the patient requiring treatment not routinely provided will be without a remedy (unless irrationality can be shown), see R v Cambridge Health Authority ex p B 1995 18 and R v North West Lancashire Health Authority ex p A 2000. 19 In reality, a patient's local Primary Care Trust (usually the commissioner of health services for a particular geographical area) can determine not to provide particular treatment subject only to the qualification that such a decision must not be founded in the well-known principles of public law – illegality, irrationality and procedural impropriety.
Nearly all Primary Care Trusts, whose delegated function is to commission health services within each area, operate a system of funding for treatment not routinely commissioned, often referred to as ‘exceptional’ treatment. Such systems have in the past existed under various guises, e.g. ‘extra-contractual referrals’ and ‘out-of-area treatments’. They are now described under the current NHS policy of ‘payment by results’ as simply ‘non-contract activity’ or ‘non-contract treatments’. Trusts are encouraged by the Department of Health to have a policy in respect of non-contract treatments which will explain in general terms the reasons for a particular treatment not being routinely available and to provide a procedure setting out how a request by a patient for a non-contract treatment will be handled. The Primary Care Trust's deliberations when considering any request for a non-contract treatment will often include that decisions should be based on the evidence about a treatment's clinical and cost-effectiveness, including any NICE guidance.
In relation to combined testing, a judicial review challenge by an individual, say a pregnant 37-year-old mother who wants the combined test (to reduce the risk of unnecessary fetal death as a result of amniocentesis or CVS and to reduce the chance of the risk of Down's being missed) but has been told that the combined test is not available at her local trust, would probably be commenced against the background of her seeking the funding of the treatment on an exceptional basis from the trust as a non-contract treatment. One would assume she could make the following arguments in support of her case:
The recommendation of the UK National Screening Committee and NHS Fetal Anomaly Screening Programme of the combined test; The apparent failure of the timely implementation of that recommendation by the Trust; The recent revision of the NICE clinical guideline, ‘Antenatal care: routine care for the health pregnant woman’ which advises that the first trimester combined test is used; The likely cost/benefits of adopting the combined test – a commensurate saving on counselling and repeat testing/unnecessary investigative procedures; The reduction in the risk of unnecessary fetal death and in the risk of Down's being missed; The pregnant mother's rights under Article 8 and Article 12 of the ECHR.
If the request was denied by the trust's non-contract treatment panel, the reasons would need to be scrutinized. Her likely challenge would then either be to the policy operated by the trust itself, or to its failure properly to apply it, and/or an argument that the conclusion not to fund is irrational. Such a challenge might follow the pattern of recent challenges to trusts on the basis of failure to properly apply an exceptional circumstances policy, such as R (Rogers) v Swindon Primary Care Trust 2006
20
where the Claimant challenged the refusal by her Trust to fund Herceptin treatment for her breast cancer. The Court of Appeal considered that the Defendant's exceptional circumstances policy was irrational because it was not possible to envisage any circumstances in which the exceptional circumstances would apply.
In the present case, given the recommendation of the treatment by the NHS, NICE and the UK National Screening Committee, the reasons for refusal would need to be compelling. Furthermore, given the likely costs savings, a refusal based on a Trust tolling the bell of ‘resources’, might well prove to be, on analysis, Wednesbury irrational.
In light of the above, it is possible that reasons other than resources would be relied upon. Two suggest themselves. First, that the reason that the combined test is not routinely commissioned is not cost, but rather the absence of trained staff to carry out the test on the NHS. Such argument might be countered by the suggestion that the treatment could be obtained privately at a relatively modest cost (£270) and that the trust should pay for this. Second, the trust may argue that allowing the costs privately for one woman would be unfair on other women in the area who would have one of the other tests, and that private testing for all patients was not affordable. Again, the rationality of such a response would have to be carefully scrutinized, having regard to the clear recommendation in favour of the treatment and the likely cost savings arising from its routine use.
It must be said though, that having regard to the relatively modest cost of private testing, the likely result of any request for combined testing by an individual – unless the trust wanted to take a stand on a point of principle – is that the test may be paid for privately, rather than incurring the costs of resisting a judicial review. The reality is that if any such case got as far as obtaining the grant of permission for judicial review it would never reach a substantive hearing. Thus, an individual challenge would be limited in effecting change for the wider community.
A challenge with a likely wider effect could be envisaged against the Department of Health for its failure to ensure the timely and widespread availability of the combined test and failure to give adequate guidance to trusts in relation to making the combined test generally available – even if trusts pay for tests privately for a temporary period while NHS staff are suitably trained and software packages introduced, et cetera. The action would be brought by a pressure group, which would be able to make a convincing case as to the irrationality or discriminatory effect of the absence of effective guidance in respect to the combined test. The discriminatory effect here referred to is the ‘postcode lottery’ inherent in the NHS commissioning system.
Guidance issued by the Department of Health can be expected to be followed unless there are cogent reasons to justify not doing so. The slowness with which the combined test is being introduced suggests that further guidance is necessary. There have been recent examples of challenges to the organs of Government both in respect of the failure to promulgate appropriate guidance and to the content of existing guidance, see for example R(Purdy) v DPP 2009. 21
In summary, there are good reasons for thinking that a challenge to the current unavailability of the combined test might be successful, whether made by an individual or pressure group against a trust or the Secretary of Health not least because of the points 1–6 mentioned above, and one might even add the rights of the unborn child, although English law has thus far refrained from accepting that a fetus has any ‘rights’ as such – see for example R v St George's Healthcare Trust ex p. S. 22
Conclusion
It is unacceptable that babies are dying unnecessarily as a result of the failure to introduce combined testing for Down's syndrome, and that women are not being given the test most likely to identify whether they are at risk of having a Down's baby. In addition, it seems that women are not being given the information about combined testing to enable them to give informed consent, which is putting trusts at risk of litigation and complaint. It is clear that the most likely action to succeed at present for failing to introduce the recommended test would be in public law. However, a woman who has lost her baby as a result of being given a false-positive result or has a Down's syndrome baby following a false negative, and was given only a blood or scan test, will be left angry and disillusioned that she is unlikely to be able to claim common law damages (save possibly in relation to consent) when she hears of the NHS's clear recommendation of the combined test, and the failure to implement it.
Importantly, this is not an argument for the introduction of a costly ‘best available’ test – something above and beyond what can reasonably be expected – simply the introduction of the test that the NHS has itself recommended. It is hoped that the test is readily available by the end of 2010.