Abstract
The implementation of a uniform service for Down's syndrome screening in England has been ongoing since it was first set out in 2001 that all women would be offered screening. The main aim has been to improve the detection rate while retaining a low screen positive rate (Spr) using a combined screening strategy. It has been a major challenge to ensure that the measurement of the nuchal translucency using ultrasound is performed well in each hospital and that sonographers are supported in meeting that challenge. The service is now beginning to meet the national policy that all units must offer combined screening and that it reaches a 90% detection rate for a 2% Spr.
Down's syndrome screening has been undertaken in this country since the mid-1980s following the first publication of markers for Down's syndrome in 1984. 1 Primarily the first set of biochemical analytes used were alphafetoprotein and human chorionic gonadotrophin commonly known as the Double test. 2 The initial provision and development of Down's syndrome screening services evolved in response to patient demand and changing clinical practice, and was driven mainly by the advertising of private testing facilities. As demand grew from pregnant women to access private testing services, as the test was not commonly available in the NHS, little acknowledgement or thought was given to how women would or could access the follow-on diagnostic test necessary to complete the clinical pathway. Diagnostic testing was not commonly available in the private service, and therefore required women to access NHS services. The introduction of unconjugated estriol in the late 1980s to form the Triple test provided a slight improvement to detection rates. The early 1990s began to see the first published studies on the sonographic measurement of the nuchal translucency early in pregnancy, which was a competing rival for the biochemical test. 3 At this time there was no national guidance; however, it was becoming clear that this demand was growing and beginning to impact on NHS services at all levels of the pathway. 4
A survey undertaken by the UK National Screening Committee (UK NSC) in 2001 5 reported that there was extensive provision of antenatal screening services for Down's syndrome in the UK, but that the nature of the service varied, with some units offering first trimester screening and others second trimester. The aim was to identify the foundations for an infrastructure that could support a coordinated approach, and have the capacity to identify and address any concerns within the NHS. The goal was to develop a comprehensive coordinated service that could comply with performance management standards and a programme of quality assurance.
In April 2001, the Minister for Public Health announced the establishment of an antenatal screening service for all pregnant women by 2004. This was followed in October 2001, with a statement in The Chief Medical Officer's Update, Number 31, that initial standards with which the antenatal screening service for Down's syndrome should comply and that:
The UK National Screening Committee (UKNSC) has recommended that all pregnant women, irrespective of age, should be offered second trimester serum screening. The target performance at this point was that the test should reach a 60% detection rate (DR) for a 5% screen positive rate (Spr) using a cut off level of 1 in 250.
The Serum, Urine and Ultrasound Screening Study (SURUSS) report commissioned by the National Institute for Health Research Health Technology Assessment Programme (NIHR HTA) in 2003 6 provided some evidence as to the performance of the screening strategies and formed the main basis of the first national policy set out in 2003, which followed the UK NSC recommendation. Responsibility of implementation and performance monitoring of the policy objectives were given to the NHS Fetal Anomaly Screening Programme (NHS FASP). Policy review is a triennial process and subsequent reviews expected and set out that the performance target would be raised. Table 1 shows expected national targets. The performance target of a 90% detection rate for a 2% screen positive rate (Spr) will remain until at least 2014. 7 The only change that will take place is that the cut-off level of one in 150 will be set for all screening strategies through both trimester tests. This should be implemented from 1 October 2011.
The national expected performance target of the test and the year it was to be achieved by
Spr, screen positive rate; Dr, detection rate
It was agreed that the main screening strategy to achieve this goal was to be the combined screening test which encompasses the ultrasound measurement of NT and biochemical analysis of the pregnant woman's blood between 11 weeks and two days and 14 weeks and one day gestation. 8 Integrated testing would also reach the expected performance target but requires a two step process in which the pregnant woman has to attend for two appointments. Therefore, in practical terms it is a much more complicated strategy to implement in an NHS setting and women are less likely to comply. In 2001, the main method of screening was using double testing in the second trimester, that is, between 15 weeks + 0 days and 20 weeks + 0 days gestation, and access to these services was variable. Working with frontline service staff, commissioners and Strategic Health Authorities there is now a 95% implementation of combined screening in the English service. Extensive dialogue is taking place with those hospitals who have not implemented to achieve full coverage. The figures for April 2011 can be seen in Table 2.
The number of hospitals that have implemented the combined screening test in England by April 2011
SHA, Strategic Health Authorities
Although a screening programme encompasses many aspects and is holistic in its terms of implementation and management, one of the key components is the performance of the test and measurement against the set target. Setting the performance target of the combined test at a 90% detection rate for a 2% Spr is an ambitious target as no published studies demonstrate that it meets this level. However, extensive analysis of the national data using our quality assurance framework the Down's syndrome screening Quality Assurance Support Service (DQASS) did provide confidence that it could be met with cohesive and comprehensive work on improvements to all areas of the components of the screening test. This most recently includes improvements to the accuracy of measuring NT and crown rump length (CRL). 9 The consensus in the most recent review of policy, when a number of changes to the service had been put into place already, was that we would not advocate the assessment of the presence or absence of the nasal bone at this point as it can be difficult to assess in every situation and requires extra training.
To address how we would know we were meeting the 2010 target a number of measures have been put into place, the first being the need to assess the performance of the test on a national basis. Some of the previous difficulties of assessing test performance in published articles were due to dissymmetry in the use of cut-off levels and age standardization. Quite often these aspects were not published and very rarely commented upon. These and a number of other variables could impact on the ability to compare like for like. In October 2006, the NHS FASP set up DQASS specifically to monitor and improve the performance of the screening test. All laboratories are required to submit their screening data on a six monthly cycle basis. There was still a variety of screening strategies in place when this commenced but it has provided the first extensive actual screening data on test performance comparing all using the same criteria. Table 3 shows the Sprs for England and performance of the testing strategies.
The screen positive rates for each testing strategy presented by six monthly DQASS cycles
DQASS, Down's syndrome screening Quality Assurance Support Service
Note: From cycle 6, the risk cut-off policy was changed from one in 250 for all tests to one in 150 for first trimester tests and one in 200 for second trimester tests
In relation to ultrasound measurements the Fetal Medicine Foundation (FMF) has a long history with developing NT as a screening test and has worked hard to monitor and improve the measurement of NT. The NHS FASP acknowledges this. FMF guidance has been accessed mostly by those who wish to provide private ultrasound screening services, which are now regulated by the Care Quality Commission. 10 However, it is understood that accurate measurement of the CRL component is imperative if the performance target is to be met, as a bias away from the set reference curve on both NT and CRL measurements can have a detrimental impact on the final risk provided to the pregnant woman. To support improvements a quality assurance system has been set up to monitor all sonographers and clinicians who perform the NT and CRL measurement for screening for Down's syndrome. A practical training framework has also been implemented with 10 Regional Obstetric Support Coordinators appointed in November 2010.
Sonographers' measurements are sent to DQASS along with the laboratory data and assessed against the FMF reference curve. Depending on their bias from that curve they are awarded a green, amber or red flag status. Those who are awarded a red flag will require retraining and supervision until an amber status is achieved.
It is this continual drive for improvements and the ability to analyse national data that provide knowledge of where changes need to be made. This knowledge and understanding can be fed back to the individual sonographers and the laboratory so they can benefit from it and provide a high quality screening service.
Conversely monitoring of the detection rate has been difficult to achieve as outcome data are not collected nationally at NHS FASP and this has been a continual objective of the programme. The detection rate is therefore a modelled rate based on the performance of the screening data and using evidenced parameter sets such as those provided in the HTA 2003 SURUSS report. 6 This is modelled to be an 85% detection rate. Anecdotal evidence from those who do measure their outcomes on a regional basis correlates with this. Data from the National Downs Syndrome Cytogenetic Register show that detection rates have steadily increased although these data are relevant only to 2008. 11 The number of diagnostic procedures has fallen substantially reflecting the decrease in the Spr in England which has been achieved through implementation of the combined screening test and careful monitoring of the programme with effective improvements put into place. In 2003, there were 36,968 invasive procedures. In 2010, this had reduced to 13,595. 9 Further information is available from the Association of Clinical Cytogenetics. 12
Conclusion
Since 2001, when the first statement was issued by the Minster for Health and a survey undertaken to establish the service provision, much has been achieved. Screening is a comprehensive and all inclusive pathway which requires a multidisciplinary approach. The main aim is always to provide choice to the woman about her pregnancy but the major point is that this has to be provided by a national screening programme which is as safe as possible. Achieving a high detection rate yet keeping the need for invasive procedures as low as possible is the minimum a pregnant woman can expect. Ultrasound has a major part to play in that.
DECLARATIONS
The authors have no conflicts of interest to declare.