Prenatal screening for Down syndrome in England and Wales and population-based birth outcomes

Transcription

1 Prenatal screening for Down syndrome in England and Wales and population-based birth outcomes Rebecca Smith-Bindman, MD, a,b Philip Chu, MS, a Peter Bacchetti, PhD, b Jonathan J. Waters, PhD, MRC Path, c David Mutton, BSc, d and Eva Alberman, MD d San Francisco, Calif, and Birmingham and London, United Kingdom OBJECTIVE: Whether the introduction of antenatal screening for Down syndrome in England and Wales with serum biochemistry or ultrasound has led to improvements in patient outcomes is unknown. The purpose of this study was to relate pregnancy outcomes to the dominant method used for prenatal Down syndrome screening. STUDY DESIGN: For the years 1989 through 1999, England and Wales were divided into geographically defined areas where specific hospitals, health authorities, and cytogenetic laboratories provided maternity care for welldefined populations. For each year from 1989 through 1999, the dominant Down syndrome screening method that was used in each area was determined. Outcomes for area-years that used serum biochemistry or ultrasound (first or second trimester) were compared with area-years that used advanced maternal age as the dominant screening method. The percent of Down syndrome cases that were diagnosed prenatally (effectiveness) and the number of invasive prenatal tests that were performed to diagnose each Down syndrome case prenatally (efficiency) were compared. RESULTS: There were 5,980,519 births and 335,184 referrals for prenatal karyotyping (amniocentesis and chorionic villus sampling) that occurred in the area-years studied, of which 12,047 pregnancies were diagnosed as Down syndrome; 5393 cases of Down syndrome (45%) were diagnosed prenatally. Invasive testing increased from 4.4% of pregnancies in 1989 to 6.4% in 1997 and declined slightly in 1999 (5.8%). Prenatal diagnosis of Down syndrome cases rose from 28% in 1989 to 53% in 1999, and the number of invasive tests that were performed to diagnose each Down syndrome case fell from 89.7 to 47.7 (P [for trend] <.0001). Areas with serum or ultrasound as the dominant screening method detected 50% more Down syndrome cases in prenatally (52% and 53% vs 36%; P <.0001) and performed fewer invasive procedures to diagnose each Down syndrome case (60.7 and 52.0 vs 88.0; P <.0001) compared with areas in which advanced maternal age screening was dominant, despite serving populations with similar mean/median maternal ages. CONCLUSION: In clinical practice, screening programs for Down syndrome that were based on maternal serum biochemistry or ultrasound were more effective and efficient than the screening programs that used advanced maternal age alone. ( 2003;189:980-5.) Key words: Prenatal screening, Down syndrome, birth outcome, serum biochemistry, ultrasound examination Prenatal detection of Down syndrome fetuses became possible after karyotyping of fetal cells that were removed by amniocentesis or chorionic villus sampling (CVS) became available in the mid 1970s. Because amniocentesis and CVS are invasive diagnostic procedures that are associated with a 0.5% to 1.0% fetal loss rate 1-4 and because From the Departments of Radiology a and Epidemiology and Biostatistics, b University of California, Wolfson EQA Laboratory, Institute of Research and Development, University of Birmingham Hospital NHS Trust, c and the Department of Environmental and Preventive Medicine, Wolfson Institute of Preventive Medicine, Barts and the London Queen Mary s College School of Medicine and Dentistry, University of London. d Presented at the Twenty-Third Annual Meeting of the Society for Maternal Fetal Medicine, San Francisco, California, February 3-8, Reprint requests: Rebecca Smith-Bindman, MD, Department of Radiology, University of California, San Francisco, 1600 Divisadero St, San Francisco, CA Rebecca.Smith-Bindman@Radiology.UCSF.Edu Ó 2003, Mosby, Inc. All rights reserved /2003 $ doi: /s (03)00721-x Down syndrome is rare, screening methods have been used to identify women who are at sufficiently elevated risk to justify invasive prenatal testing. Initially this risk was based on maternal age alone, and invasive prenatal testing was recommended generally for women older than 35 to 37 years. Although the risk of Down syndrome is higher among older women, most pregnancies occur in women younger than 35 years, 5 and most Down syndrome cases were missed when this was the only screening method used. 6,7 Subsequently, multiple first and second trimester fetal ultrasound and maternal serum biochemical markers have been found to be associated with Down syndrome Their use, in conjunction with maternal age, provides higher detection rates, at lower invasive testing rates, than the use of age alone. Such screening has become a routine part of prenatal care in both the United States and the United Kingdom Whether the introduction of population screening with biochemistry and ultrasound examination has led to the 980

2 Volume 189, Number 4 Smith-Bindman et al 981 potential benefits that are based on expected screening performance is unknown. The willingness of women to participate in different types of screening programs may depend on the gestational age when screening is offered, how long it takes to report results, and sociodemographic factors. Additionally, screening tests may not perform as well in clinical as compared with research settings, offers for screening may not be accepted, and women with screenpositive results may choose to forego invasive prenatal testing. Only a few relatively small studies have evaluated the outcomes of prenatal screening in population settings, and in clinical practice the advantage of the use of serum and ultrasound screening has been questioned. 16 The present study aimed to determine whether there were variations in population outcomes that were related to Down syndrome screening over time in England and Wales and, if so, whether these were associated with the types of screening methods that were used. Methods For the years 1989 through1999, with the use of information that was provided by the Office of National Statistics and the National Down Syndrome Cytogenetic Register, 23 a map was created that linked hospitals, health authorities, and cytogenetic laboratories in England and Wales for each year of the study. With this map, England and Wales were divided into 26 geographic areas in which maternity care was provided to well-defined, nonoverlapping populations and in which a single laboratory performed most cytogenetic testing. The mean annual number of births (including births, reported terminations, and stillbirths) per area was 22,826, and the combination of geographic area and year became the unit of analysis (area-years) for all statistical comparisons. We excluded 1 major cytogenetic laboratory, because this was a private laboratory without a defined catchment area and included patients who were not UK residents. Within area-years, we linked information on Down syndrome screening methods and population-based birth outcomes from 3 sources: (1) annual reports on all invasive prenatal tests that were performed by each cytogenetic laboratory (amniocentesis and CVS) produced by the UK National External Quality Assessment Scheme, 24 (2) preand postnatally diagnosed Down syndrome cases that were collected by the National Down Syndrome Cytogenetic Register, 23 which includes the screening method that led to prenatal diagnosis, and (3) annual maternal agespecific population births for each health authority, which was provided by the Office of National Statistics. UK national external quality assessment scheme (UKNEQAS). To satisfy proficiency requirements, all clinical cytogenetic laboratories in the United Kingdom are required to participate in an external quality assessment program. As part of an annual audit, UKNEQAS 24 collects information on the total number of amniocentesis and CVS specimens that are processed annually by each laboratory. Data from the UKNEQAS for areas outside England and Wales were excluded. National Down syndrome cytogenetic register (NDSCR). Since its inception in 1989, NDSCR has received reports from all cytogenetic laboratories in England and Wales on all pre- and postnatally diagnosed cases of Down syndrome, and audits have found that the register is >94% complete. 23 For each case, a laboratory identification number, maternal age, date and outcome of pregnancy, method of screening and prenatal testing, gestational age at diagnosis, and fetal karyotype are obtained. Information on these registered cases was aggregated with information on all prenatal diagnostic tests that were reported by the individual cytogenetic laboratories and information from the Office of National Statistics on the number of pregnancies. We then calculated, for each area and each year of the study, the annual number of Down syndrome cases that had been diagnosed, the proportion of Down syndrome cases that had been detected prenatally, and the mean/median maternal age. Method of prenatal screening. UKNEQAS data do not include information on the screening method that led to the referral for invasive testing. Instead we used the screening method that led to the prenatal diagnosis for Down syndrome cases as a proxy measure for referral screening methods. The indication for referral was classified as serum (any combination of serum markers 15 ), first trimester ultrasound (ultrasound examination at <14 weeks of gestation that reflected nuchal translucency screening), second trimester ultrasound (ultrasound examination at #14-24 weeks of gestation that reflected fetal anomaly screening), third trimester ultrasound, advanced maternal age ($35 years), family history, and all other indications combined. Using this inferred screening method, we then classified each area-year on the basis of its dominant screening method as either (1) serum, (2) ultrasound, (3) advanced maternal age, or (4) mixed (areas without a dominant method). We combined first and second trimester ultrasound because both methods are used for screening yet neither indication accounted for the dominant method in any area-year. The dominant indication was defined as the method that accounted for more Down syndrome positive referrals than all other indications combined. In area-years in which serum was dominant, serum accounted for 58%, ultrasound 17%, and advanced maternal age 19% of case referrals. In area-years in which ultrasound was dominant, serum accounted for 17%, ultrasound 57% of case referrals, and advanced maternal age for 16% of case referrals. In area-years in which advanced maternal age was dominant, serum accounted for 13%, ultrasound accounted for 12%, and advanced maternal age for 69% of case referrals. Lastly, in area-years that were characterized as mixed, serum accounted for 34%, ultrasound 29%, and advanced

3 982 Smith-Bindman et al October 2003 Table I. Pregnancies, invasive diagnostic tests (amniocentesis and chorionic villus sampling), percent of pregnancies undergoing an invasive test, Down syndrome cases and percent prenatal diagnoses, and invasive tests per Down syndrome cases prenatally diagnosed in England and Wales, Invasive diagnostic tests % Pregnancies Down Prental undergoing syndrome diagnoses Invasive tests/ Year Pregnancies* Total Amniocenteses CVS invasive test cases yz (No. [%]) case diagnosed y ,266 25,649 22,586 3, , (28.4) ,130 28,957 25,948 3, , (33.2) ,099 28,732 25,648 3, , (37.2) ,468 29,492 26,709 2, , (43.1) ,547 31,955 29,063 2, , (46.4) ,482 32,072 28,874 3, , (46.3) ,903 32,469 29,365 3, , (50.0) ,926 32,281 28,604 3, , (52.5) ,402 32,673 28,486 4, , (49.4) ,929 31,831 27,446 4, , (49.5) ,367 29,073 24,786 4, , (53.4) 47.7 Total 5,980, , ,515 37, ,047 5,393 (44.8) 62.2 *Pregnancies include livebirths, reported terminations, and still births. ytrends significant, P values < zthe increase in Down syndrome cases is a result of the increase in maternal age over the years studied as well as the use of prenatal testing. 23 Table II. Indication for referral for invasive prenatal testing for Down syndrome cases prenatally diagnosed in England and Wales, Indication for referral for prenatally diagnosed DS cases Year Prenatal diagnoses Serum (No. [%])* AMA (No. [%])* Ultrasound (No. [%])* y Third-trimester ultrasound (No. [%])* Family history (No. [%])* (6) 220 (77) 31 (11) 9 (3) 9 (3) (14) 234 (67) 39 (11) 12 (3) 12 (3) (21) 212 (53) 64 (16) 22 (6) 12 (3) (37) 178 (39) 92 (20) 10 (2) 5 (1) (40) 166 (33) 105 (21) 13 (3) 12 (2) (45) 147 (29) 98 (19) 12 (2) 12 (2) (44) 139 (26) 126 (24) 8 (2) 18 (3) (42) 150 (25) 155 (26) 14 (2) 21 (4) (46) 107 (18) 185 (31) 14 (2) 11 (2) (37) 126 (22) 202 (36) 13 (2) 9 (2) (39) 114 (19) 214 (35) 21 (3) 12 (2) Total 5,393 1,957 (36) 1,793 (33) 1,311 (24) 148 (3) 133 (2) Other indications or no indication was provided for the remaining 51 Down syndrome cases. AMA, Advanced maternal age. *Percentages are based on Down syndrome cases prenatally diagnosed. yincludes first- and second-trimester ultrasound. maternal age 30% of case referrals. Mixed area-years were not included because no screening method was dominant. To test the accuracy of our proxy measure for the determination of the dominant screening method in each area-year, we compared the actual indications for all referrals with those inferred from Down syndrome cases in 2 geographic areas over 10 years, which provided individual patient data for 34,169 invasive prenatal tests, which were derived from 685,167 pregnancies that included 582 prenatally diagnosed pregnancies as Down syndrome. In these 2 areas, the actual and inferred screening methods were highly correlated (Pearson rank correlation between the percentage inferred and actual referrals for serum [r = 0.70], ultrasound [r = 0.85], and advanced maternal age [r = 0.77]). Additionally, the inferred screening method correctly classified the dominant screening method for 19 of 20 area-years. In the single area-year in which the dominant method was misclassified, the inferred screening method characterized the area-year as serum, whereas the dominant screening method was actually advanced maternal age. Analysis. We calculated the number of Down syndrome cases that were diagnosed annually and the percentage of cases that were diagnosed prenatally. We also calculated the number of invasive prenatal tests (amniocentesis and CVS) that were performed, the percentage of pregnancies that underwent a prenatal test, and the number of prenatal tests per prenatal diagnosis (efficiency). We calculated these outcomes for each area each year and stratified the area-years by the dominant method of screening. Because we generally found no significant changes in the outcomes over time within each screening category, we pooled the

4 Volume 189, Number 4 Smith-Bindman et al 983 Table III. Prenatal Down syndrome detection rates and invasive prenatal tests (amniocentesis or chorionic villus sampling) per Down syndrome case prenatally diagnosed, stratified by the dominant method of prenatal screening Prenatal detection (%) Invasive tests/case diagnosed Year Serum* Ultrasound* AMA* Serum* Ultrasound* AMA y Total (95% CI) z 52.1 (49-55) 52.9 (45-61) 35.8 (33-39) 60.7 ( ) 52.0 ( ) 88.0 ( ) Numbers are not provided if no area had that method as dominant in that year. AMA, Advanced maternal age. *Within group trends over time not significant, P value >.10. yspearman rank correlation 78, P = zdifferences between groups by ANOVA significant for prenatal detection and invasive tests/case diagnosed, P < results across all area-years within each category. For example, we calculated the Down syndrome detection rate across all area-years in which the dominant method of screening was serum. We used analysis of variance to compare outcomes between screening categories. We compared the mean and median maternal ages between serum, ultrasound examination, and advanced maternal age areas each year. For each area-year, we plotted detection against efficiency, and each laboratory each year contributed a single point to the graph. Results There were 5,980,519 pregnant women included in this analysis, in whom 335,184 invasive prenatal tests were performed and 12,047 Down syndrome cases were diagnosed (Table I). The percent of pregnant women who underwent invasive prenatal testing each year increased from 4.4% of pregnancies in 1989 to 6.4% of pregnancies in 1997 and declined slightly in 1999 (5.8%). Between 1989 and 1999 the proportion of all Down syndrome cases that were diagnosed prenatally rose from 28% to 53%, and the number of invasive tests that were performed to diagnose each Down syndrome case fell from 89.7 to 47.7 (P [value for trends] <.0001; Table I). Over the 11 years of the study, 5393 of the 12,047 Down syndrome cases (45%) were diagnosed prenatally (Tables I and II). The use of serum and ultrasound increased, and the use of advanced maternal age decreased, over time (Table II). Nonetheless, even in 1999 overall, 19% of the prenatally diagnosed cases (114/609) were referred because of advanced maternal age screening. Areas with serum or ultrasound as their dominant screening method detected 50% more Down syndrome cases in their areas prenatally (52% and 53% vs 36%, respectively; P <.0001) and performed fewer invasive procedures to diagnose each Down syndrome case (60.7 and 52.0 vs 88.0, respectively; P <.0001) compared with areas where advanced maternal age screening was dominant, despite serving populations with similar mean/median maternal ages (Table III). The difference between serum and ultrasound areas in the number of invasive tests per prenatal Down syndrome diagnosis was not statistically significant. The graph of the effectiveness (prenatal detection rate) versus efficiency (invasive tests per prenatal diagnosis) demonstrates that serum and ultrasound area-years tended to have both better detection and improved efficiency compared with advanced maternal age area-years (Fig 1). Comment Most studies of Down syndrome screening methods have evaluated prenatal detection among women who chose to undergo the method under study or have used statistical modeling to estimate the impact of screening on a larger population. 7,8,25 Such studies will tend to overestimate the impact that would occur in clinical practice, because they typically assume 100% uptake of screening and invasive prenatal testing and ideal performance of the test under study. The best way to compare different screening methods is to evaluate outcomes within a large population, which will take into account both the accuracy of the test and the choice of women to accept, reject, or demand invasive testing. Our simple analysis of the performance of Down syndrome screening among nearly all pregnancies in England and Wales for an 11-year period confirms Down syndrome modeling results and demonstrates that, in actual clinical practice, serum biochemistry and/or ultrasound are substantially more effective and efficient than advanced maternal age screening. Over the years of the study, this translated into 50% more Down syndrome cases being diagnosed before birth, at 30% lower rates of invasive prenatal testing, in areas that primarily used serum or ultrasound as opposed to advanced maternal age

5 984 Smith-Bindman et al October 2003 Fig. 1. The percentage of Down syndrome cases that were detected prenatally (effectiveness) versus the number of invasive tests that were performed per Down syndrome case that was diagnosed prenatally (efficiency). Each geographic area is shown once each year, and the symbol that is used to illustrate that area reflects its dominant method of prenatal screening in use at that time (advanced maternal age, open circles; serum, closed circles; ultrasound, open triangles). screening. The improvements in population outcomes over the years of this study reflect the increase over time in the use of serum and ultrasound in England and Wales. 15 Only a few relatively small studies have evaluated the outcomes of Down syndrome screening in population settings, and most identified decreasing rates of Down syndrome births that were temporally related to the introduction of serum and/or ultrasound. Jorgensen et al 22 and Cunningham and Thompkinison 26 demonstrated the feasibility of large scale, population-based antenatal screening programs. However, none of these existing studies have linked population outcomes specifically to variations in the use of specific screening strategies. Wellesley et al 16 concluded that there was no advantage of an antenatal screening program based on serum biochemistry or first trimester ultrasound compared with advanced maternal age alone. However, they focused on one geographic area and included too few cases of Down syndrome to detect clinically significant differences between the different screening methods. The current analysis has several limitations. NEQAS estimates of the number of invasive prenatal tests include duplicates and some tests for other conditions, but these are likely a very small percentage of the total. Our proxy measure assumes that the patterns of screening for diagnosed cases are a reliable reflection of referral patterns; the analysis in 2 geographic areas showed a close correlation between the 2 measures. However, the categorization of area-years by their dominant screening method will tend to underestimate the differences between the screening methods, because our characterization of screening patterns based only on Down syndrome cases almost certainly overestimated the use of serum and ultrasound screening. Additionally, each area-year includes a mixture of screening methods (rather than just a single method), which accounts for some of the variability in outcomes within each screening category and might have limited our ability to differentiate between the performance of serum and ultrasound screening. We combined nuchal translucency and second trimester anomaly screening methods, because neither individually accounted for a majority of referrals in any area-year. Additional work is needed to determine which of the screening strategies is most effective. What is clear is that serum and the different methods of ultrasound screening are more efficient and effective than advanced maternal age screening alone. It is important to improve the information received by parents of all ages to maximize their ability to make informed choices and to ensure that the best (most effective and efficient) screening methods are promoted among all women. Our findings show that a policy of offering serum biochemistry or ultrasound screening (first trimester

6 Volume 189, Number 4 Smith-Bindman et al 985 nuchal screening or second trimester anomaly screening) to pregnant women of all ages will lead to substantially better detection rates and a lower number of prenatal invasive tests per case detected than the use of advanced maternal age screening alone. Advanced maternal age screening will miss Down syndrome cases that occur in women younger than age 35 years, whereas serum biochemistry and ultrasound screening will tend to miss cases among women of all ages. However, the widely available triple test will detect approximately 85% of Down syndrome cases in women age $35, 21 and newer, more sensitive methods are becoming available that will allow the detection of an even higher percentage of Down syndrome cases among older women. Individual women may choose to undergo invasive testing for Down syndrome despite a low Down syndrome estimated risk, and their personal choice should be a leading determinant of whether they undergo invasive prenatal testing. However, a general policy of encouraging invasive prenatal testing on the basis of maternal age alone, as suggested by some professional organizations, is not justified. Maternal age should be avoided as the primary method to decide which women should be offered invasive prenatal testing, because in actual clinical practice, as shown here, it is less effective and less efficient than serum and ultrasound screening. If all women were offered serum and/or ultrasound screening, overall fewer cases of Down syndrome would be missed, and fewer invasive prenatal tests would be performed than if advanced maternal age screening alone were used. We thank Travis Seawards for assistance in preparing the manuscript, UKNEQAS and its contributing laboratories for allowing us to include their data in this analysis, the UK Office of National Statistics for making their data available to us electronically to use in this analysis, the Regional Cytogenetics Centre, Southmeads Hospital, Bristol for contributing data for the validation study, and the North and South Thames NHS R & D, which currently fund the NDSCR and the data collection that was made possible by the collaboration of cytogeneticists and referring physicians. REFERENCES 1. Rhoads GG, Jackson LG, Schlesselman SE, de la Cruz FF, Desnick RJ, Golbus MS, et al. The safety and efficacy of chorionic villus sampling for early prenatal diagnosis of cytogenetic abnormalities. N Engl J Med 1989;320: Midtrimester amniocentesis for prenatal diagnosis: safety and accuracy. JAMA 1976;236: Tabor A, Philip J, Madsen M, Bang J, Obel EB, Norgaard-Pedersen B. Randomised controlled trial of genetic amniocentesis in 4606 lowrisk women. Lancet 1986;1: Randomised trial to assess safety and fetal outcome of early and midtrimester amniocentesis: the Canadian Early and Mid-trimester Amniocentesis Trial (CEMAT) Group. Lancet 1998;351: Ventura S, Martin J, Curtin S, Mathews T, Park M. Births: final data for 1998 national vital statistics reports. Hyattsville (Md): National Center for Health Statistics; Adams M, Erickson J, Layde P. Down s syndrome: recent trends in the United States. JAMA 1981: Egan JF, Benn P, Borgida AF, Rodis JF, Campbell WA, Vintzileos AM. Efficacy of screening for fetal Down syndrome in the United States from 1974 to Obstet Gynecol 2000;96: Wald NJ, Kennard A, Hackshaw A, McGuire A. Antenatal screening for Down s syndrome. J Med Screen 1997;4: Haddow JE, Palomaki GE, Knight GJ, Canick JA, Saller DN Jr, Panizza DS. Prenatal screening for Down s syndrome with use of maternal serum markers. N Engl J Med 1992;327: Snijders RJ, Noble P, Sebire N, Souka A, Nicolaides KH. UK multicentre project on assessment of risk of trisomy 21 by maternal age and fetal nuchal-translucency thickness at weeks of gestation: Fetal Medicine Foundation First Trimester Screening Group. Lancet 1998;352: Schuchter K, Hafner E, Stangl G, Ogris E, Philipp K. Sequential screening for trisomy 21 by nuchal translucency measurement in the first trimester and maternal serum biochemistry in the second trimester in a low-risk population. Ultrasound Obstet Gynecol 2001;18: Hafner E, Schuchter K, Philipp K. Screening for chromosomal abnormalities in an unselected population by fetal nuchal translucency. Ultrasound Obstet Gynecol 1995;6: Cuckle H. Integrating antenatal Down s syndrome screening. CurrOpin Obstet Gynecol 2001;13: Palomaki GE, Knight GJ, McCarthy JE, Haddow JE, Donhowe JM. Maternal serum screening for Down syndrome in the United States: a 1995 survey. 1997;176: Wald NJ, Huttly WJ, Hennessy CF. Down s syndrome screening in the UK in 1998 [letter]. Lancet 1999;354: Wellesley D, Boyle T, Barber J, Howe D. Retrospective audit of different antenatal screening policies for Down s syndrome in eight district general hospitals in one health region. BMJ 2002;325: Cheffins T, Chan A, Haan EA, Ranieri E, Ryall RG, Keane RJ, et al. The impact of maternal serum screening on the birth prevalence of Down s syndrome and the use of amniocentesis and chorionic villus sampling in South Australia. BJOG 2000;107: Palomaki G, Haddow J, Beauregard L. Prenatal screening for Down s syndrome in Maine, N Engl J Med 1996;334: Smith NC, Hau C. A six year study of the antenatal detection of fetal abnormality in six Scottish health boards. BJOG 1999;106: Verloes A, Gillerot Y, Van Maldergem L, Schoss R, Herens C, Jamar M, et al. Major decrease in the incidence of trisomy 21 at birth in south Belgium: Mass impact of triple test? Eur J Hum Genet 2001;9: The California Expanded AFP Screening Program: prenatal care provider handbook. Berkeley (CA): California Department of Health Services, Genetic Disease Branch; Jorgensen FS, Valentin L, Salvesen KA, Jorgensen C, Jensen FR, Bang J, et al. MULTISCAN: a Scandinavian multi-center second trimester obstetric ultrasound and serum screening study. Acta Obstet Gynecol Scan 1999;78: Alberman E. The National Down Syndrome Cytogenetic Register. J Med Screen 2002;9: Waters J, Waters K. Trends in cytogenetic prenatal diagnosis in the UK: results from UK NEQAS external audit. Prenat Diagn 1999;19: Wald NJ, Watt HC, Hackshaw AK. Integrated screening for Down s syndrome on the basis of tests performed during the first and second trimesters. N Engl J Med 1999;341: Cunningham GC, Tompkinison DG. Cost and effectiveness of the California triple marker prenatal screening program. Genet Med 1999;1: American College of Medical Genetics Clinical Practice Committee. ACMG position statement on multiple marker screening in women 35 and older. American College of Medical Genetics Newsletter 1994; American College of Obstetricians and Gynecologists Committee on Practice Bulletins Obstetrics. ACOG practice bulletin: clinical management guidelines for obstetrician-gynecologists: prenatal diagnosis of fetal chromosomal abnormalities. Obstet Gynecol 2001;97(suppl): US Department of Health and Human Services, Office of Public Health and Science, Office of Disease Prevention and Health Promotion. Guide to clinical preventive services: report of the US Preventive Services Task Force. Washington (DC): US Government Printing Office; American College of Obstetricians and Gynecologists. Down Syndrome Screening. Washington (DC): The College; Committee Opinion No.: 141.