|
Economic modelling of antenatal screening and ultrasound scanning programmes for identification of fetal abnormalities |
Ritchie K, Bradbury I, Slattery J, Wright D, Iqbal K, Penney G |
|
|
Record Status This is a critical abstract of an economic evaluation that meets the criteria for inclusion on NHS EED. Each abstract contains a brief summary of the methods, the results and conclusions followed by a detailed critical assessment on the reliability of the study and the conclusions drawn. Health technology Strategies of antenatal screening and ultrasound scanning for foetal abnormalities were compared. The structural abnormalities included in the analysis were anencephaly, spina bifida, cardiac defects, renal tract defects, abdominal wall defects and congenital diaphragmatic hernia. The two chromosomal abnormalities included were trisomy 21 (Down's syndrome) and trisomy 18 (Edward's syndrome). A model evaluated six strategies of serum screening and scanning in early pregnancy, including combinations of first- and second-trimester ultrasound screening and first- and second-trimester screening for chromosomal abnormalities. Strategy 1 comprised first-trimester nuchal translucency (NT) scan and screen for trisomy 18 and 21, and second-trimester anomaly scan.
Strategy 2 comprised first-trimester NT scan and screen for trisomy 18 and 21.
Strategy 3 comprised first-trimester NT scan and screen for trisomy 18 and 21, and second-trimester anomaly scan and alpha-fetoprotein for neural tube defects (NTD).
Strategy 4 comprised first-trimester NT scan and screen for trisomy 18 and 21, and second-trimester alpha-fetoprotein for NTD.
Strategy 5 comprised booking scan, second-trimester anomaly scan and double test for trisomy 18, 21 and NTD.
Strategy 6 comprised booking scan, and second-trimester double test for trisomy 18, 21 and NTD.
Strategy 6 was the base-case.
Economic study type Cost-effectiveness analysis.
Study population The study population comprised a hypothetical cohort of 50,000 singleton pregnant women. No other information was provided.
Setting The setting was secondary care. The economic study was conducted in Scotland, UK.
Dates to which data relate The effectiveness and resource use data were collected from studies published between 1991 and 2004. No information was provided on the dates to which the cost data referred.
Source of effectiveness data The effectiveness data were derived from a synthesis of completed studies and experts' opinions.
Modelling A discrete simulation model was constructed in R Language (Version 1.8.0), based on evidence from relevant articles and cost data, to calculate the costs and benefits of the six strategies for 50,000 singleton pregnancies
Outcomes assessed in the review The main outcomes assessed were:
the detection rate of abnormalities;
abnormality prevalence at second trimester;
the number of abnormalities detected in foetuses which would have reached birth;
the number of abnormalities missed; and
the number of iatrogenic losses resulting from invasive tests.
Study designs and other criteria for inclusion in the review It was unclear whether a systematic review of the literature had been undertaken to identify the primary studies. There was limited information on the designs of the primary studies. The reader is referred to Ritchie et al. 2004 for full details of the review of the literature (see 'Other Publications of Related Interest' below for bibliographic details).
Sources searched to identify primary studies Not stated (see Ritchie et al. 2004).
Criteria used to ensure the validity of primary studies Not stated (see Ritchie et al. 2004).
Methods used to judge relevance and validity, and for extracting data Not stated (see Ritchie et al. 2004).
Number of primary studies included The effectiveness data were derived from 35 primary studies.
Methods of combining primary studies Not stated (see Ritchie et al. 2004).
Investigation of differences between primary studies Not stated (see Ritchie et al. 2004).
Results of the review The prevalence at second trimester per 10,000 was:
for anencephaly, 8.1 (at first trimester) and 8.1 (at second trimester);
for spina bifida, 8.8;
for cardiac defects, 21.0 (corrected for joint incidence with trisomy 21);
for renal tract defects, 2.8;
for abdominal wall defects, 5.8;
for congenital diaphragmatic hernia, 4.0;
for trisomy 21, 16.9; and
for trisomy 18, 3.8.
The ultrasound detection rate was:
for anencephaly, 0.86 (at first trimester) and 1.00 (at second trimester);
for spina bifida, 0.8;
for cardiac defects, 0.29;
for renal tract defects, 0.61;
for abdominal wall defects, 0.77;
for congenital diaphragmatic hernia, 0.69;
for trisomy 21 at second trimester ultrasound, 45% at a 4.9% false-positive rate;
for trisomy 18 at first trimester, 90% with a false-positive rate of 0.5%.
The detection rate for cardiac defects using raised NT and normal karyotype was 50% at a 5% false-positive rate.
Second-trimester serum screening for NTD using alpha-fetoprotein was 77% at a 3% false-positive rate.
The number of abnormalities detected in foetuses that would have reached birth was 176.0 with strategy 1, 71.3 with strategy 2, 180.6 with strategy 3, 92.0 with strategy 4, 176.2 with strategy 5 and 87.1 with strategy 6.
The number of abnormalities missed per 50,000 (live born) was 95.9 with strategy 1, 200.6 with strategy 2, 91.3 with strategy 3, 179.9 with strategy 4, 95.7 with strategy 5 and 184.8 with strategy 6.
The number of iatrogenic losses per 50,000 was 7.9 with strategy 1, 7.9 with strategy 2, 7.9 with strategy 3, 7.9 with strategy 4, 15.8 with strategy 5 and 15.8 with strategy 6.
Methods used to derive estimates of effectiveness An expert provided an estimate of the second-trimester detection rate for trisomy 18. Authors' assumptions were the basis of the detection rate for trisomy 21 at second trimester.
Estimates of effectiveness and key assumptions The second-trimester detection rate for trisomy 18 at a false-positive rate of 0.5% was 85%. The detection rate for trisomy 21 using soft markers at 20 weeks was 75% with false-positive rates of 2.5% at first trimester and 6.8% at second trimester.
Measure of benefits used in the economic analysis The health benefit measure used was the number of abnormalities detected.
Direct costs The resource quantities and the costs were reported separately. The authors stated that the cost boundary adopted was that of the health care provider at the point of delivery. The direct costs included in the analysis were the costs of informed consent for the different types of tests, the costs of the different scans, serum tests and follow-up tests, and outcomes including spontaneous miscarriage and elective termination. Two sets of costs were calculated, one for an urban setting and one for a rural setting. These reflected differences in the overhead costs, cost of capital equipment, maintenance of ultrasound machines and costs of inpatient stays. The average costs were reported. Discounting does not appear to have been relevant since only the short-term costs were considered in the analysis. The source of the data was not explicitly reported and no price year was given. The authors stated that full details of the methodology used to determine the costs would be published separately.
Statistical analysis of costs The costs appear to have been treated deterministically.
Indirect Costs The indirect costs were not included in the economic analysis.
Sensitivity analysis One-way sensitivity analyses were undertaken to explore uncertainty in the model variables. The detection rate of trisomy 21 at 20-week ultrasound, the timing of the first trimester screening, and the proportion of women who request chorionic villus sampling (CVS) as the diagnostic test for chromosomal abnormality compared with amniocentesis, were all varied. No justification was given for how and why the new variables were chosen.
Estimated benefits used in the economic analysis The total number of abnormalities detected was 184.8 with strategy 6 (base-case), 328.5 with strategy 1, 202.0 with strategy 2, 292.with strategy 5, 227.2 with strategy 4 and 334.1 with strategy 3.
Cost results The total cost of screening and consequence was 5.215,386 with strategy 6 (base-case), 7,525,441 with strategy 1, 5,368,935 with strategy 2, 7,351,575 with strategy 5, 6,212,431 with strategy 4 and 8,358,632 with strategy 3.
Synthesis of costs and benefits An incremental cost-effectiveness ratio (ICER; i.e. the cost per additional anomaly detected) was calculated to combine the costs and benefits of the screening strategies. The ICERs were as follows:
strategy 6, reference value;
strategy 1, 4,790;
strategy 2, 8,927;
strategy 5, 17,525;
strategy 4, 33,472; and
strategy 3:, 148,784.
The sensitivity analysis showed that the effectiveness of the screening strategies was sensitive to variations in the detection rate, timing of screening and use of appropriate risk model. It was also sensitive to the proportion of women who requested CVS as the diagnostic test for chromosomal abnormality versus amniocentesis. However, there was no information as to how these variations would impact upon the ICER.
Authors' conclusions The preferred strategy included both first- and second-trimester ultrasound scans and a first-trimester screening test for chromosomal abnormalities.
CRD COMMENTARY - Selection of comparators The choice of the comparator was clear. Strategy 6 reflected current practice in the authors' setting. The reader should consider whether this reflects current practice in his or her own setting.
Validity of estimate of measure of effectiveness The effectiveness data were derived from a combination of a review of the literature and expert opinion. The two sources of evidence were quite mixed up. The authors did not provide information on the methodology of the review, referring the reader to another publication instead (Ritchie et al. 2004). It was unclear how the estimates of effectiveness from the primary studies were combined, and whether the authors considered the impact of differences between the primary studies when estimating effectiveness. It appears that the authors have used data from the available studies selectively, according to the amount of information provided within the primary studies. Expert opinion also provided the second-trimester detection rate for trisomy 18, although the authors did not describe the process used to select the physician. Authors' assumptions about the detection rate of trisomy 21 at second trimester were justified by referring to NICE guidelines. The latter was investigated in a sensitivity analysis, although no justification was given for the range of variables used.
Validity of estimate of measure of benefit The summary benefit measure was derived from the model and was specific to the intervention and patient group. In terms of other more generic benefit measures that could have been used, the authors noted the difficulties associated with assigning utilities to the information resulting from antenatal screening and diagnostic tests.
Validity of estimate of costs Limited information on the methods and conduct of the cost analysis was provided because the authors noted that more detailed information would be provided in a subsequent report. The authors stated that the cost boundary adopted was that of the health care provider at the point of delivery. As such, all the costs relevant to this perspective appear to have been included in the analysis. The authors also noted that to have included societal costs, which would have been considerable, would have resulted in those strategies that identify the greatest number of abnormalities appearing to be the most cost-effective.
The resource and cost data were reported separately. The resource data were based on expert opinion. Clinical directors of the 22 consultant-led maternity units in Scotland were surveyed to determine the acceptance rates for screening and follow-up tests. The authors acknowledged that while some responses were based on locally collected data, others were estimates. The use of authors' assumptions to derive the rate of women requesting CVS rather than amniocentesis as the diagnostic test for chromosomal abnormality was justified on the grounds of a lack of published data. This assumption was investigated in a sensitivity analysis, although the authors did not justify the range of variables used. No information on the price year or the source of the prices used was given.
Other issues The authors did not make extensive comparisons of their findings with those from other studies. The issue of generalisability was addressed in relation to urban and rural settings, as the authors provided different costings for the different settings. The conditions included in the model were self-selected by the authors based on the availability of data. In addition, only conditions considered serious enough to warrant the offer of a termination were included. The authors noted that other alternatives for identifying chromosomal abnormalities could have been included, but justified their absence from the model since none are routinely offered in the authors' setting.
Implications of the study The authors recommended that two ultrasound scans should be offered to all women in Scotland: one in the first trimester, which should include NT measurement for risk assessment of chromosomal abnormalities, and one in the second trimester for the identification of structural abnormalities. The authors noted that this programme should be provided in a way that does not make scanning appear mandatory. In addition, it should allow sufficient time and information for women to make informed decisions in the process of their antenatal care.
Bibliographic details Ritchie K, Bradbury I, Slattery J, Wright D, Iqbal K, Penney G. Economic modelling of antenatal screening and ultrasound scanning programmes for identification of fetal abnormalities. BJOG. An International Journal of Obstetrics and Gynaecology 2005; 112: 866-874 Other publications of related interest Ritchie K, Boynton J, Bradbury I, et al Routine ultrasound scanning before 24 weeks of pregnancy. Health Technology Assessment. Glasgow: NHS Quality Improvement Scotland; 2004. Report No.: 5.
Indexing Status Subject indexing assigned by NLM MeSH Chromosome Aberrations; Cost-Benefit Analysis; Female; Fetus /abnormalities; Humans; Models, Biological; Patient Acceptance of Health Care; Pregnancy; Pregnancy Trimester, First; Pregnancy Trimester, Second; Prenatal Diagnosis /economics; Scotland; Ultrasonography, Prenatal /economics AccessionNumber 22005008266 Date bibliographic record published 28/02/2006 Date abstract record published 28/02/2006 |
|
|
|