Antenatal ultrasound screening for fetal abnormalities: a systematic review of studies of cost and cost effectiveness

OBJECTIVE

To review systematically and critically evidence to derive estimates of costs and cost effectiveness of routine ultrasound screening for fetal abnormalities.

DESIGN

A systematic review of the literature using explicit criteria for inclusion of primary research studies, a stated electronic strategy to identify relevant material, and an explanation of why apparently relevant studies have not been included.

SETTING

All countries of origin were included. The results of this review are important to obstetricians and to health service managers in the allocation of resources, and others who are considering conducting further research in this area.

MAIN OUTCOME MEASURE

Formal economic evaluations and cost studies of routine ultrasound screening. Costs of routine anomaly scans and costs of other procedures carried out as part of antenatal screening by ultrasound.

RESULTS

One hundred and ninety-nine studies were identified in total, 24 reaching the final stage of the review. Nine studies were formal economic evaluations and 15 reported costs studies or clinical effectiveness studies with some assessment of cost. The studies were carried out mainly in Europe and in the United States. After quality criteria were applied, data were extracted from six of the economic evaluations and six of the costs studies. One economic evaluation conducted alongside a randomised trial concluded that screening for fetal abnormalities by ultrasound in the second trimester was cost effective, compared with routine antenatal care. The costs of routine scans ranged from Pound Sterling 18 to Pound Sterling 204 and for non-routine ranged from Pound Sterling 32 to Pound Sterling 113.

CONCLUSIONS

There is a lack of good quality primary studies of the costs of ultrasound screening in pregnancy. Typically, economic evaluations of ultrasound screening have been based on poor quality evidence of clinical effectiveness. There is a need for more published data on the costs and cost effectiveness of routine ultrasound screening for fetal anomalies, and of the longer term consequences of screening for anomalies.

Methodological limitations of economic evaluations of antenatal screening

A review of recent economic studies of antenatal screening reveals widespread violation of accepted economic evaluation methodology. In particular, the costs and benefits of antenatal screening are often misclassified and conflated, and the non-resource effects of averted costs are often excluded from the evaluation process. The result is a widespread violation of the explicit and systematic approaches taken by economic analysts more generally, and conclusions that may be described as misleading. This letter calls for economic analysts to be consistent in their application of economic evaluation methodology to antenatal screening programmes.

Mortality from congenital abnormality in Malaysia 1991-1997: the effect of economic development on death due to congenital heart disease

An analysis was done of available data from the Department of Statistics Malaysia, on the type of congenital abnormality contributing to death, to determine whether progress in health care over recent years was associated with any decline in mortality from congenital abnormality. A significant decline in death due to congenital abnormality was observed between 1991 and 1996. This was attributable to a decline in deaths due to congenital heart disease occurring because of improvements in cardiac surgical services for infants. In 1997 death due to congenital heart disease increased significantly. This could be attributed to improvements in the diagnosis of congenital heart disease in the neonate.

[Diagnosis-related groups and neonatal surgical patients]

Diagnose Related Groups (DRG) are defined on the basis of the principal diagnosis, secondary diagnoses, procedures, age, sex and discharge status, and were developed to improve hospital productivity and efficacy. Existing code systems do not cover all medical specialties equally well; examples are neonatal medicine, cancer treatment and rehabilitation. We have developed a prospective method to measure actual costs related to patients individually. The major element in this method is based upon the hospital stay being divided into types of treatment with different resource requirements: heavy intensive care, light intensive care, intermediate care and ordinary care. In addition, costs related to surgery and other procedures are measured. Our method was used to calculate costs related to neonatal surgery due to various inborn diseases in the gastrointestinal tract and the urinary system. All patients needed immediate care and competent medical intervention. Mean costs for the group was NOK 291,181 while total reimbursement to the hospital was NOK 100,390, resulting in a net negative balance of NOK 190,970. Neonatal surgery does not seem to be adequately covered by the DRG system. This complex patient group provides a comprehensive test of the prospective method, and after evaluation we feel that it can be used in most other patient groups to verify actual cost.

Infant mortality and congenital anomalies from 1950 to 1994: an international perspective

STUDY OBJECTIVE

To provide an international perspective on the impact of congenital anomalies on infant mortality from 1950 to 1994.

DESIGN

Population-based study based on data obtained from vital statistics reported to the World Health Organisation.

SETTINGS

36 countries from Europe, the Middle East, the Americas, Asia, and the South Pacific.

RESULTS

On average, infant mortality declined 68.8 per cent from 1950 to 1994. In the countries studied, infant mortality attributable to congenital anomalies decreased by 33.4 per cent, although it recently increased in some countries in Central and Latin America and in Eastern Europe. Anomalies of the heart and of the central nervous system accounted for 48.9 per cent of infant deaths attributable to congenital anomalies. During 1990-1994, infant mortality attributable to congenital anomalies was inversely correlated to the per capita gross domestic product in the countries studied. At the same time, the proportion of infant deaths attributable to congenital malformations was directly correlated with the per capita gross domestic product.

CONCLUSIONS

Congenital malformations account for an increasing proportion of infant deaths in both developed and developing countries. Infant mortality attributable to congenital anomalies is higher in poorer countries although as a proportion of infant deaths it is greater in wealthier countries. Conditions such as spina bifida, whose occurrence can be reduced through preventive strategies, still cause many infant deaths. The apparent increase of infant mortality because of congenital anomalies in some countries should be investigated to confirm the finding, find the causes, and provide prevention opportunities.

Fiscal impact of a potential legislative ban on second trimester elective terminations for prenatally diagnosed abnormalities

This study was designed to determine the fiscal impact of a theoretical legislative ban on elective terminations for prenatally diagnosed abnormalities at Hutzel Hospital/Wayne State University. A fiscal comparison was completed for patients who had second trimester elective terminations for prenatally diagnosed abnormalities versus not allowing the procedure. An eight-year database of genetics cases and hospital and physician cost estimates for performing elective terminations for prenatally diagnosed abnormalities, and published reports of the average lifetime costs per selected birth defects, were used to calculate the net cost. The estimated lifetime cost for an average cohort year of a legislative ban on elective terminations for prenatally diagnosed abnormalities was found to be at least $8.5 million for patients treated at Hutzel Hospital. Extrapolated, a similar ban on second trimester elective terminations would have a net cost of $74 million in Michigan and $2 billion annually in the United States.

Cost-effectiveness of neonatal surgery: a review

BACKGROUND/PURPOSE

Technological developments have revolutionized both diagnosis and treatment in neonatal surgery. However, it has been increasingly recognized that financial resources might become insufficient to provide all the medical care that is technically feasible or that patients and families might desire. The purpose of this study is to apply the theory of health economics to neonatal surgery and to explore the extent and the kind of economic evaluation done in neonatal surgery.

METHODS

To explore the work done so far, the authors undertook a literature search aimed at costs and effects of surgical interventions in newborns with Ravitch' surgical index diagnoses of congenital anomalies. Common keywords in cost-effectiveness analysis were used to search Medline.

RESULTS

Evidence about the cost effectiveness of neonatal surgery is largely lacking. This is probably because of difficulties in long-term tracking of the patients and to the problem that most generic quality-of-life measures are not applicable in children yet.

CONCLUSIONS

Further cost-effectiveness research in neonatal surgery is warranted to settle priority discussions in health care when neonatal surgery is part of such discussions. Methodology for generic quality-of-life measurement in children is badly needed.

Outcomes and resource utilization for newborns with major congenital malformations: the initial NICU admission

HYPOTHESIS

Newborns with major congenital malformations (MCM) have contributed to a significant proportion of resource utilization in a regional referral neonatal intensive care unit (NICU).

SETTING

The Children's Hospital Medical Center NICU, Cincinnati, OH.

SUBJECTS

Newborns with and without MCM admitted from August 1, 1993 through July 31, 1994. Total patients studied were 572; 147 with and 385 without MCM. No intervention was performed in this observational study.

STATISTICS

Statistics were t test, chi-squared, and rank sum analysis.

RESULTS

MCM accounted for 27.6% of NICU referrals, 32.4% of total NICU days, and 39.6% of NICU costs. Both median cost per patient and length of stay were significantly (p < 0.01) higher for patients with MCM than those without MCM. Surgery was more frequent in MCM than non-MCM cases. Thirty-three percent of the newborns with MCM received ongoing medical support at discharge.

CONCLUSION

Patients with MCM remain as one of the largest and costliest groups hospitalized in a referral NICU.

Prevalence of congenital malformations and genetic diseases in Korea

A nationwide investigation of congenital malformations and genetic diseases in Korea was conducted by analyzing Medical Insurance data for infants aged under 1 year. Medical Insurance data were obtained for 1993 and 1994 and the ICD-9 (International Classification of Diseases, Ninth Revision) code was used to classify the diseases. The coverage rate of medical insurance was approximately 95% of the total population. Anomalies of the cardiovascular, musculoskeletal, and gastrointestinal systems, in descending order of frequency, were more frequent than anomalies in other systems. The average prevalence of cardiovascular anomalies for 1993 and 1994 was 15 per 1000 infants, and ventricular septal defect, with an average prevalence of about 3.50 per 1000 for 1993 and 1994, was the most frequent cardiovascular anomaly in infants. Polydactyly was the most frequent musculoskeletal anomaly, with an average prevalence, for 1993 and 1994, of about 1.20 per 1000 infants. Anencephaly had the highest frequency of nervous system anomalies. Congenital hypertrophic pyloric stenosis was the most common of the gastrointestinal anomalies. The prevalence of the congenital malformations and genetic diseases examined was similar to that reported in other countries. Total medical expenses for the care of patients with each disease entity were also estimated. The highest medical expenses were incurred for ventricular septal defect, congenital coagulation factor VIII disorders, atrial septal defect, tetralogy of Fallot, and spinal anomalies, in descending order of magnitude. This investigation could be helpful in planning social welfare systems, as well as for elucidating the current status of congenital malformations and genetic diseases in Korea, and in other Asian countries.

Medical costs and lost productivity from health conditions at volatile organic compound-contaminated superfund sites

This paper estimates the health costs at Superfund sites for conditions associated with volatile organic compounds (VOCs) in drinking water. Health conditions were identified from published literature and registry information as occurring at excess rates in VOC-exposed populations. These health conditions were: (1) some categories of birth defects, (2) urinary tract disorders, (3) diabetes, (4) eczema and skin conditions, (5) anemia, (6) speech and hearing impairments in children under 10 years of age, and (7) stroke. Excess rates were used to estimate the excess number of cases occurring among the total population living within one-half mile of 258 Superfund sites. These sites had evidence of completed human exposure pathways for VOCs in drinking water. For each type of medical condition, an individual's expected medical costs, long-term care costs, and lost work time due to illness or premature mortality were estimated. Costs were calculated to be approximately $330 million per year, in the absence of any remediation or public health intervention programs. The results indicate the general magnitude of the economic burden associated with a limited number of contaminants at a portion of all Superfund sites, thus suggesting that the burden would be greater than that estimated in this study if all contaminants at all Superfund sites could be taken into account.

Eurofetus: an evaluation of routine ultrasound screening for the detection of fetal defects. Aims and method

The aims of this study are (1) to evaluate the efficiency of ultrasound in detecting CA in low-risk populations of pregnant women by routine screening performed in hospital ultrasound labs (level II); (2) to highlight the areas where improvement could be obtained; (3) to determine efficient timing and number of examinations; (4) to evaluate the psychological returns of detection and nondetection of CA; and (5) to evaluate the cost-effectiveness ratio of antenatal screening of CA. A European collaboration was supposed to help in meeting these objectives because results concerning the analysis of individual CAs or groups of CAs can only be statistically significant when their number is sufficiently large. It was estimated that it was necessary to collect nearly 5,000 CA; this corresponds to about 200,000 pregnant women, the prevalence of malformations at birth being estimated at 2.5%. These conditions yield worthy conclusions, given the following circumstances: a large variety of CA, the extremely low incidence of each CA, the multiple approaches for diagnosis and management, the manifold classes of defects, the differences in gestational age when anomalies are detectable and detected. We study prospectively (1) the reliability of ultrasound in detecting antenatal malformations by recording all CA, ultrasonically suspected and not; (2) the gestational age of anomaly recognition; (3) the response to antenatal diagnosis of CA; (4) the individual outcome of pregnancies; (5) the financial cost of the screening program; and (6) the psychological consequences for the parents.

Ethical dimensions of ultrasound screening for fetal anomalies

The ethical dimensions of the debate on routine ultrasound are analyzed. The central role of the informed consent process, based on a respect for the autonomy of the pregnant woman, is presented. Failure to offer quality ultrasound in clinical settings where it is available restricts access to pregnant women to the diagnosis of fetal anomalies and therefore restricts access to the options of abortion and fetal therapy. We show that beneficence- and justice-based considerations do not supersede respect for autonomy.

Managed care and ultrasound for the diagnosis of fetal anomalies

Managed care organizations and their affiliated group practices approach technology evaluation in a methodical way. This paper reviews the factors used in analysis of ultrasound for the diagnosis of fetal anomalies. It includes one group practice's strategies for ultrasound management while supporting the use of the second trimester fetal survey.

Sensitivity of fetal anomaly detection as a function of time

In this paper, we show that the ratio of the number of fetal anomalies detected by ultrasounds (US) to the total number of cases is not a consistent estimator of the US sensitivity. As Eddy pointed out, when the disease evolves over time, the sensitivity of a test also varies over time according to the development of the disease. To assess correctly the detection capability of a test, it is therefore necessary to estimate a time continuous function (sensitivity function) instead of a single parameter. From a methodological point of view, by considering the "detectability" time of a fetal anomaly as a random variable and parametrizing its distribution function, we estimate the probability that an anomaly is detected conditional upon the precise timing of actually performed US during pregnancy. We fit this model with Eurofetus data (about 7,300 abnormal fetuses), and we compare estimations for different kinds of anomalies (classification based on the system involved and/or severity of the handicap). To allow for heterogeneity of anomalies regarding the detectability time, we generally adopt mixture models. For instance, we select a bi-gamma distribution for major malformations and estimate that 63% of such anomalies are detectable quite early in pregnancy (conditional mean: 15.2 weeks of amenorrhea (WA) +/- 4.2 WA), the others becoming detectable later (30.3 WA +/- 6.4 WA). Such results are then integrated in a cost-effectiveness analysis.

Can decision analysis help us decide whether ultrasound screening for fetal anomalies is worth it?

Decision analysis is a widely used tool to improve clinical decision making when randomized controlled trials are infeasible, underpowered, or lack generalizability. We performed an exploratory decision analysis of routine second trimester ultrasound to detect fetal anomalies, focusing on the assumptions that would have the greatest impact. Six outcome categories were considered: (1) abnormal ultrasound, anomalous child, (2) abnormal ultrasound, elective abortion of anomalous fetus, (3) abnormal ultrasound, healthy child, (4) abnormal ultrasound, elective abortion of healthy fetus, (5) normal ultrasound, healthy child, and (6) normal ultrasound, anomalous child. Live birth and fetal death rates for nine sonographically detectable anomalies were obtained from the California Birth Defects Monitoring Program. The sensitivity and specificity of ultrasound were estimated through meta-analysis of recent series. Plausible ranges for the probabilities of cesarean delivery and elective abortion, by anomaly, were determined through review of the literature. Standard gamble, willingness-to-pay, and human capital estimates of utility were rescaled for comparability. We found that routine ultrasound appears to be the preferred strategy for most women. This choice is sensitive primarily to the specificity of ultrasound and women's willingness-to-pay for the reassurance of a normal ultrasound.

Reduced costs of congenital anomalies from fetal ultrasound: are they sufficient to justify routine screening in the United States?

No comprehensive benefit-to-cost analysis has been performed to date on a policy of routine ultrasound screening for fetal anomalies in the United States. We performed a preliminary benefit-to-cost analysis drawing upon our previous research on the cost or birth defects in the United States and upon the literature regarding (1) the sensitivity of ultrasound in detecting congenital anomalies, (2) the rate at which pregnancies are terminated upon detection of fetal anomalies, (3) the number of ultrasounds performed per pregnancy under a routine screening policy, and (4) the average cost of an ultrasound. We assumed a 100% subsequent replacement rate of terminated pregnancies with a normal child, an assumption most favorable to routine screening. The benefit-to-cost ratio ranged from .33 to 3, suggesting that a routine screening policy for fetal anomalies is of uncertain net societal benefit. Routine screening may be justified, however, based on standards that elude the methods for establishing societal benefits adopted in this analysis.

Contribution of birth defects and genetic diseases to pediatric hospitalizations. A population-based study

OBJECTIVE

To estimate the contribution of birth defects and genetic diseases to pediatric hospitalizations by use of population-based data.

DESIGN

Hospital discharges were categorized according to the diagnostic codes of The International Classification of Diseases, Ninth Revision, Clinical Modification. Hospitalizations that were related to birth defects and genetic diseases were compared with hospitalizations for other reasons, with respect to age, race/ethnicity, sex, length of stay, charges, source of payment, and mortality rate. Hospitalization rates and per capita charges were computed with the use of population estimates from 1990 census data.

MATERIALS

The 1991 population-based hospital discharge data from California and South Carolina.

RESULTS

Nearly 12% of pediatric hospitalizations in the 2 states combined were related to birth defects and genetic diseases. These children were, on average, about 3 years younger, stayed 3 days longer in a hospital, incurred 184% higher charges, and had a 4 1/2 times greater in-hospital mortality rate than children who were hospitalized for other reasons. The rate of hospitalizations that were related to birth defects and genetic diseases was 4 per 1000 children in both states, but these rates varied by age and race.

CONCLUSION

These population-based data are the first contemporary findings to show the substantial morbidity rate and hospitalization charges associated with birth defects and genetic diseases in the pediatric population.

IMPLICATIONS

This information is important for planning effective health care strategies, especially as the causes, treatments, and prevention of these disorders are being further elucidated by findings from human genome research and epidemiologic studies.