Questioning the costs and benefits of non-invasive prenatal testing

Abstract

Prenatal testing for Down syndrome through the use of non-invasive prenatal testing (NIPT) has been increasingly implemented in clinical practice and a recent cost analysis suggests that NIPT is cost effective when compared to other screening modalities in high risk populations. However, this anaylsis makes many assumptions regarding uptake of testing and pregnancy termination, which cannot be applied to all populations in the United States. Additionally, this cost analysis, which hinges on fewer Down syndrome births, does not align with the goals of prenatal testing to support autonomous and value consistent decisions. NIPT is an expensive new technology and more careful analysis is needed to determine the impact of NIPT on outcomes and overall healthcare costs.

• Aneuploidy screening
• costs analysis
• Down syndrome
• prenatal screening

We write in reference to the recently published cost analysis by Song et al., Clinical Utility of Non-Invasive Prenatal Testing with cffDNA Analysis in High Risk Women Based on a US Population [1]. While we agree that preliminary studies of non-invasive prenatal testing (NIPT) show promise as a screening test for aneuploidy, we believe there are some inaccurate assumptions with this analysis that deserve further consideration. Additionally, we are concerned that this cost-benefit analysis is incongruent with the foundations of current prenatal screening program goals, namely to support informed, autonomous and value consistent decisions. Efforts should evaluate the impact of these tests on patients and healthcare outcomes. In addition, research and resources are needed to improve patient education to ensure that testing is based on informed patient decisions for appropriate utilization of healthcare resources.

First, this study assumes that in women 35 and older, NIPT will replace current aneuploidy screening methods including use of nuchal translucency (NT) in the first trimester. While the more expensive NIPT may replace the Integrated or first trimester serum screening, NT screening may remain available as an adjunct to NIPT due to the ability to detect congenital heart disease and other conditions at an early gestational age [2]. Furthermore, maternal serum alpha-fetoprotein (MSAFP) in the second trimester may also remain a part of the screening paradigm for the detection of neural tube defects. The potential benefits of first trimester ultrasound and serum analyte screening (in the first and second trimester) need to be compared to newer technologies such as NIPT as we evaluate the best approach to prenatal screening. Although we recognize the use of maternal age greater than 35 as an effort to define a high-risk population, risk is a continuum with true meaning determined on an individual basis. Returning to the use of maternal age as a benchmark for offering NIPT has been criticized as inappropriate and outdated [3], and guidelines by the International Society for Prenatal Diagnosis (IPSD) recommend against using maternal age as criteria for offering NIPT [4].

For this analysis, the authors assume 70% of all women will uptake serum screening; however, the overall proportion of women electing to undergo aneuploidy screening in the United States has not been rigorously studied. The actual uptake of aneuploidy screening may vary dramatically in individual populations, and can be affected by patient beliefs and attitudes, provider knowledge and understanding of testing options, as well as the availability of technology and services in the community. Estimates from the California State Prenatal Screening program indicate that approximately 75% of residents undergo screening [5]. In an United States military population with a standardized patient education program, uptake of aneuploidy screening is 30–40% [6]. Uptake rates have been discussed more extensively in literature from Canada and Europe. Rate of uptake of aneuploidy screening in Canada varies widely by province (22–71%) [7]. Screening uptake in the United Kingdom declined in recent years with a recent estimate being approximately 41% [8]. While the actual uptake rate of such screening programs likely varies dramatically, we believe the overall uptake of testing in the United States is likely lower than the 70% used in this analysis.

The authors assume that the proportion of women who would undergo NIPT following a positive first trimester or integrated screen would be 100%, which we feel is a gross overestimation impacting this cost analysis. Since NIPT is another screening test, some women may choose to proceed with diagnostic testing or decline further testing following a high risk screening result. In a recent study on testing after abnormal serum screening, approximately 40% of women chose diagnostic testing; 40% elected NIPT; and 20% declined any further testing [9]. In the California Prenatal Screening Program, 23% of women found to have an increased chance of Down syndrome declined referral for genetic counseling and consideration of further testing [5].

The authors also assume that uptake of invasive testing following a positive NIPT result would be 99%. This is likely a significant overestimate, as only 43% of women who underwent genetic counseling for an abnormal analyte screen in California elected to proceed with diagnostic testing [5]. It is possible that some women who would not undergo serum analyte screening due to the 5% false-positive rate, may choose to proceed with NIPT even if they would not pursue confirmation with invasive diagnostic testing.

The cost benefit determined by this analysis also depends on the rate of pregnancy termination when Down syndrome is diagnosed. Termination rates for Down syndrome vary; however, recent estimates suggest that rates of pregnancy termination in Down syndrome pregnancies are declining. In a recent review of seven population-based studies, the weighted mean termination rate for Down syndrome was 67% [10]. Although this review was cited by Song et al. in their cost analysis, they used a termination rate of 75% in their calculation. Currently, it is unclear how the use of NIPT will impact pregnancy termination rates for Down syndrome, but it is possible that rates will remain stable or continue to decline. Prior to NIPT some women chose to undergo diagnostic testing based on less sensitive screening tests. As stated previously, NIPT may also be popular among women who would not consider invasive diagnostic testing or pregnancy termination. Solely looking at pregnancy termination rates may oversimplify other factors that affect a woman’s choice to undergo prenatal testing and diagnosis. A recent Canadian analysis of the cost of prenatal diagnosis stated that “defining benefit in terms of fetuses aborted is inadequate as it does not capture any of the benefits of having a child, nor does it account for the difficulty of the decision to terminate a potentially wanted pregnancy” [11]. The Canadian study defines “benefit” as the number of Down syndrome pregnancies detected with the assumption that such information would be valuable whether or not termination is elected. It is recognized that women find benefit in prenatal diagnosis for preparation purposes [12], and more research into how prenatal diagnosis impacts outcomes is needed.

We recommend insurers and policy makers take a closer look at the overall cost, benefits, risks and limitations associated with new prenatal testing technologies such as NIPT based on the additional factors outlined above. Currently, is not clear that NIPT will decrease overall healthcare costs related to Down syndrome. It is possible that NIPT may increase costs by leading to more elective screening by patients who are attracted to the test due to aggressive marketing campaigns of the testing companies, or because they are enticed by the promise of learning fetal gender at an early gestational age. These individuals may have no desire to undergo follow-up invasive testing or termination of pregnancy. Adequate pretest counseling regarding NIPT and other prenatal screening options is critically important. We recommend additional resources be devoted to investigate a broader view of the potential risks, benefits and costs associated with new technologies such as NIPT. This should include completion of more rigorous studies to assess outcome measures while continuing to support individual patient needs and values in clinical practice.

Declaration of interest

The views expressed are those of the author(s) and do not reflect the official policy of the Department of the Army, the Department of Defense or the U.S. Government.