The four pillars of the ethical framework can be used to evaluate the potential expansion of the scope of NIPT. Below, four limits are proposed to the responsible expansion of the scope of NIPT in the future. These limits provide ethical guidance for professionals and policy-makers who are working in the field of NIPT and will be shaping its development and further implementation in the future.
Limits set by the aim of prenatal screening
In the five documents it is explicitly stated that although the aim of prenatal screening is not to maximise reproductive choice indefinitely, there is room for expansion of the screening offer [
15,
17,
20]. In the interviews several professionals indicated that a broader test will contribute to the aim of prenatal screening because an expanded NIPT allows for detecting more disorders than trisomy 21, 18 and 13: “
People do not want a test for Down syndrome, but a test for a healthy child.” (I15)
However, an expanded scope might affect informed choice as a precondition of reproductive autonomy. When NIPT includes a high number of diseases, it will be difficult in pre-test counselling to discuss all possible test outcomes in detail, “including the full range of variability in the manifestations” of these diseases [
15]. Testing for more abnormalities might thus “paradoxically undermine rather than serve or enhance reproductive autonomy.” [
15] A clinical professional feared that “
people have no idea what the results [of NIPT] can be and what these could mean to them. I am sure about this, because for Down syndrome it is already the case [that people do not understand what the outcome means to them].” (I7, medical specialist)
This raises the question how to best inform pregnant women prior to the test. It has been suggested in documents and by some of our respondents that information about the possible outcomes of prenatal screening should be presented as categories of disorders: the scope of NIPT can be narrow or broad, with results pertaining to severe or non-severe disorders and early- or late-onset disorders. When the scope of NIPT expands to such an extent that it becomes impossible to describe in detail all possible test outcomes during pre-test counselling, the counsellor “should describe the general characteristics of the categories of disorders tested for (e.g., mental disability or neurological impairment). Women will receive intensive counselling after a foetal diagnosis.” [
20] This model of informed choice is sometimes referred to as ‘generic consent’, which is thought to be a solution for complex counselling and has already been discussed in the context of genetic screening. Generic consent aims to prevent ‘information overload’ and to avoid the provision of information that is “pointless or counterproductive” [
45]. The question arises whether generic consent offers enough information to enable people to make a truly informed choices [
46,
47]. The ESHG/ASHG and the Dutch Health Council have their reservations about generic consent [
15,
17], because “the feasibility of this model has not yet been empirically tested in the prenatal context” and it remains unclear how
informed generic consent would be [
15]. The extent to which generic consent can be informed consent should be studied in line with previous studies on informed choice in the context of prenatal screening. These studies showed highly variable percentages of women having made informed choices: 89%, 77,9%, 51% and 44% [
35,
48‐
50]. Some of that variation might be explained by variation in the nature and the quality of pre-test counselling practices, which will likely affect the ‘informedness’ of women’s choices to a great extent, also in the context of an expanded NIPT. In practice, it is not clear whether a sufficient number of professionals will be available to counsel large numbers of pregnant women and their partners, and whether they will have enough time to explain the details of the test and facilitate informed decision-making. In some countries, measures have been put in place to counter this problem, including the use of decision aids and the additional training of midwives in NIPT counselling [
51]. Another solution might be a change in the
focus of counselling, from technical-medical aspects to women’s values or goals related to screening. As respondents suggested, too, counselling is more than providing information; women should be triggered to think about why they would want prenatal screening and what they would do in case of an abnormal test result, to make them more aware of their attitude towards undergoing prenatal screening. Attitude is defined as the general feeling of ‘favourableness’ or ‘unfavourableness’ for testing [
34]. Triggering women to think about testing might lead to a process of deliberation and evaluation of pros and cons, which, according to several authors, should be part and parcel of an informed choice [
49]. Professionals could play a role in this deliberation and help women to formulate their values, for instance in accordance with the interpretive model of the physician-patient relationship, as described by Emanuel and Emanuel [
52]. This model entails that the healthcare professional helps to elicit the norms and values of a patient.
We would suggest that in this process, the necessary technical information about the test could support or influence the attitude, but is not sufficient or even essential to the quality of decision-making. Shifting the focus of counselling from ‘conveying knowledge about screening’ to ‘exploring women’s attitude towards screening’ might improve women’s and their partners’ decision-making processes, even in the context of an expanded scope of screening and, in combination with decision aids, takes away the time pressure to explain all clinical and technical details of NIPT.
Professionals differed in their opinions about whether women should be given a say in decisions regarding the scope of the screening offer. Some professionals suggested that a list of options should be offered from which women could choose, whereas others believed that experts should determine which (categories of) disorders should be included in the test. The main reason for preferring a predetermined offer was that women might not have the information – or the capacity to understand the information – required to make a decision about the adequate scope of NIPT. Another study of opinions of professionals showed that a majority of respondents preferred a predetermined offer or a fixed list of disorders to be tested [
53].
A second category of problems arises with the dual aim of prenatal screening within antenatal care systems [
15]. Some routinely offered prenatal screening tests are used to improve pregnancy outcomes or the health condition of the mother or the baby, such as the blood test for rhesus status in RhD-negative women. The rhesus test is currently offered as a separate test but could – for reasons of efficiency – be combined in one test with NIPT for autosomal aneuploidies. An objection to a combination of this test with screening for aneuploidies is the possible confusion in women about what test they should accept or decline, and for what reasons. Prenatal screening for aneuploidies is aimed at reproductive autonomy and requires non-directive counselling [
15]. The term ‘non-directiveness’ refers to the absence of coercion or the withholding of advice, in order to respect the autonomy of a patient [
54]. According to Ten Have, as cited in Oduncu, non-directiveness means that the expert who provides information about genetic conditions “should not, in any respect, try to influence the decision made by the persons who are counselled or screened. (…) his aim is merely to provide information and to help the patients or clients to work through possible options.” [
54] For prevention-aimed screening in antenatal care (e.g. screening for hypertension or rhesus status), it may not be objectionable for health professionals to recommend or insist on participation, because this type of screening promotes the health of the mother and the foetus, but for autonomy-aimed screening, directive counselling is not appropriate [
15,
55]. In sum, one (expanded) NIPT that combines two aims and two - opposed - modes of counselling is not desirable.
NIPT is meant to offer reproductive options, but not to screen foetuses for all kinds of medical problems. For instance, children are usually not allowed to undergo predictive testing for (untreatable) late-onset diseases because this might affect their right to an open future and their right not to know unwanted predictive information [
15,
17‐
20]. The principle to defer testing until adulthood applies to unborn children as well. Prenatal screening is not meant as a medical screening of future children: its scope should thus be limited to those conditions for which expecting parents may consider terminating the pregnancy. To protect the unborn child’s right not to know, ‘conditional access’ models have been proposed for women who want information about late-onset diseases: testing for late-onset diseases, including some sex chromosomal aneuploidies, will only be offered if women “expressed the clear intention to choose abortion if a predisposition for a late-onset diseases is found.” [
56] However, as termination of a pregnancy is, and should continue to be, the result of a voluntary decision, women who change their minds about an earlier expressed intention cannot be forced to terminate an affected pregnancy. Therefore, it cannot be excluded that children may be born in the knowledge of carrying a mutation for a late-onset disease. Further research should focus on the consequences of living with this information for both parents and children and on its effects on their relationship [
18,
19].
NIPT may contribute to the aim of prenatal screening: the promotion of reproductive autonomy. On the basis of the first pillar of the ethical framework for prenatal screening, however, limits can be set to the morally responsible expansion of the scope of NIPT: NIPT should generate only test outcomes that are relevant to reproductive decision-making, and informed choice should be (made) possible through adequate pre-test counselling.
Limits set by proportionality
The expansion of the scope of NIPT also raises questions concerning proportionality. According to the Dutch Health Council, proportionality is an important requirement of prenatal screening, and benefits of each ‘test’ (for each condition) to be included in the screening offer should outweigh the harms [
17]. Professionals noted that it may be beneficial to include more disorders in a test because that means that more reproductive choices can be made:
“There are children who are born with a severe disorder. Then we do an exome analysis to see what the cause is. Then we find, say, in 40% of the cases, a new mutation, in a crucial gene, which the parents do not have. In the future it may be possible to detect that [mutation] in maternal blood.” (I15, lab specialist) Several professionals gave the example of the 22q11 deletion, which is associated with a severe phenotype. Studies on the attitudes of pregnant women towards an expanded scope of prenatal screening showed that women thought that it may be valuable especially to include severe disorders with no or short life expectancy in a screening test [
57,
58]. Women also wanted to learn about sex chromosomal aneuploidies [
59,
60] and about specific other aneuploidies, but were hesitant about learning about conditions with unknown or variable phenotypic expression. They were uncertain about what the benefit would be of knowing about such conditions [
59].
Proportionality concerns might limit the expansion of the scope, on at least three points. Firstly, when genome-wide analyses are used in NIPT, it might be difficult to assess the clinical validity of many among the huge number of abnormalities that can be detected. Offering a test for disorders without knowing the validity might lead to false positives and false negatives, cause harm to pregnant women, and challenge the proportionality of including the disorders [
17]. Professionals mentioned that outcomes should be actionable for pregnant women. When tests are not reliable (i.e. clinically valid), they provide few actionable options. Moreover, uncertain test outcomes might lead to unnecessary anxiety or insecurity in pregnant women, which is objectionable:
“I think that, when you introduce uncertainty in the pregnancy, it will become difficult. If you [can say that you] are sure that the child is disabled, then this is understandable for people, and they will be able to prepare [for the birth of a disabled child] or to decide that they do not want this. But if you say, ‘we actually do not know what it means exactly; (…) it can turn out better than expected, but the child can also turn out severely disabled.’ Well, what should you do, as parents?” (I3, medical specialist) Several other professionals stated that in practice this should not pose a big problem, as only a small number of abnormalities that are currently being detected in labs are of unknown or little-known clinical validity. These will need to be discussed between expecting parents and clinical geneticists specialised in prenatal diagnosis.
A second point that several respondents stressed is that NIPT has shortcomings: NIPT is not a diagnostic test, and it still requires invasive follow-up. An expanded scope might lead to an increasing number of positive test results for a wide range of disorders, which will include false positive results that need confirmation by (unnecessary) invasive diagnostic testing. This is problematic, because a reduction of invasive tests as compared to FCT is seen as one of the important benefits of NIPT [
15,
18,
19].
A third point that might limit the scope of NIPT is the burden of the decision to terminate a pregnancy. Some disorders may not be sufficiently severe to justify their inclusion in the NIPT; they may not meet the first screening criterion of Wilson and Jungner: “The condition sought should be an important health problem.” [
61] However, professionals mentioned that it is hard to define what ‘serious’ or ‘non-serious’ diseases are. In the documents it is stated, for instance, that severity should not be determined at all: “It would be dangerous to create medical, legal, or social definitions of ‘serious’, because these could infringe on couples’ lives in several ways.” [
20] Expecting parents are the ones who should indicate whether they consider a disorder to be serious or not, in their life situation [
20]. Although it will be difficult in practice to draw the lines, the seriousness of disorders can serve as an (arguable) limit to the expanding scope of NIPT.
From the pillar of proportionality a few additional limits can be derived for the expansion of the scope of NIPT: in order for tests to be included in an expanded scope of NIPT, they should be clinically valid. Especially the positive predictive value should be high, as confirmatory testing through invasive procedures will still be required and is associated with risks, costs and burdens. NIPT should not be offered for trivial conditions.
Limits set by justice aspects
When using the ethical framework to evaluate an expansion of the scope NIPT, the pillar of justice is less prominent than the other three pillars. However, there are three issues that arise from the pillar of justice. Firstly, when NIPT is offered as an expanded test, it should be available equally for every pregnant woman [
17]. Equal access to healthcare is considered to be a fundamental right that should preclude the exclusion of specific groups from healthcare services [
62]. Women should not face restrictions to having reproductive options. Ideally, all women should have access to the same information about their foetus, and the scope of first-trimester prenatal screening should be equal for all women. When expanded NIPT is made available only to women who have an increased risk of trisomy 21, 18 or 13 as a second-tier test after FTC, for instance, low-risk pregnant women will not have access to information about the foetus other than the three more common trisomies detected through FCT, whereas high-risk women will [
17]. For this reason, justice would require making NIPT available as a first-tier test to all women (or restricting the scope of NIPT as a second-tier test). Also, it is important to note that diagnostic follow-up testing should be made available to women who have undergone NIPT, in line with the criterion of Wilson and Jungner that in screening programmes, diagnostic follow-up testing should be available to those found to be at risk [
38]. This is of special importance in countries in which access to follow-up testing is not self-evident.
A second aspect, according to the International Bioethics Committee, is that education is a matter of justice: “Persons with a lower education level and lower health literacy are denied the information which is required to exercise their freedom and autonomy.” [
19] Some women may not be able to understand all relevant information pertaining to the screening offer, which is necessary to make an informed choice. The expansion of NIPT will only exacerbate this inequality [
19], it is feared, as the test becomes more elaborate and more complex, and decision-making places higher demands on women’s health literacy.
A third concern is that an expanded NIPT could challenge a justifiable distribution of healthcare resources. As resources are scarce and should be distributed equally, efforts must be taken to demarcate the scope of prenatal screening tests to prevent unnecessary follow-up of clinically insignificant findings. Besides, when prenatal screening is offered within the context of a public health programme and is upheld by taxpayers, there should be transparency with regard to the utility of the test [
15]. This also underlines the importance of ensuring the proportionality of a test.
When considering the costs of prenatal screening it should be noted that a widespread implementation and uptake of prenatal screening programmes is likely to lead to the birth of fewer affected children, which reduces the costs associated with their healthcare and support. Although this should not be an aim of prenatal screening, these long-term costs savings are undeniably part of a cost-effectiveness analysis of new screening tests [
15].
From the pillar of justice another limitation can be derived: expanded NIPT should be available for all pregnant women, which may increase the costs of the programme. This limitation may change over time as the technology improves and becomes cheaper.
Limits set by societal aspects
In discussions on the expansion of the scope of NIPT, concerns are reiterated that have already been raised in the context of earlier prenatal screening programmes, such as discrimination and stigmatisation of people with chronic diseases. New societal aspects, unique to expanded NIPT, are raised as well. Professionals noted in the interviews, for instance, that a benefit of an expanded scope could be a removal of the focus of prenatal testing on Down syndrome. Down syndrome is the most common of the three trisomies and in the Netherlands first-trimester screening for chromosomal abnormalities is often referred to as a ‘test for Down syndrome’. By expanding the scope of prenatal screening this focus could shift, which might reduce concerns related to discriminatory messages conveyed by the screening programme. This benefit of the expansion is also acknowledged by parents of children with Down syndrome, who experience the focus on Down as stigmatising for their children [
63]. On the other hand, the Dutch Health Council mentioned that expanded NIPT is not free from the allegation of stigmatisation either, as, for instance, a list of selected disorders can be thought of as ‘subjective’ and vulnerable to stigmatisation of specific groups, too [
17]. According to some professionals, an expanded scope might reduce the acceptance of children with a disability: “
With 22q11 deletion, [children] can be mentally retarded, etc. When people hear a story like that, they tend to terminate [the pregnancy]. I find it very hard. Everybody wants a healthy child; I understand that. So it is good to have these options. On the other hand, I am afraid that, when more [screening] becomes possible, what space is there for children with a disability? I find it terrible that there may be no respect or no care [for these children].” (I10, midwife).
Adverse societal consequences of an expanded scope are also mentioned by pregnant women and parents of children with Down syndrome, who fear a loss of diversity in society and a ‘slippery slope’, implying that people might want to start testing for increasingly trivial abnormalities [
57,
63]. However, respondents questioned whether these consequences of an expanded scope will occur and denied that society will eventually be without children with a disorder or disability. Although it is difficult to predict the societal consequences (if any) of NIPT or how these would limit the expansion of its scope, it is clear that negative consequences for people with disabilities should be mitigated, and the practice of prenatal screening should be monitored continuously, not only with a focus on the risks and benefits for individuals, but also for its wider societal implications.