Social dimensions of preimplantation genetic diagnosis: a literature review

Abstract

The study provides a systematic overview of research on the social implications of preimplantation genetic diagnosis (PGD). The analysis focuses on empirical studies that provide insights into its heterogeneous fields of application and the concrete experiences of its users. The literature review shows that three areas of concern and controversial topics are particularly relevant for a social-scientific evaluation of PGD. Firstly, we present attitudes towards and assessments of PGD based on an ongoing expansion and transformation of its fields of application. This process includes not only more and more disorders and disease risks but also applications that are not disease-related. Secondly, there is evidence of significant gender asymmetries and financial concerns regarding the use of and access to PGD. Thirdly, the empirical studies point to a shift in normative expectations towards the idea of “genetic reproductive responsibility” and possible discriminatory consequences for individuals and families with diseases and disabilities.

Keywords:

• Preimplantation genetic diagnosis
• social implications
• risk
• inequality
• genetic responsibility
• discrimination

1. Introduction

Preimplantation genetic diagnosis (PGD) is a complex combination of reproductive techniques and genetic diagnostic procedures. Prior to establishing a pregnancy, embryos are created outside the human body using in vitro fertilization (IVF). In a second step, the embryos are screened for genetic and chromosomal characteristics using cytogenetic or molecular genetic diagnostic technologies. The purpose of genetic testing is to identify embryos that are free from a distinctive genetic or chromosomal alteration that may cause illness or disability.

Since it was first carried out successfully in 1990, PGD has generated a great deal of controversy both in public and in the academic world. Some observers in medicine and the social sciences and humanities have come to regard PGD as a new form of eugenics. According to this view, the selection of embryos prior to transfer marks a decisive threshold that goes beyond the option of terminating pregnancy subsequent to a prenatal diagnostic finding. The philosopher Jürgen Habermas, for example, has warned against “eugenic programming of desirable traits and dispositions”, which is morally reprehensible in that it “commits the person concerned to a specific life-project or, in any case, puts specific restrictions on his freedom to choose a life of his own” (Habermas 2002, 61, see also Testart and Sèle 1995; King 1999; Kuhlmann 1999). Others, in contrast, highlight the differences between the eugenics of the nineteenth and twentieth century and the current use of PGD. They argue that its current users are not aiming at population-related objectives but at the prevention of individual suffering. According to this view, the decision-making process is no longer dominated by social requirements or racial policies but embedded in a culture of reproductive autonomy (Petersen 2005; Robertson 2005). Still others even argue for an “ethical eugenics” in which PGD would be something ethically required of parents in view of their responsibility towards their potential children (Appel 2012).¹

PGD has now been available for a period of nearly 30 years. While the debate still tends to be dominated by philosophical and ethical concerns about the acceptability of creating, screening and selecting embryos according to genetic characteristics, many important societal dimensions of PGD have remained at the margins of scholarly and public attention. However, in recent years the body of empirical studies on the manifold social implications of PDG has been growing. To date no overview of this topic has been published, and our study seeks to fill this gap in the literature.² This article summarizes the findings of relevant studies that investigate social dimensions of PGD, and delineates important socio-technical trajectories and areas of concern relating to this technology.

The paper is structured as follows. To begin with, we outline the methodology and the material of our review process (2). We then briefly describe contexts and configurations of PGD, especially policy regimes and regulatory approaches in Europe (3). The next parts of the article present three significant socio-technical tendencies and controversial topics identified in the empirical studies that relate to the use of PGD. Firstly, there has been a gradual expansion of the fields of application, a broadening of the spectrum of indications to include more and more disorders and disease risks. In addition, PGD is also being used for purposes that are not disease-related. We report findings from empirical studies that address different dimensions and concerns regarding this process of expansion (4). Secondly, empirical evidence points to significant gender asymmetries and financial concerns relating to the use of and access to PGD (5), and, thirdly, empirical investigations document a shift in normative expectations towards the idea of “genetic reproductive responsibility” and possible discriminatory effects for individuals and families living with diseases and disabilities (6). The final section summarizes the results of our review and discusses possible future scenarios (7).

2. Methodology and materials

To get a comprehensive picture of the current state of research on the societal implications of PGD, we conducted a systematic literature search in the electronic database ISI Web of Science. The search was performed in March 2016 and not limited to social science and humanities publications but included the entire inventory of the database. We took into account every German and English-language journal article, book or book chapter listed in the database that met the following inclusion criteria. Firstly, it was published after 1990, the year a PGD was successfully conducted for the first time. Secondly, its title, abstract or keywords contain at least one of the following terms: “social implications”, “social consequences”, “psychosocial”, “eugenics” and “discourse” in connection with “preimplantation genetic diagnosis” or “pre-implantation genetic diagnosis”. In this way, we identified a total of 99 publications.

In the next step, we screened the titles and abstracts of all publications and removed those that focused either on the medical, scientific or technical aspects of PGD or on its ethical, legal and religious implications (e.g. possible eugenic effects of PGD, the moral status of the embryo, or the reproductive autonomy of women and couples). Ethical studies were included in the sample, however, if they discussed distinctive societal issues (e.g. sex selection, discrimination against disabled people) using empirical data.

This screening process reduced the number of publications to 78. In order to access further relevant publications not included in the ISI database, we used the bibliographies in the publications for an extended search. Additionally, we consulted other scientific databases (EBSCO Discovery Service, ProQuest Central, JSTOR). In this way, it was possible to add 40 publications to the sample that met the inclusion criteria. Thus, the following literature review is based on 118 scientific publications that either employ a social scientific research design or analyze societal dimensions of PGD by referring to empirical findings.

The data material was subsequently processed, analyzed and reflected in a series of steps. The first step consisted in a rough structuring of the relevant publications with regard to the chosen study design. The spectrum of methods used in the empirical studies on the social implications of PDG ranges from questionnaire surveys and opinion polls to interviews and participatory observation, with most studies following a quantitative research paradigm. Broadly speaking, it is possible to distinguish between three methodological approaches. The majority of the articles investigate attitudes to specific areas of application of PGD. The respondents either come from the general population or the studies focus on a specific group or segment of the population, such as medical experts or individuals directly affected by genetic conditions. The second cluster of studies documents the experiences of women or couples who have already undergone or decided to undergo PGD. Thirdly, there is a rather small number of empirical studies that examine family dynamics and psychosocial processes after a PGD and the birth of a child.

The second step in analyzing the data material consisted of close reading of the articles in order to reach an overall assessment of their content. We searched for the aims and issues raised in the empirical studies and their results, coming up with preliminary thematic groups and categorizations of domains of interest. On the basis of this rough classification and a closer examination of the textual material, we developed an analytic matrix consisting of the following categories: (1) attitudes and perspectives on different fields of application, (2) decision-making processes and psychosocial dynamics, (3) “sex selection” and gender inequalities, (4) “saviour siblings” and HLA typing, (5) discrimination and disability, and (6) financial worries and cost calculations. These categories allowed systematic text processing and comparison, while at the same time ensuring the necessary openness and flexibility of the review process. Based on this analytic scheme we identified the socio-technical tendencies and trajectories described below, which are of particular importance for an assessment of the societal implications of PGD.

The following presentation focuses on qualitative studies that are based on the concrete experiences and accounts of people with genetic disabilities, diseases and illness risks, and from carriers of recessively inherited disorders. In addition, we included investigations that cover the attitudes of the general population and medical experts regarding important fields of application of PGD. Finally, we have to point out a structural limitation that must be taken into account when interpreting the results of this literature review. Almost all studies in this field originate from European countries or the USA, Canada, Australia and Israel, and thus provide insights into specific “Western” cultural and historical contexts of PGD. However, it has to be noted that PGD is a “global form”, which has the potential for de- and recontextualization but cannot be separated from local appropriation practices and acceptance conditions (cf. Knecht, Klotz, and Beck 2012, 16–22).³

3. Contexts and configurations of PGD: policy regimes, regulatory approaches and historical trends

After the first successful carrying out of PGD in human embryos in 1990, it was initially discussed in public and scientific discourse as a procedure that could be used to avoid the inheritance of severe and untreatable monogenic diseases and chromosomal disorders. Correspondingly, the use of PGD in many countries was either restricted to this category of diseases and disorders or completely prohibited. In the course of the past three decades, this kind of strict limitation has been increasingly abandoned in many countries. However, distinctive policy regimes can be discerned. Byk (2008) provides an instructive overview by distinguishing several regulatory approaches in European countries. In Italy, Germany and Austria he identified a “prohibitive” regulation which banned PGD as a matter of principle. Since then, each of these states has revised this approach insofar as PGD has now become permissible under certain conditions (Bayefsky 2016; Gianaroli et al. 2016, 212). The majority of European countries (e.g. Portugal, Norway, Denmark) have adopted what Byk (2008, 91–92) calls a “restrictive regulatory approach” that limits the use of PGD to severe diseases. It is, however, worth stressing that this provision often includes genetic predispositions for non-fatal and/or treatable diseases.

The UK opted for the “most liberal” approach in Europe (Byk 2008, 92) and established the Human Fertilization and Embryology Authority (HFEA) for regulating and controlling the use of PGD. In the directive issued by the HEFA in 2002, PGD was initially limited to serious illnesses and chromosomal disorders (cf. Nippert 2006, 50–84).⁴ However, just one year later, the Authority pointed out that the severity of a disease could not be determined on the basis of “objective” criteria, but was rather the subject of negotiation processes.⁵ Since 2009, all conditions whose predictive testing is permitted in the UK under PGD have been published on the HFEA website. This list now includes genetic tests for about 400 diseases.⁶ Among them are several late-onset and/or treatable diseases (e.g. phenylketonuria), as well as genetic dispositions whose penetrance is significantly less than 100% (e.g. hereditary breast and ovarian cancer; see Ormondroyd et al. 2012, 4–5). In these latter cases, it is therefore ultimately uncertain whether the person in question would be affected by this disease during their lifetime (Wang and Hui 2009).

Beyond legal questions of admissibility, practical access to PGD treatment differs widely between European countries. While in the UK, as we have seen, a list of disorders for which PGD is permitted exists, other countries (e.g. Germany, Norway) rely on the work of commissions and ethical boards that decide on a case-by-case basis (Geffroy and Zerres 2016). Furthermore, most European countries limit access to heterosexual couples (e.g. France) and women of “biological reproductive age” – thus excluding different social groups and categories (e.g. single women, homosexual couples or “older” women) from PGD treatment (Ferraretti et al. 2010, 263; Bayefsky and Jennings 2015, 1–17; Gianaroli et al. 2016).⁷

The regulatory differences and discrepancies between countries result in a significant “medical migration”. Shenfield et al. (2010) investigated “cross border reproductive care” in six European countries, surveying users (N = 1230) of Artificial Reproductive Technologies (ART). While most respondents underwent gamete donation, a small number also used PGD. The study found legal restrictions regarding a particular technological option or specific user categories (e.g. age, sexual orientation) to be a major motive for traveling abroad to get treatment (see also Pennings et al. 2009; Ferraretti et al. 2010).

4. From monogenetic disorders to fertility treatment and beyond: investigating the expanding scope of PGD

Since 1997 the PGD Consortium of the European Society of Human Reproduction and Embryology (ESHRE) has been collecting data from up to 71 ART centers from all parts of the world. Their reports document 56,800 PGD cycles performed between 1997 and 2012 (De Ryke et al. 2017).⁸ The statistics not only provide evidence that the number of cycles is increasing;⁹ one can also observe a shift in the use of PGD. While between 1997 and 1998 the proportion of PGD cycles conducted for the purpose of single gene disorders (33%), x-linked disorders (25%) and aneuploidy screenings (PGS, 32%) were comparable, the proportion of the latter has increased since then to nearly 60% (Harper et al. 2012, 237; see 4.3).

In the following section, we will summarize the findings from empirical studies that investigate the social dimensions of this dynamic process. The section charts the perspectives, attitudes and concerns of different groups of users, experts and the general public leading from the focus on potentially fatal diseases via treating the problem of infertility to opting for non-medical sex selection.

4.1. From fatal conditions to disease risks

A first set of empirical studies has examined whether and, if so, how the population’s attitude to these different fields of application of PGD varies. One example is a representative survey (N = 2110) that recorded the attitude of the German general population in 2003 (Meister et al. 2005; Finck et al. 2006). Although there was no explicit legal regulation of PGD in Germany at the time, it was de facto prohibited on the basis of the German Embryo Protection Law (ESchG). The clear majority (76%) of respondents advocated the legalization of this technology in the case of illnesses that lead to death in the first year of life (Meister et al. 2005, 234). However, approval ratings dropped sharply when explicit questions were asked about the use of PGD for late-onset cancers. Only 40% of respondents supported the use of PGD for this purpose. The expansion of the indication spectrum outlined to include late-manifesting and/or treatable diseases was therefore supported by a smaller proportion of the German population than its use in the case of serious diseases that are highly likely to occur and to lead to death (see also Borkenhagen et al. 2007).

Other studies have looked at how people who are themselves affected by genetic cancer risks assess the potential uses and benefits of this technological option. A meta-analysis by Quinn et al. (2012) identified thirteen relevant studies, most of them targeting women at increased risk of hereditary breast and ovarian cancer (BRCA 1/2). The systematic overview shows that the majority of those affected support the use of PGD. At the same time, however, most respondents state that they do not want to use the procedure themselves (ibid., 195-196). This result is consistent with findings of a questionnaire survey by Rich et al. (2014) conducted in the US. More than three quarters (72%) of individuals with a hereditary cancer syndrome (N = 370) stated that affected individuals should have the opportunity to choose PGD; only 43% declared that they would consider using this technology themselves.¹⁰

However, other studies document much lower support. In a qualitative online survey (Quinn et al. 2009a) which investigated the attitude of American women who were themselves affected by hereditary breast and ovarian cancer or had a corresponding family history (N = 446), for example, approximately 70% of respondents wrote unfavorable comments related to the use of PGD for BRCA 1/2. Only 33% of the respondents could imagine using the procedure with this objective themselves (see also Lammens et al. 2009; Quinn et al. 2009b; Douma et al. 2010).¹¹

An Australian questionnaire survey provides an even more complex picture. The study by Katz et al. (2002) investigated attitudes and concerns regarding a potential future expansion of the scope of PGD. The authors asked individuals who were under treatment in a reproductive medicine clinic if they would “accept the transfer of an embryo identified as being a healthy carrier” (ibid., 1119). Such a carrier of a recessive disorder is not herself or himself affected by the disease but has a 25% chance of having an affected child if she/he procreates with another “healthy carrier”. The authors compared three groups of patients: (1) individuals presenting for PGD for a single gene disorder (N = 41), (2) individuals presenting for an unspecific Preimplantation Genetic Screening (PGS) (N = 48), and (3) individuals who were not presenting for PGD or PGS, but were about to start IVF treatment (N = 32). The findings document a significant difference between the first group on the one hand, and the second and third group on the other. While only a small minority of the individuals presenting for PGS or IVF treatment would accept a “healthy carrier”, 63% of individuals presenting for PGD for a single gene disorder stated that they would accept the transfer of such an embryo. The authors suggest that the majority of individuals presenting for IVF or PGS might misunderstand the concept of carrier status. Since half of individuals presenting for PGD for a single gene disorder were themselves at risk for or affected by this disorder, the different attitude might thus be attributable to a difference in knowledge.¹² However, rejecting a “healthy carrier” might also indicate a wish “not to pass on abnormal genes” (ibid., 1121) independent of the actual health condition of the future child.

4.2. “Saviour siblings”: selecting embryos for the health of another human

In addition to the prevention of disorders and disease risks, PGD is also used for the selection of so-called “saviour siblings” who could later serve as organ and tissue donors for another person. In this case, it is not the health of the future child but its medical use for another, already living person that is the decisive criterion for selecting an embryo.

The basis of extracorporeal selection is the determination of characteristics of the human leukocyte antigen system (HLA system). Compatibility of these characteristics between recipient and donor is necessary to reduce the risk of rejection after the transplantation of organs and tissues (see e.g. Rechitsky et al. 2004). Today, most European countries in which the use of PGD is approved also allow HLA typing (Geffroy and Zerres 2016). Reports from the ESHRE PGD Consortium have documented 200 PGD cycles performed exclusively for this purpose since 1997 (De Ryke et al. 2017, 1982–1983).

The use of PGD for this purpose is controversial both in the scientific literature and in public discourse. Borkenhagen et al. (2007) investigated the attitudes of couples who were in treatment at a fertility clinic in Germany (N = 265). The majority of couples (60%) were in favor of legalizing PGD for the purpose of selecting “saviour siblings”. However, this high level of approval is put into perspective by a US study (Kalfoglou, Scott, and Hudson 2005), which explored the attitudes of (potential) users (N = 13) on the one hand and of experts in reproductive medicine (N = 19) on the other by means of in-depth interviews. In this case too, a majority supported the use of PGD for the purpose of HLA typing in cases where the sick child would die without a tissue donation. A different result emerged, however, when a concrete scenario was described in which the intention was for the conceived child to donate not only bone marrow but also organs in the future. While some respondents also supported the implementation of PGD even in these circumstances, others rejected it in principle with reference to the health risks to the donor child. The vast majority of respondents adopted an ambivalent position in this case. The concern of medical risks to the “saviour sibling” was counterbalanced by the determination to save the threatened child at any cost.

The normative ambivalences and social implications of this controversial area of application of PGD are also the subject of the novel My Sister’s Keeper by Jodi Picoult (2005) and a movie based on the story. It tells the fictional life of Anna, a young woman who was conceived using HLA typing. After the protagonist has to donate organic material several times in her life for her sister, who suffers from leukaemia, she finally tries to defend herself against these interventions and to assert her body’s rights by legal means. Raz et al. (2017) analyzed German and Israeli reviews of this film in order to shed light on the normative framing of the topic in a culturally comparative way. While the German reviews accused the film of not adequately highlighting the ethical pitfalls of HLA typing and argued that the protagonist was being instrumentalized as a stock of raw material and spare parts, this criticism was hardly to be found in the Israeli film reviews. In accordance with the liberal regulation in Israel, the reviewers highlighted the medical significance of HLA typing and presented it as a life-saving technology. According to Raz et al. (2017), this different assessment of HLA typing corresponds to the diversity of family concepts and ethical notions: while family members in Germany are primarily addressed as singular persons with individual rights, the family in Israel is principally understood as a “unified body of members with similar interests” (ibid., 13; see also Hashiloni-Dolev and Shkedi 2007).

4.3. PGD as infertility treatment: improving IVF success rates

Whereas in the uses of PGD outlined so far, the health of a future or already living human was the criterion for embryo selection, PGD is also employed for purposes that transcend this medical framework. One important area is the use of PGD to increase the chances of success of IVF.¹³ In this case, PGD is not operating as a testing procedure by a specific medical indication, but functions as a screening instrument called Preimplantation Genetic Screening (PGS).

Despite ongoing scientific research efforts, the likelihood of success of IVF is still relatively low. For example, the pregnancy rate after IVF treatment in Germany remains at about 30% (Blumenauer et al. 2017). Many scientists assume that numerical chromosome aberrations, so-called aneuploidies, are responsible for these difficulties, especially in women over 35 years of age (e.g. Verlinsky et al. 1999). It is therefore suggested that the success rate of IVF in this group of persons could be increased significantly by aneuploidy screening (cf. Kollek 2000, 94–105). Following the ESHRE data again, most of the PGD treatments are currently performed as PGS.¹⁴ The reports document a total of 32,832 PGS cycles since 1997 (De Ryke et al. 2017, 1985–1986). In line with the hypothesis that aneuploidies are responsible for the relatively low success rate of IVF treatments, women’s age is the most common indication for such screening (cf. De Rycke et al. 2015, 3–14; see also Harper et al. 2010; Mastenbroek and Repping 2014).¹⁵

Given its enormous medical and societal significance, it is quite surprising that no empirical investigations of attitudes and expectations towards this field of application or the concerns and experiences of its users have yet been carried out. A discourse analysis of the media debate in Germany (Rödel 2015, 2018) does not directly address the issue of PGD as a means to improve the likelihood of establishing a pregnancy; however, the author shows convincingly that in a rather restrictive regulatory and political environment the argument that presents PGD as a technological option to fulfill the desire to have children has been growing in importance in the past ten years. It successfully generated support for the use of PGD in reproductive medicine – finally leading to more liberal legislation in Germany that approved the employment of PGD under specific conditions. As Rödel demonstrates in her analysis of German newspapers and weeklies published between 2000 and 2011, the charge of (eugenic) selection (PGD as a means to prevent the birth of a sick or disabled child) was increasingly replaced by framing PGD as the technological facilitation of a natural desire – even a right – to have children. The author argues that this discursive shift “strategically naturalizes” gender and reproduction: the “wish to have one’s own biological child” seems to be naturally given and self-evident – thus effectively ruling out alternative options and solutions (e.g. adoption; Rödel 2015, 223–226).

4.4. PGD as an instrument of sex selection

PGD is also used for purposes that have no connection to illness or health at all. Here, embryo selection aims at socially desirable physical or psychological characteristics of the future child, which is especially relevant in the context of sex selection. This chromosomal determination and selection of sex without medical indication is referred to in the literature as “social sexing” or “social sex selection” and has been heatedly discussed.¹⁶

The ESHRE reports document 791 PGD cycles conducted for the purpose of “social sex selection” since 1997. Their share of the total number of PGD cycles performed per year has remained relatively stable at less than one percent (see De Rycke et al. 2015, 14, 2017, 1984). This relatively low proportion is not surprising, since non-medical sex selection is prohibited in all European countries. It is legal, at least under certain circumstances, in Israel¹⁷ and the USA, where nine percent of all PGD cycles are carried out for this purpose (Bayefsky and Jennings 2015, 1–17; Bayefsky 2016, 42). In some other countries, this rate seems to be much higher. There is evidence that in Saudi Arabia (Abotalib 2013) and Lebanon (Farra et al. 2014) non-medical sex selection is even the most common indication for PGD.¹⁸

Interestingly, it is not until the seventh report of the ESHRE that information on parents’ gender preferences is recorded. The reports show that the vast majority of couples wanted a male child.¹⁹ Despite the relevance of these data for evaluating the societal implications of PGD, the pertinent data has since disappeared from the consortium’s publications. Remarkably, this is justified by referring to the controversy that followed after this information was first made public (De Ryke et al. 2017, 1984; see also Ray et al. 2003; Robertson 2003; Seif 2003; Sermon 2003).²⁰

From a feminist perspective in particular, the concern was expressed early on that this non-medical sex selection by means of PGD was propagating and deepening sexist discrimination (see Hollingsworth 2005). As it fits into a long history of technologies designed to influence the child’s sex and increase the likelihood of a male child being born, it has been argued that an evaluation of the use of PGD in this context cannot be made without an analysis of underlying societal gender relations (see Sills and Palermo 2002, 433–434; Couture et al. 2013; Gammeltoft and Wahlberg 2014, 205–207). In the bioethical debate, this feminist criticism is often countered with the thesis that, in Western societies at least, there is no longer any preference for male children (Sureau 1999; Hank and Kohler 2000; Savulescu and Dahl 2000; Dahl et al. 2006). The argument is that non-medical sex selection in these societies is motivated less by a devaluation of the female sex than by the desire for “family balancing”. According to this reasoning, it is not a given gender that is preferred, but a balance between sexes. Empirical support for this assessment is found in an Israeli study (Hashiloni-Dolev et al. 2010), in which married men and women of reproductive age (N = 687) who already had one or more children of the same sex were interviewed. 42.6% of those who were planning in principle to have more children expressed a desire to choose the sex of the future children. These persons stated almost exclusively that they wanted to decide on the sex that their already born children did not have (ibid., 1021–1024). However, only 14.8% of them were willing to undergo IVF for this purpose. This study confirms the desire for a “balanced” gender composition of the family, but it also makes it clear that the necessary IVF treatment represents a high barrier to the actual realization of the wish.

However, as Bhatia (2010) has pointed out, the argument that gender preferences have lost their social relevance is often based on abstract juxtapositions of Western and non-Western cultures, as well as on a normative distinction between “good” and “bad” sex selection practices that does not stand up to critical scrutiny.²¹ Furthermore, she argues that in the case of family balancing genetic sex selection is based on and reinforces a “geneticization of gender”, since linking chromosomal variation with gender stereotypes is precisely the prerequisite for the establishment of this technology (see also Bhatia 2018a, 2018b; Whittaker 2011). In addition, the use of the term family balancing renders problematic a family in which the gender of the children is unequally distributed. The “balanced” family thus becomes the norm (see also Holm 2004, 31–33; see also Whittaker 2015).

Against the background of this controversy, numerous studies have examined the perceptions and attitudes of the population towards sex selection. While the use of PGD for medical purposes meets with moderate to high approval, the expansion of the application spectrum to non-medical sex selection is usually met with greater scepticism. Accordingly, this application of PGD was rejected by a majority in the already mentioned representative German survey (cf. Meister et al. 2005; Finck et al. 2006). In the US study by Kalfoglou, Scott, and Hudson (2005), as well, only a minority of respondents supported the use of PGD for non-medical sex selection: “Most participants who were against non-medical sex selection argued that the goal of PGD is to avoid disease and prevent the suffering of a child. Selecting for sex does neither.” (Ibid., 492; see also Klitzman et al. 2013; Abbate et al. 2014) By contrast, an Australian study (Katz et al. 2002, 1120) came to a different conclusion: 69% of the users of reproductive medicine technologies surveyed in this study (N = 121) saw no problems in using PGD for sex selection.

Of particular interest is a British study (Scully, Banks, and Shakespeare 2006), which investigated the underlying deliberative processes and argumentative structures by means of group discussions and interviews. The majority of the study participants opposed the legalization of non-medical sex choice. In their reasoning, they ultimately mobilized a specific notion of “good parents”: good parents are those who do not want to control their children, but rather protect them. Given this premise, the renunciation of sex selection does not contradict its reproductive autonomy; it is rather the condition of the same:

The identity of the good parent is constituted by this voluntary self-limitation. Parental autonomy can only operate within the limits set by this framework. Otherwise the choices, however freely made or in line with the individual’s life goals, do not foster the autonomy of a good parent, but of an individual failing to be an adequate one. (Ibid., 30)

The group discussions and interviews thus express an idea of relational autonomy that more closely resembles concepts of feminist ethics than the dominant Western liberal bioethics (see Wolf 1996; Haliburton 2014). In accordance with this conceptualization, the study participants understand the concrete societal conditions as an indispensable framework for individual decision-making processes.²²

5. Gender asymmetries and social inequalities

A PGD involves the biological parents of the child thus conceived in very different ways. Women not only have to bear the physical burden of IVF treatment alone, but are also more exposed than their partners to the mental stress of the procedure. Nevertheless, only a few empirical studies address these gender asymmetries, often implicitly documenting the unequal burden-sharing between the sexes (5.1).²³ Several empirical studies also document financial concerns and cost calculations that influence the decision-making processes of PGD users. These findings suggest that under certain institutional, legal and social conditions this technology has the potential to perpetuate socio-economic inequalities (5.2).

5.1. Between empowerment and distress: women’s experiences with PGD

As early as the beginning of the 1990s, there was evidence that individual reproductive history significantly influences the attitude of women towards PGD. Women who had opted to terminate a previous pregnancy after a prenatal diagnosis had indicated a risk of disease or disability of the future child often saw PGD as a suitable alternative to undergoing prenatal diagnosis. From their perspective, this method had the advantage of guaranteeing they would not have to suffer again the emotionally very stressful experience of the previous abortion (e.g. Pergament 1991; Palomba et al. 1994; Van Rij et al. 2011).

An Australian study (Karatas et al. 2010a) shows on the basis of qualitative interviews (N = 14) that women often experience PGD as simultaneously a relief and a stressful experience. Against the background of traumatic experiences with past miscarriages and abortions due to a genetic indication, many of the women interviewed saw PGD as a form of empowerment and control. At the same time, however, PGD itself also contributed to specific psychological distress. For example, respondents report frustrations after pregnancy failed to occur after the first PGD cycle, or fear after the embryo transfer when a pregnancy determination and embryo tests were pending. Particularly for women who had experienced problematic reproductive experiences in earlier biographical phases, it proved difficult to establish a relationship with their foetus after successful PGD treatment. Here, the phenomenon of a “tentative pregnancy” (Rothman 1986) was observed, which is already known from prenatal diagnosis. This is how one of the interviewees put it:

I felt sometimes I didn’t want to get too close in case there was something wrong and knowing how distressing it is to lose a pregnancy. And I also felt my husband didn’t get very close for the same reasons … But I mean I guess I bonded more in terms that I knew when it was kicking and would feel it … feel it more and … but I didn’t talk or play music or anything like that in terms of wanting to get close in terms of that. (Karatas et al. 2010a, 775; see also Karatas et al. 2010b)

Another source of concern is the fear of health risks for the women undergoing fertility treatment. A Dutch study (Derks-Smeets et al. 2014) documents, by means of group discussions and interviews, the perspective of women and their partners who were aware of the presence of a BRCA 1/2 mutation associated with breast and ovarian cancer and who had used either a PGD (N = 6), a PND (N = 4), or no genetic diagnosis (N = 8). The respondents expressed particular concern that hormone stimulation may increase the risk of cancer in women and ultimately contribute to the onset of the life-threatening disease. One of the couples stated: “That’s actually your biggest concern, right? That you bring a child into this world and then you fall ill yourself, due to the hormones … ” (ibid., 1109; see also Snowdon and Green 1997; Chamayou et al. 1998; Lavery et al. 2002; Dekeuwer and Bateman 2013).

The physical and psychological stress associated with PGD treatment can also have a negative impact on the relationship between the partners. A British study (Roberts and Franklin 2004; Franklin and Roberts 2006) which, in addition to interviews, also uses document analyses and participant observation to elaborate on gender inequalities in detail, provides deeper insights into the couple dynamics. The concerns of the participants related not only to the uncertainty of whether pregnancy would occur, but also to social life and career prospects. It becomes clear that women are not only physically burdened to a much greater extent by PGD treatment; the uncertainty and difficult planning of their careers and family life also affect them much more than their partners (cf. Roberts and Franklin 2004, 290–291). In addition, women are faced with the paradoxical need to “relax” in the face of this stressful situation in order to increase the chances of success of IVF treatment. An interviewee accentuates this gender asymmetry by comparing her burdens with those of her husband:

And this is something that I find really hard to come to terms with and no disrespect to Ben [her husband], but like it’s easier for him because he can go to work as normal, because he can forget about it! I think it’s a bit different for men. I’m the one that’s having the injections. I’m the one that’s getting the hormone treatment. He can go to work and kind of switch off a little bit, whereas I’m the one that’s thinking ‘I’ve got to be at home. Got to try and relax’. (Roberts and Franklin 2004, 291; emphasis in original)

The gender asymmetries elaborated go beyond the implications of PGD in the narrower sense; they structure the field of reproduction work as a whole. For example, women are currently also much more confronted with social expectation that they will have to provide the increased care work that could be necessary in the event of a (future) child’s illness or disability (e.g. Brekke and Nadim 2016). Against this backdrop, PGD could in turn appear as a technological option to avoid negative effects on women’s employment biography and social relations. Accordingly, the overarching forms of gender division of labor also shape the contexts of use of PGD and the couples’ decision parameters, so that the technology ultimately risks reproducing the prevailing gender inequalities (see Gammeltoft and Wahlberg 2014, 205–207).

5.2. Financial concerns, unequal access and cost calculations

PGD has the potential to perpetuate socio-economic inequalities if users have to pay privately for the costs of treatment. This is the case, for example, in the US, where this treatment is not covered by most health insurance companies. Drazba, Kelley, and Hershberger (2014) conducted qualitative interviews with 18 couples who were at increased risk for genetic disorders and either considered PGD, were undergoing appropriate treatment, or had already completed it at the time of the survey. The cost of PGD treatment was the greatest barrier to its use for the vast majority of couples. In the words of an interviewee: “[We] liked the idea of PGD from the start. I think the cost was what was holding us back. […] The whole decision was really just either spending the money or not.” (Ibid., 206) In addition, some expressed anger and disappointment at the fact that the costs of treatment – unlike the expenses for the treatment of the diseases in question – were not covered by their insurance. Despite these financial concerns, however, the interviewees’ hope of being able to rule out the transmission of the disease in question by means of PGD ultimately prevailed (see also Rubin et al. 2014).

Karatas et al. (2010c) interviewed 50 women who underwent a PGD in Australia, where the costs also have to be covered by those undergoing the procedure. Although almost all the study participants had an above-average household income, they were concerned about the financial burden of the treatment. In another Australian questionnaire survey (Katz et al. 2002), by contrast, the majority of couples examined who had a PGD or IVF performed stated that they had no financial worries about the cost of treatment. However, there was no information on the socio-economic background of the interviewees. Remarkably, dimensions of social inequality play only a minor role in the studies. We are not aware of any study that investigates possible relationships between ethnicity and the use of PGD. Only a few studies highlight the correlation between the level of education and attitudes towards this technological option, although the results partly point to a negative correlation between level of education and acceptance of this procedure (see e.g. Meister et al. 2005; Borkenhagen et al. 2007).²⁴

However, the high cost of PGD treatment does not only have the potential to effectively exclude people with lower and middle incomes from using it if these expenses have to be covered privately. Concerns about the resulting costs can arise even if they are paid for by a public health insurance system. Järvholm, Broberg, and Thurin-Kjellberg (2014) interviewed 19 Swedish couples who had visited a clinic for the purpose of PGD treatment. For some, the fact that their costs would be borne by health insurers raised the question of whether this expenditure of social resources could be justified in their case. One of the women interviewed expressed this weighing-up process as follows:

And then you can have feelings about PGD that … there are other people with cancer and things like that – so you can feel a bit guilty in some way that this [PGD] is such an expensive treatment … but the society has several layers of problems and if people are feeling well, then society gets better. You can’t just only help the patients with cancer or heart diseases. (Ibid., 65)

In fact, the economic rationality expressed in this interview passage is already shaping research and health policies today. Corresponding cost calculations are an essential basis for the legitimization and delegitimization of health policy decisions and research policy priorities (e.g. Rothgang and Preuss 2008). In health economics so-called cost-effectiveness models are circulating that assess the economic potential of a comprehensive application of PGD and contrast the therapy costs of diseases and disabilities that could be prevented in advance by means of genetic and reproductive medicine procedures (cf. Clarke 2010, 372–377). Tur-Kaspa et al. (2010) carried out an exemplary cost–benefit analysis of IVF/PGD treatments for the USA using the example of autosomal recessive inherited cystic fibrosis (CF). The starting point is the consideration that couples who know about their status as carriers of a genetic variation often do not decide on termination of a pregnancy after a prenatal diagnostic CF diagnosis. The authors prefer a preventive strategy based on PGD, in which the decision for or against abortion by the preceding extracorporeal embryo selection is not necessary (ibid., 187). In their calculations, they conclude that a corresponding national prevention program could save approximately $ 33.3 billion in 37 years, the underlying statistical life expectancy of a CF patient.

In summary, offering IVF–PGD to all CF carrier couples who wish to conceive without facing the dilemma of possible pregnancy termination or raising a sick child is highly cost effective and will save billions of dollars in direct health expenditures. Delivering a healthy baby instead of one affected with CF means avoiding not only direct medical treatment expenses, but also avoiding the significant loss of productivity and quality of life for CF patients and their caregivers over a lifetime. (Ibid., 195)

Some countries (e.g. Saudi Arabia and Iran) in which the autosomal recessive blood disease beta thalassemia is particularly common have now introduced mandatory screening programs for people of reproductive age prior to marriage (cf. Zlotogara 2009, 250–251). The carriers of the genetic variation detected in this way are also being offered PGD in order to prevent the inheritance of the disease and thus reduce the relatively high treatment costs (cf. German Ethics Council 2011, 121).

Such economic rationality is also found in the narratives of physicians and people who are themselves affected by disease risks. This is documented in the study already cited by Kalfoglou, Scott, and Hudson (2005). Both the patients interviewed and the reproductive medical experts saw PGD as a way to reduce medical treatment costs in the short term and to “relieve” society of the burden of genetic diseases in the long term. In the following interview passage, the hope of being able to “eradicate” genetic diseases in the future by means of selective reproductive technologies is particularly striking:

I am hoping that, like vaccination, PGD will eliminate some of these ravaging diseases from occur[ing] … . [M]aybe fifty years in the future, we will look upon genetics, manipulation of embryos, as the earliest form of perinatal health care. (Ibid., 489)

PGD appears to be comparable here to immunological vaccination, which in the 19th and 20th centuries helped to eliminate numerous viral infectious diseases, and as a pioneer of “perinatal health care”, in which genetic diseases can be combated even earlier and more effectively in the future.

6. Responsibilization and discrimination

Social-scientific research not only shows that the social dynamics of PGD cannot be separated from risk management techniques or prevention practices, but also suggests that its technological availability structures expectations of “responsible” health or reproductive behavior (6.1). These expectations can in turn evoke fears and negative categorizations, and may thus provide a fertile breeding ground for stigma and discrimination. The latter also affects individuals currently living with chronic diseases and disabilities (6.2).

6.1. Normative expectations and the dynamics of genetic responsibility

In a study on the societal implications of prenatal and genetic diagnosis, the sociologist Elisabeth Beck-Gernsheim (1994, 325) pointed to an “expansion of responsibility” that already existed in the 1990s. She argued that this term was increasingly being used “in the direction of a qualitative selection, already starting before birth, perhaps even before conception” (ibid., 326, authors’ translation). From this perspective, it appears to be “responsible” to avoid the birth of presumably disabled or sick children by abandoning the wish to have children, or by terminating pregnancy after a medical finding (see also Beck-Gernsheim 1996; Ruhl 1999; Hallowell 1999).

In the empirical studies, there are a number of indications that PGD perpetuates this concept of “genetic responsibility” (Kollek and Lemke 2008, 223–287; see also Leefmann, Schaper, and Schicktanz 2017). The survey by Wüstner and Heinze (2007) shows that a significant proportion of German and Japanese students (N = 720) articulates such a normative expectation. They overwhelmingly supported access to PGD in this comparative cultural survey (Germany: 51.3%, Japan: 75.2%). A quarter of German (25.1%) and 15% of Japanese students also believed that PGD should be mandatory for couples with genetic disease risks.²⁵

Zeiler (2004) argues that individuals living with a genetic disease or genetic disease risks are confronted with perceptions of specific reproductive responsibility. The geneticists and gynaecologists from the UK, Italy and Sweden who were interviewed in the study reported that some couples affected by disease risks experience a non-use of this technology as culpable and irresponsible, as one doctor makes clear on the basis of her own consulting experience: “Simply because this technology is there, many couples seem to feel that they must take it, that they are denying their future children [something] if they don’t take it.” (Ibid., 181) PGD thus seems to mobilize a normative dynamic which Zeiler (ibid.) understands as an “internalized technological imperative” (see also Hershberger and Pierce 2010; Kalfoglou, Scott, and Hudson 2005).

Many empirical studies on the motives of a possible use of PND or PGD show that those affected usually seek to structure their reproductive decisions in such a way that, as far as possible, they do not pass on genetic disease risks. In this way, they try not only to prevent their future children from suffering from health problems but also to counteract feelings of guilt and possible rebukes. This is exemplified in the study by Derks-Smeets et al. (2014) based on focus groups and interviews with couples (N = 18) carrying BRCA 1/2 who considered PGD and/or PND subsequent to professional counseling on reproductive choices (see 5.1). In considering PGD, the couples most frequently articulated the wish to protect the future child from physical and psychological burdens, independent of their final decision for or against one of these technological options. Half of the couples also felt that, given their knowledge of the hereditary nature of the disorder on the one hand and the technological options on the other, it was their moral duty to spare their future children suffering. Accordingly one of the interviewees, who had decided against both PGD and PND, anticipated that she would develop feelings of guilt if her child also had an increased genetic risk of breast and ovarian cancer:

What I was afraid of myself, or still am actually, are those feelings of guilt. They might not be so relevant now, but in about twenty or thirty years when my child would go for a DNA test … Imagine it will be positive, then I would have to relive this all over again. And then, I would tell myself: it’s your own fault and you could have prevented this …  (ibid., 1107).

The authors thus conclude that the “mere possibility of PGD and PND can cause an emotional burden once people become aware and choose to refrain from it” (ibid., 1110).²⁶

It has to be noted, though, that many dimensions of this discourse of “genetic responsibility” are part of a more comprehensive prevention logic, which is articulated not only with regard to PGD but also in the context of prenatal diagnosis and predictive genetic knowledge (e.g. Freedman 1998; Arribas-Aylon, Sarangi, and Clarke 2011; Lemke 2013, 79–106). However, there is one important feature in the application of PGD that opens up a new field of normative expectations and responsibilities: the question of how to deal with those embryos that are not immediately implanted in the uterus. The concerns and self-doubts associated with the practice of storing “redundant” embryos are shown in the Australian study by Karatas et al. (2010a). The moral ambivalences articulated by the women interviewed initially referred to the embryos that will be available for further treatment in the future. Some participants experienced the obligation to “give” them the “chance” to live (ibid., 775; see also Millbank 2017). However, the respondents felt that the question of what to do with embryos that were affected by the genetic condition under investigation was a particular burden. One respondent described this feeling as follows:

I can remember at the clinic asking very clearly what do you do with the embryos that are (chromosomally) unbalanced and she said discard them. And that had this huge clunk emotionally with me because I didn’t draw a line between an embryo with unbalanced translocation and my daughter who had died who I loved. (Ibid., 775)

6.2. Impact on individuals living with disabilities and chronic diseases

In addition to the change in social and institutional expectations outlined above, PGD also has consequences for individuals and families affected by disabilities and chronic diseases. For example, it has been critically noted that the selection of embryos might also imply a moral value judgement about the life of those who are carriers of the trait in question (e.g. Roberts 2002, 5). The technological option of PGD is therefore often understood by representatives of the disability rights movement, but also by some bioethicists, as humiliating and discriminating (cf. Petersen 2005; Krones 2008).

Some of the individuals whose disease or disability is to be prevented by PGD actually associate this practice closely with a judgement on their life value. While the majority of women interviewed by Quinn et al. (2009a) stated that PGD provided options for them to pursue parenthood instead of refraining from having a biological child due to fear of transmitting the mutation, the study also shows that the use of PGD, at least indirectly, affects the self-interpretations and identity constructions of people who fall within the scope of this technological option. One of the respondents expresses this link when contrasting her parents’ reproduction decision with current diagnostic possibilities: “These things (breast/ovarian cancer) can be diagnosed and treated early. I would hate to think my parents would have chosen not to have me if they knew I carried the gene for breast cancer.” (Ibid., 445)²⁷ Another person opposes the use of PGD to select the BRCA 1/2 mutation. She de-dramatizes her own disease risk by referring to the normality of genetic variation: “I do not think a woman should avoid having children if she is BRCA positive. Everyone has genetic mutations, those of us who are BRCA positive just happen to know what ours means.” (Ibid.)

The reproductive medical health experts and PGD users interviewed by Kalfoglou, Scott, and Hudson (2005) expressed concern that the availability of PGD could further impair the social acceptance of people with disabilities or chronic diseases in the future: “(S)ociety might become less tolerant of the disabled and their parents and that, as a result, couples may feel pressured to use PGD to avoid having an affected child” (ibid.: 491). In addition, society’s willingness to cover medical treatment costs for illnesses that are considered to be preventable could also diminish further, and health insurance companies could refuse to provide cover to those affected. This in turn would shift the parameters of the decision-making process for or against PGD in such a way that the uncertainties and fears associated with a negative decision would intensify (see also Vergeer, van Balen, and Ketting 1998).

In principle, however, PGD also allows the targeted selection of specific characteristics that are commonly regarded as diseases or disabilities. Embryos with the desired genetic variation would not be excluded in these cases, but rather selected and transferred to the uterus. This selection interest, which goes against the grain of hegemonic concepts of normality and the logic of prevention, can be exemplified by a lesbian couple who in 2002 wanted a child who – like themselves – would be deaf. They fulfilled their wish by selecting a sperm donor who was affected by the same hereditary form of deafness (cf. Savulescu 2002; Dennis 2004). In this specific case, no PGD was used; in principle, however, the procedure would have presented a much more precise alternative to the selection of a sperm donor, so that PGD is also likely to encounter interest in this area. Mand et al. (2009) investigated the attitudes and experiences of (adult) children of deaf parents who themselves had no hearing impairment. All the study participants (N = 66) saw facets of a distinct culture in deafness and disapproved of its characterization as a disability. A clear majority stated that they had no preference for or against deafness among their own children (72.3%) and rejected the use of PGD/PND for this purpose (60%). Those who did not oppose these diagnostic procedures as a matter of principle were more likely to accept their use in situations where the couples choose a child with the same characteristics as the potential parents.²⁸

7. Conclusion

Our literature review on social dimensions of PGD documents considerable ambivalence about the use of this technology, not only amongst the general public but also amongst professionals and affected individuals. It presents sometimes contradictory empirical evidence that indicates not only divergent attitudes and perspectives, but also fundamental affective and moral ambiguities. We have traced these ambiguities in three important areas of concern that we identified in the literature.

First, the empirical studies provide different assessments of PGD based on the expansion and transformation of its fields of application. While PGD was initially almost exclusively framed in scientific and public discourse as a method for preventing severe and untreatable genetic diseases and disabilities, it is now also used to select genetic variations that are associated with treatable illnesses and/or only increased disease risks, so it is uncertain whether those affected will ever suffer from this disorder. The empirical surveys show that the attitudes of the general population, as well as affected individuals, towards these different purposes of PGD vary significantly. While there is strong support in the general public for employing PGD in order to avoid severe and fatal diseases, the approval rates are considerably lower when it is a question of using PGD to prevent late-onset diseases and disease risks. It remains unclear if the majority of individuals affected by late-onset diseases or disease risk approve of this selection purpose, as the evidence is contradictory. While some studies show significant support, there is also evidence of rejection amongst this group. In addition, a considerable number of those individuals who support the application of PGD for late-onset diseases and disease risks do not want to use this technology themselves.

There is also contradictory evidence regarding the applications of PGD that exceed the narrower medical framework (e.g. HLA typing) or do not have any connection to health and illness (e.g. social sex selection). The vast majority of surveys show strong disapproval of social sex selection, which corresponds to the restrictive regulations in Europe. Nevertheless, a few studies point to acceptance of and even significant demand for social sex selection, at least in particular countries (e.g. Lebanon, Saudi Arabia) – a practice that has been criticized by feminist scholars, on the grounds that this application of PGD risks reproducing and reinforcing sexist discrimination and the “geneticization of gender”.

This account corresponds with the second area of concern documented by the literature review: PGD risks reproducing existing gender asymmetries and social inequalities. It not only places a much greater psychological and physical burden on women than their partners, but might also have negative consequences on their social life and professional careers. Nonetheless, many women prefer PGD as an alternative to prenatal diagnosis and pregnancy termination after a medical finding, especially if they have had traumatic experiences during previous pregnancies. At the same time, it should be noted that it is primarily women who, in the context of the present gendered division of labor, are confronted with the expectation of providing the required level of care work in the case of illness or disability of the child. Women thus seem to experience PGD as both a form of empowerment and a cause of distress. In addition, empirical studies have provided evidence that users of PGD articulate financial worries related to the availability of this technological option. Access to PGD might depend on the socio-economic status of the users if the costs of treatment are not covered by the public health insurance system. Simultaneously, empirical investigations show that – if PGD is covered by public health care – some individuals have the concern that the costs of that treatment might not be justified in their case. Furthermore, there are indications that societal pressure on potential users could increase if the implementation of PGD proves to be cost-effective in relation to medical treatment. In the course of the ongoing restructuring of public health systems in many countries, corresponding economic calculations have long been included in social and health policy decision-making processes.

This development goes along with a trend towards the individualization of responsibility, according to which the use of PGD to “prevent” diseases is considered morally “the right thing to do”. This process of responsibilization forms the third area of concern. In the material analyzed, not only the reports by medical experts but also the narratives of those affected suggest that individuals and families who are aware of an increased genetic risk of disease are already confronted with corresponding moral expectations. In addition, there is empirical evidence of moral concerns and self-doubts related to the question of what to do with embryos that are produced during IVF but not transferred into the uterus. The selection of genetic variations associated with specific disorders or disabilities can also have implications for individuals and families living with them today. In the interview material the fear is manifest that, with the technological feasibility of their “avoidance”, the social acceptance of genetically caused diseases and disabilities will diminish and the willingness to accept the costs of therapy and care in solidarity, as well as the necessary investment in medical research projects, could decrease.

Whether and how the socio-technical tendencies outlined here continue depends not least on the further development of IVF. This is not only the “technological platform” of PGD, but also the infrastructural prerequisite for research into regenerative medicine, embryonic stem cells, and somatic cell nuclear transfer (Franklin 2013b, 36–38). It thus links PGD with other fields of the life sciences, so that their dynamics indirectly determine the future prospects of PGD. The future transformation of genetic analysis and genome editing is furthermore of particular relevance. In recent years, sequencing technologies have emerged that enable a much faster and, at the same time, more cost-effective analysis.²⁹ These procedures have now reached a stage that enables their use not only in basic medical research but also in everyday clinical practice. In the area of PGD, differentiated gene panels might increasingly take the place of single tests. Such panels could cover not only the dispositions of a large number of widespread polygenetic diseases (e.g. diabetes) but also heterozygous carriers for recessive inherited conditions (Campbell and Porteous 2018).³⁰ Should their use become established in the context of PGD, it would increasingly take on the character of a non-specific screening strategy moving further and further away from the original goal of preventing serious and untreatable disorders (see Hens et al. 2013).

In any case, our literature review shows that the social dimensions of PGD cannot be reduced to the question of whether or not it represents a new form of eugenics. Rather, it contributes to a different and more comprehensive assessment of PGD that brings to light users’ concerns and problematic socio-technical tendencies. The findings point to the many ambivalences and uncertainties that go along with the use of PGD. It may be experienced simultaneously as a means of empowerment and relief on the one hand and as a source of psychological stress and health risk on the other. We have also identified significant issues that have largely escaped scholarly attention. The vast majority of empirical studies have only indirectly, if at all, recorded socio-economic and gendered inequalities. The latter is especially surprising, as feminist research has repeatedly exposed the gender inequalities and discriminatory effects of the use of genetic and reproductive technologies (e.g. Haraway 1997; Kuhlmann and Kollek 2002; Thompson 2005), and the “medicalization of the female body” (e.g. Duden 1991; Reed and Saukko 2010) represents an important topos of criticism. Studies that systematically address these dimensions, particularly on a global scale, are urgently needed.

There is also a substantial lack of studies that investigate the use of PGD to increase the chances of success of IVF. While the ESHRE data suggest that the majority of the PGD treatments are now performed to improve the likelihood of establishing a pregnancy, we did not find any empirical study that engages with the social aspects of this screening procedure. Finally, the existing literature focuses almost exclusively on the use of PGD in Western countries. Since this technology is only one element of “selective reproductive assemblages” (Wahlberg and Gammeltoft 2018, 13) consisting of particular technologies, sites, people and regulations, we strongly advocate comparative investigations that trace PGD globally in different countries and cultural contexts.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was supported by Federal Office of Public Health of the Swiss Confederation [grant number 604.0001-460].

Notes

1 For the history of eugenics see Weingart, Kroll, and Bayertz (1992), Kevles (1995), Paul (1998), and Bashford and Levine (2010). For the relationship between classical eugenics and current practices in reproductive medicine and human genetics, see Lemke (2016).

2 The study was initiated by the Federal Office of Public Health of the Swiss Confederation. In the course of a revision of the Swiss Law on Reproductive Medicine (FMedG), in late 2015 the Federal Office commissioned – in addition to ethical and jurisprudential assessments – an expert opinion on the societal implications of preimplantation genetic diagnosis, which was completed in November 2016 (Lemke and Rüppel 2017). This article provides an overview of the key findings of this expertise.

3 It is worth stressing that the empirical studies reviewed in this paper are not entirely independent of regulatory regimes; on the contrary, attitudinal surveys and opinion polls in particular, which are often conducted under the label of “ethical, legal and social implications” (ELSI), form part of current styles of regulation in science and technology. These studies are often criticized on the grounds that they serve as source of legitimization and acceptance and a way of channeling potential public resistance (see Jasanoff 1995; Lopez and Robertson 2007; Rehmann-Sutter 2011).

4 In addition, PGD was also approved for sex chromosome-linked diseases, and, under certain conditions, for HLA typing.

5 “It is expected that PGD will be available only where there is a significant risk of a serious genetic condition being present in the embryo. […] The seriousness of the condition is expected to be a matter for discussion between the people seeking treatment and the clinical team.” (HFEA Guide CH(03)04); http://hfeaarchive.uksouth.cloudapp.azure.com/www.hfea.gov.uk/2686.html (last accessed 29 January 2018)

6 The list can be found on the following website: https://www.hfea.gov.uk/PGD-conditions?page=1 (Last accessed 29 January 2018).

7 The regulation of PGD in the United States is characterized by a “laissez-faire approach” (Bayefsky 2016, 42). It is also employed for highly controversial purposes such as non-medical (elective) sex selection (see Bayefsky 2015; Bayefsky and Jennings 2015, 11–12).

8 This number includes PGD and Preimplantation Genetic Screening (PGS) cycles (see 4.3.). When we refer here and in the following to the data of the ESHRE, the reference value is always the number of PGD cycles that have reached at least the stage of oocyte retrieval.

9 The first data set of the ESHRE (January 1997 to September 1998) documents 366 cycles (ESHRE PGD Consortium Steering Committee 1999); the latest data set (January 2011 to December 2012) documents 11,637 cycles (De Ryke et al. 2017). However, it has to be noted that the number of ART centers participating in the PGD Consortium increased from 16 to 71 during the same period.

10 They included adults affected by a genetic mutation associated with hereditary breast and ovarian cancer, Lynch syndrome, familial adenomatous polyposis, and multiple endocrine neoplasia type 1 or 2.

11 Rubin et al. (2014) traced the complex decision-making processes of women with a genetic risk for hereditary breast and ovarian cancer (BRCA 1/2) on the basis of qualitative interviews. It becomes clear that these cannot be understood without the specific family dynamics, the influence of experts, and contextual factors (see also Brüninghaus 2011; Heyen 2011; Hurley et al. 2012).

12 The structured questionnaire, however, did not record the genetic knowledge of respondents. The authors argue that individuals who were affected by or at risk for a single gene disorder might also accept a healthy carrier because they “value their own genetic status” (Katz et al. 2002, 1121).

13 IVF was initially established and is generally seen as an infertility treatment (Franklin 2013a). Approximately 186 million people, or rather between 8 and 12% of reproductive-aged couples worldwide, suffer from infertility, defined as “a disease characterized by the failure to establish a clinical pregnancy after 12 months of regular, unprotected sexual intercourse or due to an impairment of a person’s capacity to reproduce, either as an individual or with his/her partner” (Borght and Wyns 2018, 1).

14 Between 1997 and 2009 their share in the total number of PGD applications increased annually up to 58%; since then its proportion has decreased slightly to 52%. De Ryke et al. (2017, 1991) attribute this decrease to scientific doubts as to whether such screening is beneficial (see also Gianaroli et al. 2012, 1388–1389). According to a relevant meta-analysis (Checa et al. 2009), when compared to simple IVF a combination of IVF and PGD shows even a lower success rate (see also Mastenbroek et al. 2007). Proponents of screening counter this position by arguing that its failure is ultimately due to (until now) inadequate technology (see Harper et al. 1995; Vanneste et al. 2009; Munné, Grifo, and Wells 2016).

15 In the first ten years covered (1997–2007), the woman’s age was identified as an indication in almost 50% of cases (Harper et al. 2012, 241). An age of over 37 years is regarded as high (cf. Harper et al. 2010, 821).

16 Sex selection in the context of PGD can also be used to rule out the inheritance of a sex-related disease (e.g. an X-linked recessive inherited disorder such as Duchenne muscular dystrophy) (cf. Kollek 2000, 27–28).

17 In Israel, non-medical sex selection is prohibited in principle, but couples and single women can legally apply to the “Israeli National Committee for Sex Selection by PGD for Non-Medical Reasons” to grant an exception. This committee includes, among others, physicians, ethicists, a lawyer, a clergyman and a psychologist appointed by the Director General of the Ministry of Health. It can approve applications if the applicants are married, have at least four children of the same sex and none of the other and “there is real and immanent risk of significant damage to the mental health of one or both parents, or to the expected child, if the procedure is not conducted.” (Pessach et al. 2014, 2).

18 In a retrospective cohort study, Farra et al. (2014) documented 192 PGD cycles performed in Lebanon between 2004 and 2007. In 96.3% of these cases (185 cycles) non-medical sex selection was mentioned as indication. In 174 cycles the PGD users chose the male gender.

19 It should be noted, however, that a large part of the data came from a US-based center that offered PGD as well as sperm selection using the MicroSort^® method. The accuracy of this procedure is higher in determining an X chromosome than a Y chromosome. Accordingly, the selection of predominantly male embryos by PGD could also be attributed to the fact that couples who wanted a female child opted for the MicroSort^® method as an alternative, and thus dropped out of the statistics (cf. Harper et al. 2012, 243).

20 Another contentious issue is the use of PGD for the selection of embryos to which intersexuality is attributed at the chromosomal level. This also transcends the narrower medical framework, as in many cases no health restrictions exist (for this debate, see Gupta and Freeman 2013; Haramia 2013; Trafimow 2013; Sparrow 2013a, 2013b).

21 A social preference in reproductive decision-making for the male gender can be observed not only in India (Dey and Chaudhuri 2009; Malhi et al. 1999) and China (Chan et al. 2002; Chi et al. 2013); a few studies conducted in Western societies point in this direction as well (e.g. Van Balen 2006). In a US questionnaire study, Sensibaugh and Yarab (1997) found a tendency to favor a majority of boys in the case of an unbalanced gender relationship within a family. Data from the Israeli Ministry of Health show that PGD for sex selection is more in demand from couples who have already had several girls and want a son than vice versa (see Hashiloni-Dolev et al. 2010, 1024).

22 The assessment of the societal consequences of sex selection is also important because it is interpreted in public discourse as a precursor to other non-medical fields of application. Media articles often outline a future scenario in which PGD is used for the selection of further physical and psychological characteristics and is thus increasingly developing into a technology for the production of “designer babies” (e.g. Brandenburg 2011; see also Roberts 2002, 10–13; Jones and McMahon 2003, 271–274; Wüstner 2006, 93). Franklin and Roberts (Roberts and Franklin 2004; Franklin and Roberts 2006, 1–24) showed in an ethnographic study that the users of PGD are facing this scenario. They defended themselves against such ideas by emphasizing that they did not want to produce “designer babies”, but only to prevent serious illness or disability.

23 The reviewed studies focus on heterosexual couples to address the gender asymmetries, as legal access to PGD in many countries is restricted to this group of individuals (see 3.).

24 The study by Finck et al. (2006), however, did not find any clear correlation.

25 The item in question was: “PGD should be made compulsory for couples who are at risk to pass on gene abnormalities.” The average age of the two groups differed: 88% of the German students were between 21 and 30 years of age; 71.6% of the Japanese students were 20 years or younger (Wüstner and Heinze 2007, 8).

26 This sense of responsibility can also be found in couples affected by other genetic conditions. The study by Klitzman et al. (2007) focuses on individuals where Huntington’s disease (HD) had occurred in the family. The material documents how the idea of not only moral but also social responsibility materializes in the desire to prevent the birth of a child with the HD-typical genetic variation. As one respondent put it: “It seems unfair to ask people to forgo childbearing, but at the same time, you’re burdening society with people that are going to get sick.” (ibid., 358; see also Decruyenaere et al. 2007).

27 A significant proportion of respondents also feared that as a result of this technological selection option, the search for an effective therapy for breast and ovarian cancer would lose relevance and that necessary investments in corresponding medical research projects could decrease (Quinn et al. 2009a, 445).

28 No respondent voiced a preference for a child with deafness; 27.7% wished for a child without hearing impairment. In a British questionnaire study (Middleton, Hewison, and Mueller 2001), 644 deaf people were questioned about their attitudes towards a genetic test for hereditary deafness in the context of PND. It concluded that about 21% of those affected would seriously consider such a test; as many as two percent preferred a child with deafness and – should this attribute be absent – would consider abortion.

29 Another innovation boost in terms of the costs and duration of genetic analysis is so-called karyomapping, which was successfully performed for the first time in December 2015 (see Natesan et al. 2014; Thornhill et al. 2015).

30 For the use of PGD in healthy carriers, see Katz et al. (2002). In future, PGD could be used subsequent to an expanded carrier screening that is increasingly available (see Wehling 2014; Vaz-de-Macedo and Harper 2017).