Clinician views and experiences of non‐invasive prenatal genetic screening tests in Australia

Background Non‐invasive prenatal screening (NIPS) is being increasingly used by expectant parents. Much provision of this test in Australia is occurring in clinical settings where specialised genetic counselling is unavailable, such as general practice. Potential psychosocial consequences from this kind of prenatal genetic screening remain largely unexplored. Aims To explore clinicians' experiences with NIPS for aneuploidy, their perspectives of the benefits and harms of NIPS, clinicians' information needs, and their perceptions of the needs of expectant parents. Materials and Methods Qualitative, semi‐structured interviews with 17 health professionals (clinical geneticists, obstetricians, genetic counsellors and general practitioners) who request and counsel for NIPS in Australian hospital and private practice settings, conducted between June 2019 and February 2020. Results Five themes were identified relating to clinicians' perceptions and experiences of NIPS in their practice: perceived benefits of NIPS, perceived harms of NIPS (with two subthemes: clinical harms and psychosocial harms), financial and equity‐related concerns, counselling as a protective buffer against perceived harms, and clinicians' unmet education needs. While clinicians view NIPS as a useful and high‐quality screening test, especially for detection of common trisomies, many participants had concerns about how NIPS has been implemented in practice, particularly the quality (and often absence) of pre‐/post‐test counselling and the routinisation of testing for sex chromosome aneuploidies, microdeletion and microduplication syndromes. Conclusion These findings support the need for targeted clinician training around NIPS, and for a shared decision‐making approach to support expectant parents' autonomous decisions about NIPS.


INTRODUCTION
Non-invasive prenatal screening (NIPS), commercially available in Australia since 2012, uses analysis of placentally derived cell-free DNA in maternal plasma to detect genetic conditions in the fetus.
NIPS was initially used to screen for trisomy 21 (T21: Down syndrome), T18 (Edwards syndrome), T13 (Patau syndrome), but is now also used to screen for a larger number of anomalies, including sex chromosome aneuploidies (SCA) and sub-chromosomal abnormalities such as microdeletion and microduplication syndromes (MMS). 1 Non-invasive prenatal screening has high accuracy for the three most common trisomies, with detection (sensitivity) in singleton pregnancies of >99% of fetuses with T21 and T13, and 98% with T18. The combined false positive rate (1-specificity) is 0.13%. 2 Screening for SCA was added initially for monosomy X, before other SCA, including 47,XXX, 47,XXY, and 47,XYY were also added. 3 Testing for selected MMS has been clinically available since 2013. 3 Diagnostic accuracy for SCA and sub-chromosomal conditions is not as high as the more common trisomies. 1,2,4,5 Consequently, screening for these conditions may cause a larger number of false positive results, especially if widely used in the pregnant population. Additionally, the phenotype of many of these conditions, especially the SCA, spans a wide range of neurocognitive and physical symptoms that vary in severity, with many individuals remaining undiagnosed in their lifetime. 6 Despite recommendations from professional bodies to limit the genetic conditions routinely screened for, 7,8 the boundaries of NIPS continue to expand through new technology, including the burgeoning use of genome-wide testing. 9,10 There are two models for implementing NIPS: a first-tier test offered to all pregnant people, or a second-tier test offered only if there is an increased chance result on the combined first trimester screening (cFTS) test (11)(12)(13) weeks nuchal translucency ultrasound and serum markers). 3,9,11 Both models are used in Australia, 1 with provision largely on a user-pays basis given NIPS is not funded by Medicare. Despite limited public funding, approximately 25-30% of pregnant people in Australia currently undergo NIPS 12 (50-75% for private obstetric care). 13 The ad hoc, commercially driven implementation of NIPS in Australia 1 and rapid evolution of new testing options and technologies present challenges for clinicians, introducing more complexity into clinical discussions and decision-making around prenatal screening. 10,14 Data are lacking on Australian clinician views and experiences of NIPS, and particularly their experiences with issues that may arise when expectant parents receive a high chance or other unexpected result. To address this gap, we aimed to explore clinicians' experiences with NIPS, with a particular focus on high chance and unexpected results, benefits and harms, information needs, and clinicians' perceptions of expectant parents' needs.

MATERIALS AND METHODS
We used a qualitative approach to capture clinicians' personal experiences and perspectives of NIPS in relation to their clinical practice. We purposively sampled clinicians who use NIPS (clinical geneticists working in public hospitals, obstetricians, (maternalfetal medicine (MFM) specialists) and genetic counsellors working in public or private practice, general practitioners (GPs) in shared care antenatal programs, and non-shared care GPs), with particular focus on clinicians with greater depth of experience in dealing with high chance and unexpected NIPS results. While we originally intended to recruit more participants without specialist knowledge of NIPS, our initial interviews with clinicians in this group demonstrated limited experience with the high chance/ unexpected results that were the focus of this study. Therefore, our sample was predominantly (but not entirely) made up of clinicians with high level knowledge about NIPS and genetic testing. Clinicians were recruited via email invitations, professional newsletters (Primary Health Networks and antenatal shared care programs), and author networks, from three public hospitals in New South Wales (NSW) and five private (general practice and ultrasound) practices in NSW and Queensland. Ethics approval was obtained from South Eastern Sydney Local Health District (18/283) and the University of Sydney (2019/243). Written consent was obtained pre-interview.
The semi-structured interview guide (Appendix S1) was developed by the multi-disciplinary research team with expertise in MFM, clinical genetics, genetic counselling, bioethics, health policy, psychology, and clinical epidemiology, drawing on relevant literature around the potential psychosocial consequences of NIPS, and piloted with clinicians with and without specialised genetics knowledge (GPs and genetic counsellors). The interview guide was designed to be used by a non-clinician interviewer and covered a broad range of clinical situations and experiences that may be discussed by participants, with the actual questions asked determined by the participants' previous responses. All interviews were conducted by telephone by a trained qualitative researcher (SM), audio-recorded and transcribed verbatim. Transcriptions were cross-checked to ensure data integrity. Data collection ceased when thematic saturation was reached (ie once newly collected data had become broadly repetitive of previously collected data in regard to the developing thematic framework). 15 Data analysis commenced concurrently with data collection to facilitate decisions about purposive sampling and saturation.
We conducted an inductive (data-driven) thematic analysis, using the framework analysis method 16 (Box 1). Rigour was addressed through an iterative process of constant data comparison, double coding during initial framework development, regular analysis discussions with all authors (including practising clinicians -a midwife, genetic counsellor, and MFM specialist), and cross-checking data against thematic findings for consistency by SM and NJ.

Seventeen clinicians were interviewed between June 2019 and
February 2020, including five obstetricians (MFM specialists), five genetic counsellors, four GPs and three clinical geneticists ( Table 1).
The majority of participants had a high level of knowledge of NIPS.
We identified five themes: perceived benefits of NIPS, perceived harms of NIPS (subthemes: clinical and psychosocial harms), financial and equity-related concerns, counselling as a protective buffer against perceived harms, and clinicians' unmet education needs around NIPS (subtheme: relationship between education gaps and commercially driven NIPS). Table 2 presents a summary of perceived benefits and harms; Table 3 provides illustrative quotes from participant interviews.

Perceived benefits of NIPS
Participants perceived NIPS allowed earlier, and more accurate, detection of increased chance for common autosomal aneuploidies (T21, T18 and T13), with less need for invasive tests and possible risk of miscarriage. They perceived that NIPS gave expectant parents reassurance, the ability to make informed decisions about their pregnancy, or to prepare to have a baby who may have health complications. Finding out biological sex earlier was thought to facilitate early bonding. While some participants had concerns about sex-selective termination, none had encountered this in practice.

Perceived harms of NIPS
Many participants perceived potential clinical and psychosocial harms from NIPS, particularly regarding screening for SCA and MMS. The GPs interviewed for this study had limited experience with high chance NIPS results and did not often identify or voice concerns about potential harms of NIPS compared to genetic counsellors, obstetricians, and clinical geneticists.

Clinical harms
Perceived clinical harms of NIPS included changed screening behaviour when expectant parents erroneously thought NIPS fully replaced the 12-13-weeks (first trimester) ultrasound, potentially resulting in fetal structural abnormalities being identified at later gestation than had the first trimester ultrasound taken place.
Some participants attributed this to false reassurance from a low chance NIPS result because NIPS was perceived as a diagnostic test for a broad range of conditions, rather than a screening test for specific conditions.

Relationship between education gaps and commercially driven NIPS
Some participants drew a connection between the need for more consistent information/education for clinicians and the commer-   'I tell them and I remind them again that you need the scan as well. But often I find that they are just so fixated on the blood test and they think the blood test will tell them everything.' (ID16 GP) 'A number of times we find something later in the baby and they say, "But I had a NIPT †, how could you be thinking about an amnio now?" (…) trying to get that concept across that they have not just ticked off everything to do with genetics.' (ID11 obstetrician) 'We've seen so many so-called high risk 22qs (…) that were ordered by GPs, and we are yet to get a true positive. We've just done a whole bunch of diagnostic procedures for so- Counselling as a protective buffer against harms 'She was so upset and confused. She somehow had missed that it was a Triple X, thought it was Turner's. She did not even know she'd opted into the sex chromosome; she did not even know there was a choice, but she was very upset 'cause she did not know what to do next and (…) none of this had been explained to her, none of it (…) I just felt like this patient has not been treated the way I think this should be handled. She was so lost.' (ID6 genetic counsellor) 'I think in early pregnancy, there's so much information. So, this will often be given in the first or second appointment where they are getting huge amount of information. So, it can be pretty overwhelming.' (ID19 GP) Clinicians' unmet education needs around NIPS 'GPs need education and clear direction and guidelines about how to just direct people about it (…) I think that all the education I've had has been about the trisomy issues, that's it, even all the latest updates, they put all the other stuff into the genetic counselling basket, which it's not. It's in our basket.' (ID18 GP) 'If it's available for lots of people to recommend and request, then there needs to be clear guidance on how they should be used and training so that people are counselling appropriately, so patients know what they are or aren't getting done.' (ID19 GP) 'The main impact that it's had on me is that I think there is no right answer about how it should be used in conjunction with the current tests that we already have.' (ID7 obstetrician) Relationship between education gaps and commercially driven NIPS 'One of the problems is that it's very much driven by industry and by the companies. So, there's been this explosion of different tests, so obviously the technology precedes the legislation, and the implementation, and the good practice around it (…) it needs to be much better education and more timely, so that it's not 6 months or a year after all these tests have come out and the GPs are scrabbling to keep up, and try [sic] and understand what they are ordering.' (ID3 clinical geneticist) 'It's kind of -seems to be pretty strongly marketed around GPs and that sort of thing, which lends itself to people doing testing possibly without good information about what test they are actually doing, which leads to all the misunderstanding and ultimately that sort of stuff can lead to patients having a bumpier ride.' (ID15 genetic counsellor) † Non-invasive prenatal screening (NIPS) is also commonly referred to as NIPT (non-invasive prenatal testing).
driven by industry rather than regulators, consistent with how NIPS has been implemented in countries with market-based healthcare systems such as the USA. 19 Our findings suggest that clinicians who see expectant parents after they have received a high chance result perceive that many expectant parents are not receiving adequate counselling from the clinician who requests the test (often a non-specialist), and may not sufficiently understand the implications of NIPS. This is supported by previous Australian research that found that a significant minority of NIPS users gave a neutral or negative response when asked about the adequacy of the pre/post-test counselling they received. 20 We have also conducted research with people who have undergone NIPS to confirm how they understand NIPS and its implications; these results will be separately reported. The majority (75%) of pregnant people in Australia receive public hospital-based maternity care 21  should be considered, including making the offer of screening for SCA and MMS from NIPS strictly opt-in (with specific counselling) and/or identifying regulatory mechanisms to ensure that current professional guideline recommendations against routinely using NIPS to detect MMS and genome-wide chromosome abnormalities 7 are followed in practice. There also needs to be more specific guidance for GPs on the use of NIPS for screening (those currently available are focused on screening for T21 24,25 ), and a national system to collect routine data on NIPS requests. This will allow ongoing evaluation of NIPS uptake, and, through data linkage to other administrative datasets, a better understanding of potential downstream consequences (both benefits and harms). Finally, should NIPS become listed on the MBS, a practice note for the item could signal best practice: that the test should be offered by practitioners with appropriate training or expertise and done in conjunction with high-quality counselling.
Strengths of this study include a rigorous analysis process, and the involvement of a multi-disciplinary team throughout the research process. Limitations include that self-report may differ from actual practice, and all participants worked in metropolitan areas with predominantly middle-to high-income populations.
Our sample did not include midwives (although our research team did) or general obstetricians, and the interview guide was developed without the input of clinicians from these two groups.
Our findings reflect the fact that our sample was predominantly made up of clinicians with a high level of knowledge about NIPS and genetic testing, who regularly care for expectant parents who have received high chance NIPS results. Their experiences, views and perceptions of psychosocial harm are likely to be distinct from clinicians without specialised knowledge of genetics whose experience with NIPS may be more centred on pre-test counselling and whose patients generally receive low chance NIPS results.
While our sample included GPs without specialist knowledge, it is unlikely that we reached thematic saturation in regard to the experiences and needs of GPs around NIPS. Further research is needed to ascertain the knowledge, views, and specific needs of clinicians without specialised genetics knowledge (including GPs, midwives and general obstetricians), if new educational resources are to be developed for them.
Our findings support the need for targeted training around NIPS, a shared decision-making approach to support expectant parents' autonomous decisions, and policy initiatives to protect against potential harms while extending potential benefits.