Couples experiences of receiving uncertain results following prenatal microarray or exome sequencing: A mixed‐methods systematic review

Abstract Background Tests in pregnancy such as chromosomal microarray analysis and exome sequencing are increasing diagnostic yield for fetal structural anomalies, but have greater potential to result in uncertain findings. This systematic review investigated the experiences of prospective parents about receiving uncertain results from these tests. Methods A systematic search of three electronic databases was conducted. Data extraction was performed for studies that met the eligibility and quality criteria. Results were synthesised following the principles of thematic analysis. Results Fourteen studies (10 qualitative, 4 quantitative) were included. Findings were grouped into three overarching themes. Sources of uncertainty included the testing procedure, the diagnosis and prognosis, and health professionals' own uncertainty. The clinical impact of the uncertainty included parents struggling to make clinical decisions with the information available, the emotional impact included decisional‐regret, shock, worry and feeling overwhelmed. To manage the uncertainty, parents sought support from healthcare professionals, friends, family, the internet and other parents as well as remaining hopeful. Conclusions Prospective parents experience a myriad of uncertainties in the prenatal setting, which must be handled sensitively. Future research should explore optimal ways of managing uncertainty to minimise harm. Recommendations are made for discussing uncertainty during pre‐ and post‐test counseling.


| BACKGROUND
Fetal anomalies occur in 2% to 5% of pregnancies and cause around 21% of perinatal deaths. 1,2,3 Initially, prenatal testing for fetal anomalies was limited to karyotyping and targeted genetic testing. 4 Chromosomal microarray analysis (CMA), which is able to evaluate the sub-microscopic structure of chromosomes is now being offered routinely in many countries, and prenatal exome sequencing (ES), which provides resolution down to the single base-pair, is beginning to be used clinically to increase diagnostic rates. 5 There are a number of benefits in getting a result from prenatal testing. This includes the potential to provide a definitive diagnosis during pregnancy which can then inform genetic counselling, pregnancy and delivery management, and pre-and post-natal care. 6,7 Eleanor Harding and Jennifer Hammond contributed equally to this work.
Whilst genomic technologies such as CMA and ES increase the number of genetic diagnoses made in pregnancy, there remain practical and ethical challenges in interpreting results in a way that is meaningful for parents. 8 Furthermore, tests such as CMA and ES have a greater potential to result in uncertainty. [9][10][11] This is particularly challenging in the prenatal setting as many parents enter into prenatal testing hoping for and expecting reassurance and may use prognostic information to make a decision about pregnancy termination. 12 Uncertainty may arise for a number of reasons. There may be uncertainty due to a variant of uncertain significance (VUS) being identified where the relevance of that variant to the health of the baby is unknown. 10 Some conditions have variable expressivity, incomplete penetrance or fetal phenotype information may be limited meaning that even where a variant is known to be significant, it is not possible to predict the prognosis. 13,14 If no significant variant is found following an abnormal ultrasound, parents may feel they are still in a state of uncertainty around the health of the baby. 15 The last decade has seen a number of studies looking at prospective parents' experience of uncertainty in the prenatal setting. Parents frequently state that they are interested in receiving uncertain results but are surprised when they receive them, 16,17 sometimes experiencing shock, confusion and anxiety. 18,19 Here, we describe a systematic review to synthesise the literature around parents' experience of receiving uncertain results in pregnancy following CMA or ES.

| Ethical approval
Ethical approval was not required for this study.

| Design
We have undertaken a systematic review to bring a formal structure to the identification, evaluation and synthesis of research findings. As qualitative, quantitative and mixed-methods studies have been sought, an integrative approach to data synthesis has been used. 20

| Study selection
The study selection process followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines ( Figure 2). 21  What does this study add?
• Here we synthesise the current research on parents' experiences of receiving uncertain results in pregnancy including the sources of uncertainty, clinical and emotional impact of uncertainty and how uncertainty is managed.
were independently reviewed against the inclusion/exclusion criteria by two researchers. The full text of any potentially relevant studies were retrieved for further review and considered against the inclusion and exclusion criteria independently by three researchers. Any discrepancies regarding study inclusion were discussed until consensus was reached.
Studies were included if they were:

| Quality assessment
The eligible studies were critically appraised for biases using the standard quality assessment criteria developed by Kmet et al, which allows the assessment of both qualitative and quantitative research. 22 Eligible studies were appraised by two researchers (E.H. and M.H.). Checklists for qualitative (10 criterion) and quantitative (14 criterion) studies are scored as 'met' (2 points), 'partially met' (1 point), 'not met' (0 points) or 'not applicable'. The total score is converted to a percentage. We used a low cut-off point of 55%, described as liberal by Kmet et al and following the approach of other mixed methods systematic reviews. 22,23 2.6 | Data extraction and synthesis Study details, including the aim, study design, demographics and findings, were extracted into a summary table (Table 1). NVivo12 software was used to facilitate coding and analysis. 24 The quantitative and qualitative data were analysed using the principles of thematic analysis. 25,20 The results section of each of the studies which related to the experience of receiving an uncertain result was coded. For qualitative studies direct quotes from participants, themes and

Screening
Records screened by title and abstract after duplicates removed (n = 1969)

Eligibility
Full-text articles assessed for eligibility (n = 28)

Included
Total eligible studies before quality analysis (n = 14) Records excluded due to one or more criteria (n = 1941) Full-text articles excluded (n = 14): Hypothetical (5), no content specific to uncertainty (2), not CMA (3) who independently coded two randomly selected included studies.
The codes were compared and discussed until a consensus was reached. At this stage, codes that were similar were grouped into broad categories, which were then refined and grouped into overarching themes.
Once all the studies were coded, the researchers reviewed each of the codes, categories and themes and some minor changes were made (eg, splitting or combining codes, renaming themes).

| Study characteristics
Fourteen studies representing the views of 914 participants (678 women, 236 partners) were included in the review (Table 1).
Eight studies were from the USA, 6 33 and a desire for more information. 18 The reasons in one paper were not stated. 27 Three overarching themes relating to uncertainty following CMA/ES results emerged during analysis and are described below.

| Sources of uncertainty
Sources of uncertainty included women and partners' uncertainty around the testing procedure itself, uncertainty about what the results meant including the diagnosis and prognosis, uncertainty about whether online information was accurate, and healthcare professionals' (HCPs) uncertainty. More detail is provided in Table 2.

Testing procedure
Seven studies described uncertainty stemming from the testing procedure and the possible results. 6,15,18,19,26,30,31 Some women were uncertain about what a microarray test was. 15,30 In one study, a participant described not being aware of the possibility of receiving inconclusive results, 18 and there was uncertainty about how the test results would be delivered and by whom. 6,19 Results: Including the diagnosis and prognosis In two studies, 15,30 participants described not receiving enough information following their results due to the unavailability of accurate information, or HCPs limiting the amount of information they fed back due to concerns around upsetting the participants. In one study, some participants had difficulty recalling the result they were given. 18 In seven studies, participants received a VUS following CMA, which led to uncertainty. 18

Online health information
Two studies 15,27 highlighted that parents often searched for further information online, but were not clear as to whether the information they found was accurate, hence it did not resolve their uncertainties.

Healthcare professionals' own uncertainty
Three studies indicated that uncertainty for parents could arise from HCPs lack of knowledge or uncertainty around the diagnosis or condition identified. 15,27,30 Participants also described receiving conflicting information from different HCPs. 15 In some cases, participants were unable to obtain any further information about their result from HCPs, as they assumed their HCP did not know anything further. One participant expressed shock that their clinician was unable to provide any certainty about the meaning of their result. 30  Having uncertainty surrounding the prognosis for the baby, as well as the general lack of information about the future, made it difficult for participants to make decisions. For example, Bernhardt et al found that many women felt they needed more support when working out the next steps for their pregnancy. 15 In addition, participants struggled to deal with having to make decisions in such a short amount of time.

| Impact of uncertainty
Werner-Lin et al found that participants felt burdened with the pressure of managing this complex information, while dealing with their anxiety, within the limited time period. 18 Participants felt that they were unable to plan for the future with a lack of information or resources to alleviate their concerns or answer their questions. 27

Emotional impact
The emotional impact of uncertainty could create feelings of worry, affect relationships and could continue to affect parents after the child was born. Participants from six studies reported feeling shocked and worried on receiving uncertain results. 15,18,19,27,30,33 Participants across three studies 15,18,27 described wishing that they did not have the information about uncertain results, which Bernhardt et al and Rubel et al referred to as 'toxic knowledge'. 15 27 This emotional overload was often replaced with ongoing anxiety, which was reported in nine studies, 6,15,18,19,[26][27][28]33,34 along with lingering worries and uncertainties. In addition, Halliday et al found state anxiety scores to be slightly higher in women who had received an extended analysis report, which included VUS, compared to a targeted analysis, although this difference was not statistically significant. 34 Halliday et al found that decisional regret scores regarding the decision to undergo genetic testing, were higher for participants who chose to receive VUS compared to those who did not choose to receive VUS results. 34 This was also reflected by Desai et al, who reported that participants who received VUS felt less satisfied with their decision 36 months after birth, compared to those who received normal and clearly abnormal results. 29 Five studies reported participants feeling overwhelmed by the future as well as a lack of control over the uncertain situation following uncertain results. 18,27,30,32,33 Many participants questioned what would happen next as they struggled to comprehend the information and look to the future. Furthermore, Hillman et al found that participants were concerned that the issue of uncertainty could be repeated in a future pregnancy. 30 However, in one case, a couple expressed that uncertain results still provided extra information that could be beneficial for their future, stating that 'at least we know more, we are going to be prepared'. 30 One study described the impact of uncertain results on the relationship between the pregnant women and their partners. 6,15,18 There could be conflicting opinions between partners, with one wanting to discuss the pregnancy with friends and family, and the other preferring to keep the pregnancy private. However, couples also found that these difficult experiences could strengthen their relationship as the long, emotional conversations resulted in an 'aligning of their priorities'. 18 The emotional impact of uncertainty could continue after the child was born. One study described how mothers would be in a state of 'watchful waiting' as they would monitor the health and development of their child, scrutinise their child's appearance, and make comparisons against their unaffected children. 28 One participant commented 'when things weren't as advanced as my first daughter, we would question, "do you think it's that?"'. 28

| Managing uncertainty
Parents had differing levels of tolerance when it came to receiving uncertain information. Some parents reported wanting to know as much information as possible despite the potential for receiving uncertain results, whilst other parents did not want to receive such information. [31][32][33][34] Three studies reported the experience of participants who were not additionally concerned by receiving an uncertain result. 17,26,28 For example, one participant, described by Werner-Lin et al, explained how she did not give the VUS a second thought after birth. 28 The majority of participants, however, reported not receiving as much information as they wanted, which they felt prevented them from gaining definitive answers and 'grasping the significance' of their results. 26 In dealing with uncertainty, parents were found to seek support and further information, whether this was through speaking with a HCP such as a genetic counsellor, their friends and family, other parents or searching for information online. 6,15,[17][18][19][26][27][28]30 The majority of participants appreciated support from their HCP, including a referral to a genetic counsellor for emotional support, particularly as uncertain results could need longer, more specialised or more frequent counselling. 6,15 Participants also relied on friends and family, especially during the period of time straight after receiving results when they were most distressed and scared. However, Werner-Lin et al reported that participants sometimes did not want to share the uncertain information with family members, for fear of stigma towards their child after birth and a lack of understanding from others with one patient commenting 'My dad would treat [child] differently even though the results don't say anything definitive'. 28 Participants also reported the utility of speaking to other parents in similar situations and this was mainly achieved on the internet, through online communities and advocacy groups. Furthermore, Wou et al found that many participants would have liked to be connected with another family with a similar experience, for mutual support and understanding. 6 One study illustrated how uncertainty could also be managed as a couple, with both the pregnant woman and her partner playing an important role in the process. For example, Werner-Lin et al found that, within a couple, the pregnant woman often acted more as a seeker of information, while her partner provided emotional support, to help with decision making. 18 Three studies reported how participants' spoke of remaining hopeful. 15,28,30 They remained hopeful that they would eventually find enough information to make informed decisions and hoped that their test results could be used by researchers to provide answers for women in the future. 15

Pre-test
• Be aware that there are multiple sources of uncertainty linked to prenatal testing and parents differ in their willingness to be told uncertain results.
• Parents should be told of the potential for uncertain outcomes. This includes: o Uncertainties related to the results e.g. VUS, no genetic result, uncertain prognosis following clear genetic diagnosis; testing process e.g. diagnostic yield, o Uncertainties related to the limitations of potential for false negatives; o Uncertainties related to our current knowledge and understanding e.g.
• To manage the uncertainties related to test procedure, parents should be told all steps of the testing process, including how long they will have to wait for results and how they will be contacted.
• Parents will vary in their tolerance for uncertain information and this may change over the course of the pregnancy. Ideally, healthcare professionals should discuss patients' preferences for being told uncertain information prior to testing.

Post-test following an uncertain result
• Uncertain findings can be challenging to explain and as such flexibility may be needed for how long the post-test counselling session takes to allow time for explanations, discussing implications and addressing questions from parents.
• Some non-genetics clinicians may need input from genetics specialists to explain uncertain findings and their implications.
• If VUS are reported, it is important to discuss the possibility that the classification of the variant may change as new knowledge is gained. 39,49 In addition, policies for reanalysis when there are VUS or no significant findings should be made clear to parents.
• Parents may value being told what is known and certain as a way of relieving anxiety and establishing normality in pregnancy. This could include explaining that many syndromes and severe disorders have been ruled out through testing, and/or through taking extra care to point out what is structurally normal at any follow-up ultrasounds 50 .
• Parents who have been told uncertain findings may value psychological support e.g. through a genetic counsellor or other psychological support service.
• Understand that parents may want to ask further questions after the initial posttest counselling session as they digest the information. Ensure that they have appropriate contact details to access follow-up care.
• Where appropriate, signpost parents to support groups including those for parents without a diagnosis.  42 The lack of educational resources to support patients is also an issue. 43 Uncertain findings can have a negative impact on the doctor-patient relationship, as parents sometimes react angrily when they are struggling to make decisions about the pregnancy. 35 Whilst patients might be informed during pre-test counselling of the possibility of receiving uncertain findings, it may be that the reality of such findings is not being properly considered prior to testing. 15 It has been suggested that clinicians could perhaps discuss with parents their tolerance for ambiguity as part of pre-test genetic counselling, to ascertain whether information that is uncertain will be useful or problematic for them personally. 15,44 Biesecker et al suggests that examining patients' tolerance of uncertainty, resilience and optimism alongside their expectations about genomic testing, may help to identify those more likely to appraise uncertainty as a threat, and to alleviate negative responses. 45 In some settings, parents have been offered the choice between 'targeted' and 'extended testing' whereby CNVs with incomplete penetrance and VUS are reported. 32,34 Furthermore, research in this area would be valuable.
Finally, another area for future research relating to uncertainty is parent experiences and views of the reanalysis when there is a VUS or no findings. Previous research with clinicians, scientists, genetic counsellors and patient groups/charities, has found that patient representatives supported reinterpretation of results over time, more so than other participant groups. 35

| Strengths and limitations
Strengths include the systematic and rigorous approach taken to identify and appraise the studies, that all were high quality studies, and the inclusion and integration of results from qualitative and quantitative research which provides rich data on parents' experiences.
Limitations include that the sample is predominantly comprised of white, educated participants. Therefore, the findings lack the perspectives of minority ethnic groups and those from lower educational backgrounds, both who experience inequity in access to healthcaresystems and disparities in understanding prenatal testing options. 46,47 The experiences of partners are underrepresented in this study, accounting for only 26% of the total sample. Finally, only two studies investigated the experiences of those undergoing ES, which makes it difficult to make comparisons between the experiences of parents' undergoing ES and CMA.

| CONCLUSION
The findings of this review highlight the different types of uncertainties that prospective parents experience in the prenatal testing setting, and the implications of these uncertainties. Whilst many of the uncertainties relate to our current knowledge and understanding of genotype-phenotype correlation, there are some uncertainties that can be managed during pre or post-test counselling for example, parents not aware that the test could reveal uncertain results. Moreover, we identified evidence of good-practice when managing uncertain results for example, additional support. In light of these findings, we have developed a set of recommendations for HCPs as a guide for best practice when offering prenatal testing ( Figure 3). 39,48,49 Whilst there are some guidelines on mitigating for these issues, further research should look to explore optimal ways of managing uncertainty in the prenatal setting to minimise the potential for patient harm. 50