Social and cultural influences on genetic screening programme acceptability: A mixed‐methods study of the views of adults, carriers, and family members living with thalassemia in the UK

Abstract As population‐level carrier screening panels for reprogenetic information emerge globally, conditions to be included, and the timing of implementation is widely debated. Thalassemia is the only condition for which population‐based prenatal carrier screening is offered in the UK. However, little is known about the views and experiences of the UK thalassemia‐affected community toward this screening or other forms of genetic screening for thalassemia (newborn, preconception), despite the range of direct consequences of screening programmes for this group. Using a mixed‐methods integrative analysis (qualitative interviews n = 20 and quantitative survey n = 80), this study outlines the experiences and attitudes of adults with thalassemia, their family members, and screen‐identified thalassemia carriers toward preconception, prenatal, and newborn screening for thalassemia. The majority of participants described thalassemia as a burdensome condition with a range of negative impacts, which contributed to their strong support for screening in all its potential formats. However, the data also highlight the challenges of each screening mode for this group, reflected in the high level of value conflict in participants' accounts and decisions. Cultural, social, and (to a lesser extent) religious factors were found to mitigate against the advantages of early screens, particularly within faith communities. Social stigma emerged as key to this process, informing the way that thalassemia severity was not only perceived, but also experienced by affected adults, which ultimately influenced screening uptake and outcomes. These findings suggest that cultural and social sensitivity is as important as the mode of screening delivery itself, if the iatrogenic and unintended harms of screening—particularly the social/psychological burden of value conflict—are to be adequately addressed and minimized.


| INTRODUC TI ON
With the expansion and increasing integration of genomic medicine (e.g., whole-genome sequencing (WGS)) into state-funded healthcare programmes (e.g., the NHS), trials of population-wide carrier screening programmes are beginning to emerge internationally (e.g., MacKenzie's Mission, Australian Genomics Health Alliance, 2018).
These programmes are designed to identify carriers of genetic conditions either before conception or during pregnancy, on the premise that information about genetic risk will expand the limited reproductive options of carrier couples. However, given the swathe of data that WGS generates, the question of which genetic variants make suitable candidates for carrier screening is widely debated (Chokoshvili, Janssens, Vears, & Borry, 2016), as well as the best timing for delivery of this information (Delatycki et al., 2019).
Thalassemia has long been considered a prime candidate for preconception genetic screening (PCGS) (Trager-Synodinos & Harteveld, 2017) and is currently the only genetic condition for which universal prenatal genetic screening is available in the UK. Thalassemia is one of the most common inherited single-gene disorders globally, with an estimated 270 million carriers worldwide. Mediterranean countries, Southeast Asia, India, Africa, Central America, and the Middle East, have the highest prevalence, with carrier frequencies estimated to be as high as 1:7 to 1:20 (Cousens, Gaff, Metcalfe, & Delatycki, 2010). Carrier couples have a 1:4 chance of having an affected child with each pregnancy. Efforts to reduce the incidence of thalassemia have included the implementation of mandatory premarital screening programmes (e.g., Cyprus, Iran, Saudi Arabia), with varying success rates (Cousens et al., 2010). However, approximately 7,000 infants are born with the condition each year (Al Sabbah et al., 2017).

| Thalassemia aetiology
Thalassemia is an autosomal recessively inherited early-onset blood disorder that results in the under-production of hemoglobin (used by red blood cells to transport oxygen), causing severe anemia and a lifetime reliance on blood transfusions. People with thalassemia typically also require regular chelation therapy to remove excess iron from their bodies. Different types of thalassemia exist, depending on the hemoglobin gene affected (alpha/beta) and severity, with beta-thalassemia major being the most common, but also one of the most severe, types. Thalassemia is invariably fatal without treatment, but affected individuals with regular access to medicine are now described as living into their 50s, 60s, and even 70s, suggesting that thalassemia is no longer the severely life-limiting condition it once was (Vitrano et al., 2017). Moreover, the development of hematopoietic stem cell transplantation means that some people with beta-thalassemia are now being described as cured of the condition, although the widespread use of this therapy remains hampered by a lack of suitable donors and therapy complications (Srivastava & Shaji, 2017).

| Thalassemia in the UK
In England, beta-thalassemia major is thought to affect around 1,000 people, with approximately 20-25 babies born with the condition every year (Public Health England, 2018). An estimated 300,000 people are carriers of thalassemia. Thalassemia major commonly affects people of Cypriot, Indian, Pakistani, Bangladeshi, and Chinese origin; in the UK, 80% of babies born with thalassemia have parents of Indian, Pakistani, or Bangladeshi ancestry (University College London, 2019).
Population-level carrier screening for thalassemia during early pregnancy has been available in England since 2007 (Hewison et al., 2007). Pregnant women are offered a blood test as part of routine antenatal checks, and if found to be a thalassemia carrier, genetic testing of the father is offered. In instances of a carrier couple being identified, a diagnostic test of the fetus is offered. While uptake of diagnostic testing is relatively high, termination rates in England remain notably lower than for other prenatally diagnosed conditions: Only 40% of fetuses with a genotype diagnostic for beta-thalassemia were terminated between 2016 and 2017, compared to 90% of those diagnosed with Down's syndrome (PHE, 2018). Newborn screening (heel prick test) is another route for diagnosis: Although thalassemia is not formally included on the newborn blood spot screen routinely undertaken, it is occasionally discovered incidentally. Carrier couples identified through prenatal screening, but for whom prenatal diagnosis is not possible/acceptable can access newborn diagnostic testing to determine the status of their child.

| Attitudes of thalassemia-affected individuals and families to screening
Despite thalassemia being one of the most commonly screened-for conditions internationally (Cousens et al., 2010), few studies have explored the views of thalassemia-affected families and screenidentified carriers. Most studies have focused on the views of pregnant women from the general population (Tsianakas, Atkin, Calnan, Dormandy, & Marteau, 2012) and/or at-risk ethnic minority groups Atkin, Ahmed, Hewison, & Green, 2008;Darr et al., 2013;Hanprasertpong et al., 2018), to whom thalassemia screening is targetted. Nevertheless, the accounts of affected families, adults, and carriers offer important insights into the daily realities of life with thalassemia as well as the experience of receiving, and responding, to screen-positive results. As a such, the perspectives of this group are highly relevant to the wider population undergoing thalassemia screening.
No studies thus far have incorporated the views of people with thalassemia themselves into the screening literature, despite growing evidence that having a screened-for condition oneself dramatically impacts reproductive attitudes (Boardman & Hale, 2018;.
The literature is therefore lacking the perspectives of those with the most intimate and realistic knowledge of what life with thalassemia is like, insight of great relevance to the design of screening programmes, and those screened.
In response to this gap in the literature, this study examines the views of adults with thalassemia, family members of people with thalassemia, and screen-identified thalassemia carriers toward different types of screening programme for the condition.
Attitudes to screening and the social and cultural context within which these reprogenetic decisions are grounded are examined through the use of mixed data analysis (20 qualitative interviews and 80 quantitative survey responses). By outlining the social, ethical, and cultural challenges to screening implementation, this study contributes to the wider debate regarding the timing, suitability, and applications of whole-genome sequencing as a population screening tool.

| MATERIAL S AND ME THODS
An exploratory mixed-methods sequential design was adopted (Creswell & Plano Clark, 2006), using three distinct phases, as set out below. Firstly, exploratory interviews were conducted with adults with thalassemia, family members and screen-identified carriers between June and December 2017. The resulting analysis of these interviews was then used to inform the development of a national survey, which was implemented between January and June 2018. The quantitative analysis of the survey data was conducted separately at first, before being integrated with the phase one qualitative interviews, through a process of mutual illumination and direct integration techniques (Kaur, Vedel, Sherif, & Pluye, 2019). The initial qualitative themes developed in phase one were then revised to incorporate the findings of this mixedmethods analysis. Ethical approval for the study was granted by the Biomedical and Scientific Research Ethics Committee (University of Warwick) and by the NHS Ethics Committee and Health Research Authority (IRAS ID: 226508).

| Phase 1: Qualitative interviews and survey development
Qualitative interviews were initially conducted with 20 people living with thalassemia in some capacity. Participants were included if they were aged over 18, able to communicate in English and either had thalassemia themselves, had the condition in the family, or were a carrier (with or without family history). Calls for interview participants were advertised through the UK Thalassaemia Society and within a large NHS regional genetics center in England. Twenty-five people responded to these advertisements; however, two responders were excluded as they did not have experience with thalassemia, and attempts to set up interviews with a further three responders were unsuccessful despite repeated efforts.
Twenty participants took part in in-depth interviews (see Table 1 for a breakdown of characteristics). The interviews explored participants' experiences with thalassemia and their attitudes toward reprogenetic technologies that identify carriers and diagnose fetuses/ newborns with thalassemia, as well as their perceived uses of that information.
FB, an experienced qualitative researcher, was responsible for all interviews, and an interview guide was used to inform their content.
A separate interview guide was prepared for participants with thalassemia and their families (see Appendix S1), and thalassemia carriers identified through antenatal screening (see Appendix S2), due to the contrasting levels of familiarity with the condition, although both guides covered the same topic areas.
Fifteen interviews were conducted over the telephone and five face-to-face, depending on participant preference. The five face-toface interviews took place either in participants' homes (3) or in a hospital clinic (2). Interviews lasted between 40 and 70 min, were audio-recorded, and transcribed verbatim. Transcripts were returned to all 20 participants to give them the opportunity to check/ amend the resulting text, although in practice only one participant took up this offer.
The analysis was conducted using a modified grounded theory approach to ensure the resulting themes were data-driven (Glaser & Strauss, 1967). An iterative process of theme identification and refinement, with the aid of Nvivo software (v.11), was conducted until theoretical saturation had occurred.
Following completion of this analysis, the core themes were used to develop a survey instrument, the Thalassaemia Screening Survey (UK) (see Appendix S3), with verbatim text from the interviews used directly in combination with Likert scales where possible. The survey was designed to measure attitudes toward two different models of thalassemia screening currently unavailable in England: PCGS and NGS. Demographic questions were replicated (or modified) from the 2011 UK Census.

| Phase 2: National survey and quantitative analysis
Survey data collection took place between January and June 2018.
Inclusion criteria were as follows: Over 18, UK resident, diagnosed with thalassemia, known to be a thalassemia carrier, or has thalassemia within the family. No restrictions were placed on the nature of the familial relationship as experience with the condition was prioritized over biological relationship (i.e., step, adopted, and fostered family members were all eligible to be included).
A paper version of the survey was mailed to 500 households known to the UK Thalassaemia Society, and a link to an online version was distributed through their online networks, as well as the social media accounts associated with the research project. Participants were encouraged to distribute the survey to eligible family/friends.
Postal returns were processed using data scanning technology to reduce inputting error. Eighty surveys were returned and included in the quantitative analysis.
Due to the small numbers of participants in some subcategories, Fisher's exact tests were used to compare responses between groups, for example, by religious faith. Responses were stratified into agree (agree and strongly agree), disagree (disagree and strongly disagree), and other (don't know, neither agree or disagree, and prefer not to say). Missing data were excluded from analyses.

| Phase 3: Mixed-methods data analysis
In the final stage of analysis, the qualitative interview data were revisited, using direct and indirect mixed-methods integration techniques (Creswell & Plano Clark, 2006) to compare and aid understanding of the quantitative findings. This mixed analysis resulted in refined qualitative themes where the survey data provided additional or contrary insights than those generated by the qualitative analysis alone. The findings reported in this paper are derived from this final analysis and are structured around the qualitative themes that emerged from the mixed integration.
Quotations were selected that most eloquently represented the key theme being presented.

| RE SULTS
The results of this paper are organized to reflect phase three of the analysis strategy. Demographic information is firstly presented before moving on to a presentation of the final core overarching themes.

| Qualitative data
The qualitative dataset comprised of 8 people with thalassemia (3 were also family members as they had affected siblings), 11 carriers who had been identified through antenatal screening (7 of whom had affected children), and 1 family member (a parent) who had adopted a child with thalassemia (Table 1).
The affected adults, adoptive parent, and screen-identified carrier parents (n = 16) were all recruited through the UK Thalassaemia Society, while the four screen-identified carriers (without affected children) were recruited through an NHS regional genetics center. Of the 15 participants who knew the form of thalassemia in their family, 13 (90%) were affected by beta-thalassemia major and two by thalassemia intermedia. Interviewees were mostly female (70%), and all reported belonging to a religious faith. The majority (11; 55%) were Muslim, seven (35%) belonged to a denomination of Christianity (Catholic, Church of England, and Greek/Eastern Orthodox), and two (10%) were Hindu.

| Quantitative data
Of the 80 people who completed the survey, 23% were family members and 76% had thalassemia. Eighty-five per cent had experience with beta-thalassemia major, and 63% were female ( Table 2). The majority of the sample (77%) reported that they belonged to a religious faith, with a similar spread to the qualitative sample; the largest group of religious participants was Muslim (35% of the whole sample), followed by Christian (28%) and Hindu (14%). Forty-seven participants in the survey (59% of the sample) were of reproductive age (defined as 18-45 years of age), and of those of reproductive age, the majority were Muslim (24 participants; 51%). In contrast, the majority of the Christian (64%) and Hindu (55%) participants fell into the older age brackets (45-66+) (

| Prenatal screening for thalassemia: Experience and the moral minefield
While participants generally supported prenatal screening, citing 'choice' and 'information' as the key drivers for support, the interviews also highlighted the inherent difficulties involved in discovering an inherited genetic condition during pregnancy, particularly in the context of this highly religious population. Talia While Ayra was diagnosed with thalassemia shortly after birth, which Farrah described as 'devastating', Ayra initially appeared to be mildly affected and did not require blood transfusions until she was four years old. These experiences led Farrah to describe thalassemia as 'not that bad really', a factor in their decision to proceed with a second pregnancy without intervention. However, this child, Sajid, was also diagnosed with thalassemia at birth and

| Newborn screening for thalassemia: Gender and responsibility
Support for newborn screening for thalassemia was very high among survey participants (95%) ( Table 4) and slightly higher than was reported for preconception screening (92%) ( Disagree Other a 0 (0) 4 (15) 0 (0) 0 (0) 4 (5) Identifying thalassemia at birth would interfere with early bonding between parent and child Agree 1 (4) 2 (8) 0 (0) 1 (8) 4 (5 Disagree Even though parents would not know for sure how severely affected their newborn baby will be, it is still better that they know about thalassemia straight away Disagree Other 2 (7) 4 (15) 1 (9) 1 (8) 8 (10) Identifying thalassemia at birth is important as it would enable parents to make informed decisions about any future pregnancies widely acknowledged as important within the quantitative data on newborn screening (92% agreement, Table 4).
As well as a felt sense of heightened responsibility for Jawhara's diagnosis, Afra was also pivotal to the distribution of genetic risk information and policing of testing decisions within her own family: While Afra appreciated the value of early diagnosis for hers and her family's subsequent reproductive decisions, it came at a high personal cost, and one she was unable to anticipate.

Resistance and deflection
The qualitative and quantitative data together demonstrate the significance of social, religious, and cultural factors in informing attitudes toward, and experiences of, thalassemia and screening.
However, there was evidence that these influences were not rigid and could be challenged or deflected.
While Zania, for example, acknowledged the prevalence of social stigma around thalassemia, she was able to transform this into a positive assertion about her identity: …..But I also think that in some way it [thalassaemia] helped me shape my personality. Zania's perceived difference to her peers supported her emerging sense of self-determination, separating her from 'societal norms' and expectations, and enabling her to challenge and redefine restrictions.
Rather than accepting a negative label, Zania emphasized the character-building nature and unique insights gained from growing up with a condition that made her stand out from the crowd.
As well as social and cultural ideas around thalassemia, there was also evidence of participants challenging religious ideas, particularly around pregnancy termination, as Sa'id commented: I know a lot of Muslims will probably be telling you that

| D ISCUSS I ON
This study, to the best of our knowledge, is the first to examine the attitudes of people with thalassemia, their family members, and thalassemia carriers (with and without family history) toward population genetic screening in different formats. The findings help to explain the ambivalence, and sometimes contradictions, across and between qualitative and quantitative datasets, by demonstrating the complexity of views and experiences, the range of impacts associated with genetic screening, and the outward ripple effect of screening information to extended family and wider community.
The survey demonstrates that support for thalassemia screening is high in all formats among affected families and higher than reported among families living in the UK with another inherited blood disorder, hemophilia (Boardman, Hale, Gohel, & Young, 2019). Newborn screening garnered particularly high levels of support among participants (95%) (compared to 77% support for newborn screening among families affected by hemophilia), even though it would not reduce the prevalence of thalassemia, instead acting as a form of tertiary prevention (Belhoul, Abdulrahman, & Alraei, 2013). Instead, newborn screening was presented as important for the early provision of support for thalassemia-affected families, the perceived lack of which, as well as awareness of thalassemia, was strongly evident throughout the qualitative and quantitative data as well as the surrounding literature (Cousens et al., 2010;Sapountzi-Krepia et al., 2006). While survey participants did not perceive negative impacts on infant/parent bonding of an early diagnosis, a common ethical concern of newborn screens (Frankel, Pereira, & McGuire, 2016) the qualitative data suggested that there may be implications for parental guilt and blame particularly for mothers, which may have been dispersed across both partners in cases of preconception or even prenatal screening (James, Hadley, Holtzman, & Winkelstein, 2006;Hallowell et al, 2003).
Prenatal screening was broadly supported; however, for many this was considered far too late in reproductive pathways to impact pregnancy outcomes and was the screening method most highly associated with value conflict. Of the seven interviewees identified as being a carrier couple through prenatal screening, all went on to have affected children, reflecting the trend against pregnancy termination for thalassemia in the UK and beyond (PHE, 2018). This trend against pregnancy terminations among thalassemia carrier couples has often been associated with religious convictions (Al Sabbah et al., 2017;Karimi et al., 2007), although this has recently begun to be challenged (Ahmed, Atkin, Hewison, & Green, 2006;Atkin et al., 2008).
Indeed, this study confirms the suggestions of other research in this area that negative perceptions of the severity of thalassemia were far more likely to inform attitudes to pregnancy termination, over and above religious imperatives (Ahmed, Atkin, et al., 2006;Atkin et al., 2008;Shaw, 2011). Beyond this, however, this study also highlights the significant role that social and cultural stigmas play in shaping-and even creating-that disease severity. Both the qualitative and quantitative data reinforced previous findings that  (Hughes, 2004), highlighting that social factors, such as stigma, are integral to the profile of genetic conditions and consequently of direct relevance to the design of carrier screening programmes in different global contexts.
In countries such Iran and Saudi Arabia, punitive social consequences, associated with unfulfilled marriage arrangements or pregnancy terminations, have a large influence on screening uptake (Alswaidi et al., 2012;Chattopadhyay, 2006;Karimi et al., 2007)  Social, religious, and cultural influences not only create moral and personal tensions (Chattopadhyay, 2006;Boardman et al, 2017), but, importantly, also shape the lived reality of the condition itself.
Therefore, incorporating lived experience (including social/cultural influences), into the concept of disease severity (rather than biological factors alone), is an important first step toward a contextualized understanding of screening impacts.
Only when these factors are addressed may people whose lives are touched by the condition be appropriately supported in managing the value conflicts that emerge through screening with the least social and psychological burden possible.

| Study limitations
The sample size for this study was relatively small, with a poor response rate from the postal and online survey. This low response rate may be due to the high degree of stigma and secrecy surrounding a thalassemia diagnosis which was evident throughout the research process. Biases may have been introduced by the recruitment of most participants through a support organization, the UK Thalassaemia Society. An NHS regional genetics center was used to counterbalance this recruitment bias, but only procured seven interview participants.
The sample achieved also reported a high degree of religiosity with 100% of interview participants and 84% of survey participants reporting belonging to a religious faith-the majority identifying as Muslims. However, when considered alongside ethnicity data from the study (95% of survey and interview participants were of South Asian or Mediterranean backgrounds), this high degree of religiosity appears consistent with that reported among these ethnic minority groups, which are the ethnic groups most commonly affected by thalassemia.
Both within the survey and interviews, female participants were over-represented (70% of interview participants and 64% of survey participants), which may reflect a greater willingness on the part of women to discuss reproductive attitudes and behaviors than men.
Indeed, women are social positioned as more heavily implicated in reproductive outcomes than men, which may have impacted the perceived relevance of the study to women's lives.
Despite these limitations, qualitative and quantitative participants nevertheless demonstrated a wide range of experiences with thalassemia which offset somewhat the limitations of recruitment, for example, carriers with/without family history, those with/ without affected children, and participants with differing religious convictions and views on pregnancy termination. Perceptions of thalassemia may have been influenced by the fact that 80% of survey participants and 90% of interview participants had experience with beta-thalassemia major, the most severe form of thalassemia, although this is nevertheless the most common form of the condition so is unlikely to reduce the relevance of the findings.

| Practice implications
This study has a range of practice and policy implications. Firstly, genetic counselors who are likely familiar with the complexity of patient views around genetics might usefully be included in the design of population-wide carrier screening programmes. As genomic medicine expands to encompass whole population screening, the role of the genetic counselor is similarly shifting, encompassing counseling of increasing numbers of people who lack prior knowledge of the conditions screened for, and who are likely unprepared for a positive result (Ioannou, Delatycki, Massie, Hodgson, & Lewis, 2015). Genetic counselors have much expertise and insight to bring to the development of informed consent processes within such carrier screening programmes. The contribution of personal stories and insights, or providing an advisory role within the development of patient literature, for example, may enable patients to better imagine the scenarios that may arise from participation in genetic screening programmes, within the context of their own lives, before screening is undertaken.
This study also highlights the importance of sensitivity in genetic counseling practice to the way in which social stigma may play a role, not only in the way that genetic testing decisions are approached, but also its role in creating and sustaining particular lived experiences with genetic conditions. Genetic

| Research recommendations
Further research is needed to explore drivers of and resistance to social stigma and cultural and religious factors (including gender, social class, nature of the impairment, and mode of inheritance) on the lived reality of genetic disease and reproductive decision-making.

AUTH O R CO NTR I B UTI O N S
All authors contributed to the analysis, paper writing, and final approval of the manuscript. Boardman was responsible for data collection, and Boardman, Clark, Jungkurth, and Young all contributed substantially to data analysis and integration. Boardman is in agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.