Introduction
As reprogenetic medicine advances, and technologies such as whole genome sequencing and Non-Invasive Prenatal Testing (NIPT) are increasingly becoming part of mainstream NHS healthcare (Genomics England
2012; Wells et al.
2014; UK National Screening Committee
2013), important social and ethical questions emerge around their usage. Chief among these concerns is the question of which conditions the technologies can justifiably be employed to prevent, particularly as the number of genetic disorders that it is possible to detect through these means is rapidly burgeoning (Plantinga et al.
2016; Leo et al.
2016).
People currently living with the genetic conditions that are potential candidates for such population screening have much to contribute to answering these questions. However, their perspectives have been vastly under-explored in the literature (Allyse et al.
2015; Barter et al.
2016; Nuffield Council on Bioethics
2017). Indeed, the literature around expanded genetic screening has instead largely focused on the intended administers and recipients of such screening; the general public (Pei-Jung et al.
2017; Plantinga et al.
2016), new/expectant parents (Norton et al.
2014; Green et al.
1993) and/or health care professionals (Watson et al.
1991). This oversight is striking given the potential for substantial impacts on people with genetic disabilities should screening be introduced. These impacts might include: changes in the public profile of the disease they live with, emotional harm associated with having a condition that wider society seeks to avoid (Boardman
2014; Barter et al.
2016) reduced public funding for biomedical research into treatments for the condition (asscciated with declining numbers of people born with the condition), as well as reductions in the availability of peer and community support (Nuffield Council on Bioethics
2017).
Where the views of adults with genetic disabilities towards screening and testing have been explored, conclusions have been somewhat contradictory, with some studies revealing reticence, ambivalence and even active hostility towards screening (e.g. Barter et al.
2016; Benjamin et al.
1993; Middleton et al.
1998; Stern et al.
2002) and others revealing far more supportive and accepting attitudes (Chen and Schiffman
2000). This strong diversity of views is perhaps unsurprising given the heterogeneity of adults with genetic disabilities, both in terms of the nature of their conditions, but also in terms of the way(s) they are experienced in everyday life. In spite of this, however, the relationship between the nature of a person’s impairment, and their attitudes towards screening has remained under-explored within the literature. Indeed, this has persisted despite the widespread acknowledgement that everyday experiences are critical to understanding the varying attitudes towards treatment and cure amongst different impairment groups (Shakespeare
2006; Bogart
2014; Bogart et al.
2012; Hahn and Belt
2004).
This study, using mixed methods research techniques, addresses this gap in the literature by exploring attitudes towards genetic screening amongst people diagnosed with a condition for which population screening could feasibly soon be offered, Spinal Muscular Atrophy (SMA).
Spinal Muscular Atrophy (SMA) and Population Level Genetic Screening
Spinal muscular atrophy is an autosomal recessive neuromuscular disorder that has been described as the most common genetic cause of infant death (Butchbatch
2016). As such, it is has been argued that it is a prime candidate for inclusion in expanded genetic screening programmes (Prior
2010). Indeed, the recent FDA approval of nusinersen as a treatment for SMA has re-kindled calls for newborn screening programmes to be introduced in the United States and beyond (Ottesen
2017).
SMA is sub-classified into four main types, based on age of onset and severity, although the boundaries of these classifications are contested (Dubowitz
1991). Type I SMA is the severest form, with onset within the first few months of life and death usually occurring before 18 months. Type II SMA (intermediate) is the most divergent form, with onset usually within the first two years of life. With improved support, lifespan for people with SMA Type II can remain near-normal, although such indviduals remain vulnerable to respiratory infections and complications throughout their lives, which may lead to premature death. Type III SMA is usually diagnosed after the age of 4 years, with the majority of those affected able to sit and stand unaided. Type IV SMA is diagnosed in adulthood, with patients developing generalised muscle weakness. In both Types III and IV there is a gradual deterioration of abilities over time, although life span is usually unaffected (Wang et al.
2007).
All four sub-types are caused by functional loss of the
Survival Motor Neuron1 (
SMN1) gene, with the clinical severity of the disease believed to be (at least in part) regulated by the copy number of a second
SMN gene,
SMN2. As current genetic analysis can detect
SMN1 deletions and assess
SMN2 copy number, they have the potential to accurately sub-type at diagnosis (as confirmed by several small pilot studies in the US). However, as there has been no longitudinal large scale trial, the sensitivity and specificity of a
SMN1 / SMN2 prognostic and diagnostic algorithm is unknown. As a result, there is a wide variety of SMA screening practices in the international arena. In the UK, although prenatal testing and cascade carrier screening are routinely offered to families with a known history of SMA, there is currently no screening programme in place for the general population (UK National Screening Committee
2013; ACOG,
2009; Cartwright
2012). In comparison, some countries have implemented compulsory pre-martial SMA carrier screening programmes (e.g. Qatar) and Israel and Australia offer screening through state-sponsored health care plans (Sukenik-Halevy et al.
2012). In the United States, SMA is currently being considered for inclusion on the state-wide newborn screening panel. It seems reasonable, therefore, to assme that popualtion screening for SMA could one day feasibly be offered within the UK, particularly in the context of emerging therapies.
This study, using mixed methods research techniques, explores attitudes towards genetic screening amongst people diagnosed with SMA. As SMA can present as an adult-onset, childhood onset or congenital impairment (with vastly contrasting levels of severity within these sub-types) a focus on SMA as a candidate genetic screening condition allows an analysis across a broad range of impairment experiences, while maintaining a meaningful comparison between them, within the broad remit of neuromuscular impairment. By so-doing, this study makes an important contribution to an understanding of the views of people with genetic disabilities in an age of expanding genomic medicine.
Methods
The data reported in this study are derived from a larger study of attitudes towards screening for SMA amongst both adults with SMA and their family members, the findings of which are reported elsewhere (Boardman et al.
2017). While family members are an important stakeholder group in debates around genetic screening in their own right, people with genetic conditions have unique experiential knowledge of the condition in question and are set to affected by population genetic screening in very specific ways (Chen and Schiffman
2000; Allyse et al.
2015). Therefore, the views of this group of adults was considered worthy of a separate and focused analysis. An exploratory sequential mixed methods research design was adopted, and the research took place in three distinct phases.
Within all phases, participants were asked about three potential screening programmes for SMA:
1.
A pre-conception screening programme (whereby members of the general population are offered screening for their SMA carrier status before conceiving a pregnancy)
2.
A prenatal screening programme (whereby pregnant women and their partners are offered screening for carriers status, and the foetus tested for SMA where indicated)
3.
A newborn screening programme (whereby parents of newborn babies are offered genetic screening to determine whether the baby has SMA).
Phase I: Qualitative Interviews
In-depth qualitative interviews were undertaken with 36 people who either had SMA or had a diagnosis of SMA in their family. These participants were recruited through advertisements placed in the publications of the main support and advocacy group for families living with SMA in the UK, ‘SMA Support UK’. Fifteen interviews were conducted with adults with SMA, and 21 with family members of people with SMA (see Table
1 for breakdown of participants). For the purposes of this study, only the 15 interviews with affected adults were included for analysis. Interviews were designed to explore experiences with SMA, views around/uses of genetic testing technologies and selective termination, as well as perceptions of the possible introduction of population screening for SMA. Participants were eligible for interview if they were aged 18 or over, English speaking and either had SMA themselves, or had at least one diagnosis of SMA in their family. Pregnant women were excluded from the study due to the increased sensitivity of the topic area.
Table 1
Qualitative interview participants
Type II SMA | 3 | 5 | 4 | 3 | 1 | 8 |
Type III SMA | 1 | 4 | 0 | 4 | 1 | 5 |
Type IV SMA | 1 | 1 | 0 | 1 | 1 | 2 |
Totals | 5 | 10 | 5 | 8 | 3 | 15 |
Data were analysed using Nvivo 10 software by an experienced researcher under the supervision of two senior academics who provided feedback on the developing coding framework. A constructivist approach to grounded theory data analysis was used (Charmaz
2008). This process was inductive, allowing the themes to emerge directly from the data, although unlike traditional grounded theory approaches, the literature was consulted during the data analysis to facilitate refinement. After the initial ‘open coding’ of the data, higher level hierarchical coding was undertaken. A process of coding, refinement of concepts (through data interpretation and consultation with the literature) and re-coding was carried out over a period of five months until ‘saturation’ had occured (i.e. no new concepts were emerging and all of the data were incorporated within the coding framework) (Glaser and Strauss
1967).
Phase II: SMA Screening Survey (UK)
Following completion of the qualitative analysis, a survey- the SMA Screening Survey (UK)- was developed in order to gauge attitudes across a wider population of families living with SMA. Details of the development of the SMA Screening Survey (UK) have been presented more fully elsewhere (Boardman et al.
2017). In brief, the overarching themes from the qualitative analysis were used to develop both the key domains, but also the individual questions within the survey. Questions that were included to capture demographic information (e.g. religious faith/ethnicity) were either directly replicated from, or appear as modified versions of, questions from the 2011 UK Census survey.
The SMA Screening Survey (UK) contained 22 items. Questions pertaining to views on screening took the form of an attitude statement derived direclty from the qualitative analysis (for example, ‘It would be a loss to society to have less people with SMA coming into the world’) presented in conjunction with a lickert scale. Cognitive interviewing was undertaken with six people with experience of SMA to explore the mental processes that participants used to answer the survey questions (Willis
2005). In addition to the cognitive interviews, the survey was independently reviewed by two expert panels (one professional, one made up of people living with SMA and their families) and the survey questions were further developed in line with their feedback.
Phase II data collection was carried out over a period of ten months, September 2014–June 2015. Two versions of the survey were made available, an online version (hosted on a secure website) and a paper version. The paper version was posted to all members of SMA Support UK (1500 households) in September 2014 and participants were encouraged to distribute it within their networks of friends and family affected by SMA. Potential participants were invited to complete the survey if they were over 18, had not taken part in a Phase I interview and either had SMA themselves, or had at least one diagnosis of SMA in the family.
Demographic variables were stratified as follows: 1) Gender (male (1) v female (0)); 2) Highest qualification (> = degree (1) v < degree (0)); 3) Religious (any) (yes (1) v no (0)); 4) Do you have children (own, fostered or adopted) (yes (1) v no (0)); 5) How do you rate your current health (good (1) v not good (0)); and 6) Are you currently trying to conceive (yes (1) v no (0)). For all questions relating to views on SMA or screening programmes, responses were stratified into two groups: 1) “Agree” which contain strongly agree and agree responses; and 2) “Other” which contained disagree, strongly disagree and neither agree nor disagree responses.
Demographic impacts on responses were assessed using logistic regression analysis; when interpreting the results positive drivers were indicated by an odds ratio > 1; negative drivers were indicated by an odds ratio < 1 (for logistic regression to cut off for significance was set at p < = 0.05). The demographic responses were stratified as binominal variables (as highlighted above) to allow analysis using binominal univariate logistic regression (SPSS software v22, IBM).
The attitudes of adults with SMA on the disease itself and the three proposed screening programmes (pre-conception genetic screening, prenatal screening and newborn screening) were compared to determine if there were any statistical differences. Responses from adults with Types II, III and IV were compared. The individual questions were assessed and then responses correlated against support for screening. For each question the number of “agree’ v “other” responses were reported and statistical differences between distribution of responses for the adults with the different subtypes were assessed using a chi-squared analysis (Graphpad Prism software, v6).
Phase III: Re-Interrogation of Qualitative Data
Key findings that emerged as significant from the quantitative analysis were explored further within the qualitative data in Phase III of the study. Returning to the qualitative data in an exploratory sequential mixed methods research design has been identified as a research technique particularly useful in drawing out the nuances, complexities and contradictions in participants’ views that would otherwise be missed by monomethod research (Plano Clark and Creswell
2008). Excerpts from the qualitative data were selected for inclusion in this paper if they particularly eloquently communicated or clarified a key finding. All of the qualitative findings reported in this paper are derived from Phase III analysis. Pseudonyms have been used throughout, and all identifying information was removed at the point of transcription in order to safeguard- as far as possible- the anonymity of participants.
Discussion
This study is the first to describe and compare the views of adults with different types of SMA (II-IV) towards three potential population level screening programmes for SMA (pre-conception, prenatal and newborn). The analysis has underscored significant differences between adults with Type II SMA and their counterparts with SMA Types III and IV, a finding which has been alluded to in other studies of adults with SMA (e.g. Kruitwagen-van Reenen et al.
2016; Jeppesen et al.
2010). Indeed, despite participants with Type II experiencing the severest clinical presentation and earliest onset of SMA represented within this sample, adults with Type II reported far more positive views of the condition than those with milder presentations. More severely affected participants were more likely to report high quality of life, to refute the claim that SMA necessarily involves suffering and were more likely than their less severely affected counterparts to see positive attributes associated with SMA, such as heightened intelligence (Von Gontard et al.
2002).
Given this relative positivity displayed by adults with Type II SMA, it is perhaps unsurprising that, as a group, they were more supportive of newborn screening (which would not alter the number of children being born with SMA) compared to prenatal screening (which would potentially increase the number of SMA-related terminations). Indeed, the qualitative and quantitative data both provided clear evidence that the lower levels of support for screening amongst people with Type II SMA stemmed from a fundamental conviction that life with SMA is of considerable value, and consequently that it would be of detriment to have less people with SMA being born. However, it is important to bear in mind when interpreting these data that in spite of this conviction, most still agreed that a pre-conception (63%) or prenatal programme (52%) should be available to the general population, even if they might not make use of genetic technologies within their own reproductive decisions.
These findings are supported by the (somewhat limited) literature exploring the views of affected families and individuals towards population-level genetic screening, where such screening is broadly supported in spite of ambivalence towards their usage (Maxwell et al.
2011; Skinner et al.
2003).
Jeppesen et al. (
2010) have argued that due to the early onset of Type II, families with affected children are able to access childhood disability services which are typically more substantial in their provision than those available for disabled adults (Campbell et al.
2016). Early supportive environments have been described as critical to the installation of an enduring positive self-image amongst children with disabilities (Hauser-Cram et al.
2001). Adults with Type II SMA who were interviewed for this study emphasised the importance of these early positive experiences. In particular, the critical role of parents in supporting the transition to adulthood was emphasised. The introduction of the Equality Act (2010) and the protection of the rights of disabled people to work, have children and otherwise participate in society has also meant that opportunities for adults with all types of SMA have improved considerably in recent years. Many such adults are now entering higher education, living independently (through the use of self-directed care plans), undertaking paid employment and becoming parents themselves (Jeppesen et al.
2010). In spite of this, however, most adults with SMA Type II (81%) reported feeling unsupported by wider society, suggesting that their positivity about SMA cannot solely be attributed to improved social and environmental arrangements. Rather, it appeared to be inextricably bound up with how they viewed and experienced their impairment and how they incorporated it within their lives.
Adults affected by the clinically milder forms of SMA (Types III and IV) were more likely to hold negative views of the condition and to more strongly support the forms of screening with the potential to reduce the number of births of SMA children (pre-conception and prenatal genetic screening). Adults with these forms of SMA typically have normal gross motor development prior to the onset of symptoms, which in some instances may not start until late middle-age. Kruitwagen-Van Reenen et al. (
2016) have argued that discrepancies between self-reported quality of life across SMA types are due primarily to the delayed onset of the condition in its milder forms. Becoming disabled in early-mid adulthood is a very different experience to being born with the condition, with the onset of the condition invariably involving a re-calibration of an individual’s hopes, dreams and expectations of their life (Locock et al.
2009).
Writing on the concept of cure, Tom Shakespeare (
2006) has argued that different ‘impairment groups’ have highly contrasting views on the possible amelioration of their condition, depending, largely, on the nature of their experience with that condition. Those people living with relatively fixed and/or congenital impairments, he argues, are typically more accepting and well-adjusted to them than people whose conditions onset later in life, or whose symptoms fluctuate and/or deteriorate (Shakespeare
2006: 106). This argument is supported by studies within the psychological and rehabilitation literature, where similar differences in attitudes have been observed between people born with their condition and those who acquire it (for example through traumatic spinal cord injury) with the latter group more fervently pursuing treatments and cure (e.g. Bogart
2014; Bogart et al.
2012; Hahn and Belt
2004).
It has been postulated that personal identification with the condition, and the adoption of a ‘disabled identity’ (Watson
2002) is critical to understanding this phenomenon. Indeed, people born with their impairments are more likely to view it as an integral aspect of their personhood (having always been there), which in turn, invariably impacts their views towards its ammeloriation, whether this be through cure or through genetic screening programmes (Kruitwagen-van Reenen et al.
2016). For those who strongly identified with their impairment, it is not difficult to see how the practice of screening could be interpreted as a negative evaluation of their own lives (Sinason
1992). As Edwards (
2004) has argued, the ‘expressivist objection’, that is, the hurt and offence that many disabled feel towards the practices of prenatal testing and selective pregnancy termination (Parens and Asch
2000), only makes sense if the disability is considered to be ‘identity constituting’ in some way. Indeed, similar objections are typically not made in relation to other areas of preventative medicine, such as childhood vaccinations (Malek
2010).
Differences in the degree of personal identification with SMA was also evident in the data relating to health. It is noteworthy that when compared to adults with Type II, participants with Types III or IV were more likely to rate their health as poor. This finding might be considered surprising given that people with Type II SMA are (clinically) more severely affected by the disease; they are more likely to suffer chest infections and respiratory complications, are more likely to need nutritional support and surgical interventions for orthopaedic complications (such as Scoliosis and join contractures) than people diagnosed with SMA Types III or IV (Wang et al.
2007). However, this finding can be explained by the observation that adults with Type II SMA separate out their (relatively static and ever-present) disability from their understanding of health and illness, a phenomenon which has been referred to as ‘response shift’ (Kruitwagen-van Reenen et al.
2016: 5). As adults with Types III and IV SMA are likely to have spent a large proportion of their lives symptom-free, the SMA-onset is more likely to be experienced as a ‘threat’ to their health and wellbeing (Shakespeare
2006: 107), rather than entirely separate from it, leading to perceptions of SMA as an ‘illness’ rather than disability, and consequently as something in need of correction or treatment (Boardman
2013).
In conclusion, this study has revealed that the type of SMA affecting a person, and consequently the nature of their experiences with that impairment, has the greatest influence on genetic screening (non)support than any other factor (age, gender, educational background, religion etc.), with support for genetic screening declining the more severely affected a person is. It has been argued that personal identification with, and acceptance of a disabled identity (which is in turn associated with an earlier onset of the condition and a relatively static disease trajectory) is key to understanding this finding. While this concept has previously been explored in relation to the notion of cure (Shakespeare
2006; Hahn and Belt
2004), this study underscores the need to transfer this analysis and understanding of the experiential realities of the lives of disabled people to the arena of selective reproduction. As genomic medicine advances, ability to detect (and consequently screen for) genetic conditions is now far outstripping ability to treat and cure them, rendering the perspectives of people set to be directly affected by this expanded screening (people with genetic disabilities themselves) ever more important in determining, and implementing, the screening agenda.
Practice Implications
This study emphasises the need of genetic counselors to be attentive to the experiential dimensions of impairment and personal identification with genetic disease in counseling contexts. Disabled people identify more or less with their impairment for a range of reasons, and age of onset/severity are key components of this relationship that may have serious implications for reproductive attitudes. People with disabilities have long been understood as having a fraught relationship with genetic medicine more broadly (e.g. Catalina within this study), with the association with eugenics posing particular challenges (Peterson
2012). Identity politics are key to understanding these tensions, and this study highlights the need for open dialogue between genetic counselors and disabled patients surrounding the nature, meaning and significance of their impairment experiences and how these relate to decisions to use, or not use, genetic technologies.
As capacity to offer screening for ever-larger numbers of rare genetic conditions expands, people living directly with these conditions have an increasing role to play in the concomitant decisions around which conditions should, and which should not, be included on such expanded screening panels. Representing the ‘best experts’ (Petersen
2006) on their own conditions, it is critical that policy makers, clinicians and scientists both value, and make use of, the experiential knowledge of disabled people to inform such decisions. Indeed, as this study highlights, reliance on clinical disease severity to determine a condition’s suitability for inclusion on genetic screening panels (e.g. Leo et al.
2016) may not adequately target those conditions which have the largest negative impact on a person’s day-to-day life.
Indeed, disabled people’s experiences are also relevant to the micro-level decisions made by members of the general population undergoing screening (Shakespeare
2005). Involving people currently living with the screened-for conditions in the education and training of the clinicians who will deliver screening (for example involving them in the production of patient literature and facilitating contact between prospective parents and those living with the screened-for conditons as appropriate) is a key means through which the experiential knowledge of disabled people and their families can be utilised and valued in screening contexts. Different online methods for conveying these insights have been developed (Ahmed et al.
2007; Telling Stories
2007), however, further research is warranted to explore the most effective and appropriate means of delivering this information in the context of high yield genetic screens.
Research Recommendations
Further research is warranted to explore how adults with conditions that have contrasting presentations to SMA (e.g. those involving chronic pain or behavioural/cognitive symptoms), those that are treatable (e.g. Haemophilia) or those which are early onset, but that which can nevertheless significantly fluctuate or deteriorate from the outset (e.g. Cystic Fibrosis, Duchenne Muscular Dystrophy) view the possibility of genetic screening. Research is also required to explore the quantity, nature and preferred mode of delivery of information about the genetic conditions that can now be screened for within the general population. While some of these concerns are currently being addressed within studies exploring consent processes for additional findings in genomic sequencing studies (e.g. Cornelis et al.
2016), whether and how information needs alter and fluctuate over the course of reproductive decision-making has yet to be thoroughly explored.
Study Limitations
Due to confidentiality and data protection issues, no identifiable data were asked of individuals who participated in the SMA Screening Survey (UK), including IP addresses (where the survey was completed online). This meant that there was no mechanism in place to prevent an individual completing multiple surveys. Moreover, there was no way of verifying that the participant fitted the inclusion criteria to participate in the survey. Participants were furthermore accessed through a national support group, personal networks and a patient registry rather than neuromuscular clinics, which may have introduced bias. Due to the very poor prognoses associated with Type I SMA, the adults with SMA who participated in this study were largely affected with clinically milder forms of SMA which invariably will have altered their percetpions of the disease. However, we feel that this sample bias does not negate the value of the perspectives of more mildly affected adults, and indeed, even in its milder forms, SMA is still a condition with significant implications for those who live with it.
A further potential source of bias within the sample relates to parental status, with a higher number of parents within the Type III and IV groups than Type II. This is in spite of the increasing number of peope affected by Type II SMA becoming parents overall (Pugh et al.
2000). Parental status might have influenced perceptions of screening amongst these more mildly affected adults as they were more likely to have previously considered the possibility of SMA in their own child