Autistic perspectives on the moral and ethical considerations of genetic testing for autism

Abstract

This study aimed to characterize the nuanced perspectives expressed by Autistic adults about genetic testing for autism. In this secondary analysis of an existing dataset, we used inductive content analysis of free-text responses that Autistic adults wrote in response to questions asking about perceptions of genetic testing for autism that had been included in a previous survey. Participants who completed the survey had been recruited online. Three-hundred eighty-nine participants wrote a total of 28,774 words in response to the questions of interest. While some respondents wrote about potential benefits of genetic testing (79/389), more (333/389) wrote about potential negative outcomes of genetic testing. Participants wrote about concerns of eugenics (281/389), described a lack of trust in how genetics information would be used (125/389), and raised concerns that genetic testing could increase discrimination (66/389). This is the largest study to date to conduct an in-depth analysis of Autistic adults’ opinions on genetic testing for autism. Most respondents raised serious concerns. Additional work is needed to collect more Autistic opinions about genetic testing for autism.

Introduction

Autism is a neurodevelopmental condition with a broad range of phenotypes. It can be characterized by challenges with social interaction and communication skills, as well as distinctive, repetitive behaviors and interests. Autism is complex, with interactions between genetics and environment contributing to the condition [1]. The heritability of autism is estimated to be in the range of 50–90% [2]. In recent decades, a growing body of evidence has elucidated some of the genetic underpinnings of autism.

Currently, chromosomal microarray testing and Fragile X testing are recommended by both the Canadian and American Colleges of Medical Genetics [3, 4] as first-tier tests for identifying genetic syndromes that can underlie a presentation of autism. It is currently standard-of-care in many countries to perform genetic testing to identify the cause of autism, and recent studies have reported a diagnostic yield of about 25% [2, 5].

Studies aimed at understanding perspectives of Autistic adults on genetic testing for autism are now starting to emerge [6–8]. In previous work, we and others have found that Autistic adults have mixed views about genetic testing – with many expressing concerns, especially around eugenics [6, 8, 9]. Other work has shown that Autistic adults who had already undergone genetic testing had significantly more positive beliefs about genetic testing for autism, while beliefs about testing for other conditions were not different between groups [7].

In our previously published work, we reported quantitative survey data from 461 English-speaking Autistic adults about their opinions on their diagnosis and genetic testing for autism. In the same study, we provided participants with freeform text boxes in which to provide any additional thoughts on these topics. These text boxes allowed participants to express a broader range of opinions than can be captured in a survey format. In this paper, we seek to provide richer insight into the perspectives of Autistic adults on genetic testing for autism by analyzing these text responses.

Methods

Participants, recruitment, and approach

This study was performed using data collected from the participants who were recruited for the previous study, and was conducted by a team of genetic counselors, led by an Autistic adult [8]. In brief, English-speaking individuals who had been clinically or self-diagnosed with autism/as Autistic were recruited through social media posts on Facebook and Twitter from November 2021 to March 2022. Posts included a brief description of the study and a link to the survey. The survey was designed based on previous studies of parents of Autistic individuals. This study was approved by The University of British Columbia research ethics board (H21-01937).

Free text responses to questions about experiences with and perceptions of genetic testing for autism that were collected as part of the survey [8] were analyzed using inductive content analysis. Specifically, in this study we analyzed participants’ responses to two questions: “If you have any religious, ethical, or moral concerns with respect to genetic testing for ASD, please elaborate here” and “Do you have any more comments you would like to share regarding your opinions for genetic testing for ASD?” An inductive approach to content analysis is best suited to situations where little is known about the topic of study as is the case here. Given that this was the aim of our study, this methodological approach was a good fit.

Coding

To begin, two authors (MH and LB) performed independent readings of the dataset to identify an initial set of codes. Once the two coders agreed upon the initial codes, these codes were applied to the dataset. During the application of codes, both authors identified new codes which were brought to discussion. Any newly agreed-upon codes were then applied to the dataset, and again new codes were identified. This process was repeated iteratively until the authors agreed that all recurring ideas in the responses had been captured. Codes were developed to capture both manifest content (i.e. what did the respondent say?) and latent content (i.e. what did the respondent mean?).

As we worked through the data and our understanding of the content developed, some codes were combined while others were split apart. For example, an overarching code was created to combine several codes describing specific negative outcomes related to genetic testing, and a general code which captured a range of potential benefits was split apart to describe the specific benefits that were raised. Sub-codes were used to capture ideas that were only raised in the context of different, larger ideas.

Given the open-ended nature of the questions and the sample size, not every concept could be given a code. Concepts that arose only once in the dataset were not given a code and were instead labeled as “other.”

Achieving consensus

If consensus could not be reached through discussion between MH and LB, JA was included to resolve disputes. Ultimately, consensus was achieved for all codes.

Statistical analyses

We described the frequency of each code, and a correlation table was made using Pearson Correlation to evaluate the strength of correlation between individual codes. A logistic binomial regression was used to evaluate if demographic factors were predictive of the most commonly occurring codes.

Results

Demographics

A total of 641 participants completed the initial survey. Incomplete (n = 134), duplicate, and spam responses (n = 53) were removed. Participants who did not provide text responses to at least one of the two questions that constituted the focus of this analysis were also removed (n = 72). This left 389 individuals who wrote meaningful responses to these questions. Of these, most had not had genetic testing (369/389), while a small number had been tested (20/389). Just over two-thirds were professionally diagnosed with autism (281/389), while nearly one third were self-diagnosed (108/389). Almost all preferred to use identity-first language (i.e. “I am Autistic”) rather than person-first language (i.e. “I am a person with autism”) (367/389). Other demographic data are provided in Table 1.

Table 1.

Participants’ demographic characteristics.

Demographic characteristic	N = 389 [1]
Gender
Woman	212 (55%)
Different identity	101 (26%)
Man	67 (17%)
Prefer not to answer	9 (2.3%)
Sex
Female	302 (78%)
Male	62 (16%)
Prefer not to answer	25 (6.4%)
Level of education
Bachelor’s degree (ex. B.A., B.Sc., LL.B, B.Ed.)	118 (30%)
Master’s degree	85 (22%)
Some post-secondary education	50 (13%)
Diploma or certificate from a community college	30 (7.7%)
Diploma or certificate from trade, technical, or vocational school	27 (6.9%)
Doctorate	26 (6.7%)
High school diploma	23 (5.9%)
Other	15 (3.9%)
Less than high school	10 (2.6%)
Prefer not to answer	3 (0.8%)
Degree in Medicine, Dentistry, Veterinary Medicine or Optometry	2 (0.5%)
Relationship status
Married or common law	155 (40%)
Single	123 (32%)
In a relationship	79 (20%)
Other	17 (4.4%)
Prefer not to answer	15 (3.9%)
Living situation
I live with a partner/spouse/significant other	195 (50%)
I live alone	70 (18%)
I live with family members (Parents, siblings, grandparents, etc….)	66 (17%)
Other	30 (7.7%)
I live with roommates	23 (5.9%)
Prefer not to answer	5 (1.3%)

Participants wrote a total of 28,774 words in the free text boxes that were analyzed for this study.

Responses

A total of 18 recurring ideas were identified, plus an “other” category (which comprised ideas that we only noted once each). This led to 19 codes being applied to the dataset. Consensus on codes applied to individual responses was achieved between MH and LB for all but 36/389 (9%) responses. Consensus for these 36 was achieved after consultation with JA.

Topics discussed by survey respondents as well as the number of respondents who discussed each topic are presented in Table 2. Codes that are indented under a previous code had a complete overlap with that previous code and were considered sub-codes. A single response could have as many or as few codes as were necessary to capture the content. The median number of codes applied a response was 3, and the number of codes assigned to a response ranged from 0–10. Some respondents left simple answers, needing only one code. For example, Participant 261 (a 30-year-old woman) wrote: “My biggest concern is the risk of eugenics” and was coded with 2 codes: having concerns about eugenics and describing broad negatives regarding the potential harms of genetic testing. Other participants wrote long-form responses which required many codes, like Participant 211 (a 27-year-old non-binary person) who wrote 552 words, to which we attached 10 codes.

Table 2.

Code definitions, counts, and examples.

Highest category level	Sub-category	Count (N = 389)	Quote(s)
Broad Negatives: Described potential harms of genetic testing		333	“There are many autistic families that are against genetic testing. It will cause harm to families who may not yet understand that spectrum people tend to procreate with spectrum people often whether they realize it or not and that there are too many unsuspecting people who are actually autistic and don’t know because of how well they have been able to fit into society.” – Participant 396, a 41-year-old woman
	Eugenics: Concerned that genetic testing could lead to eugenics	281	“It will lead to pre-natal screening and eugenics. Autistic people deserve to live.” – Participant 47, a 28-year-old woman
	Down Syndrome: Compared genetic testing for autism to genetic screening for Down Syndrome	49	“Society’s ableism is responsible for the genocide being committed against people with Down Syndrome. I never want genetic testing to exist for genetic ways of being or neurotypes that are not fatal.” – Participant 270, a 42-year-old non-binary person
	Pro-Choice: Stated they are pro-choice or supportive of abortion	15	“Parents will be pressured to abort their autistic children simply because they are autistic- I support abortion wholeheartedly, but there is a line between ‘I don’t want a child’ and ‘I want a child, just… not this one.’ It will decrease the autistic population and increase the frequency of abusive ‘autism moms’ who believe they should be given a trophy for raising their autistic child” – Participant 166, a 19-year-old trans man
	Trust: Expressed a lack of trust in the people doing genetic testing, or in what would be done with the information	125	“I don’t think genetic testing in itself is bad, it’s all in how that research is used and I have major concerns about the goals of most autism research at this point in time (the cure narrative, for example).” – Participant 264, a 20-year-old woman
	Discrimination: Concerned that genetic testing could increase discrimination against Autistic people	66	“My concerns are what society will do with the information. What is intended to improve and equalize usually increases harm and stigma.” – Participant 21, a 51-year-old woman
	Cure: Worried that genetic testing might be used to cure autism	65	“I have an ethical objection. I feel a search for a genetic cause predicates finding a ‘cure’ for a natural variation, which would fuel desires for eradicating autism.” – Participant 37, a 49-year-old man
	Broad Division: Thought genetic testing may increase societal division	35
	Diagnosis Legitimacy: Division about what constitutes a valid diagnosis	32	“All this would do is lead to a “cure”, and have people aborting their children that may have autism. It will also screw over people who are diagnosed with autism if the testing doesn’t support the diagnosis.” – Participant 217, 29-year-old man
	Disagree with Community Norms: Division within the Autistic community	3	“I don’t think it [genetic testing] is beneficial to the Autism Community. But I think that is a good thing. I hate that community (At least the loud vocal one that usually doesn’t even have a diagnosis) I only call myself autistic instead of person with autism what I actually prefer because I have been “educated” (Means being talked down by) people that are not even diagnosed about how I should be as an actually diagnosed person that can’t live independently. But these wanna be special people that “speak for ALL autistic people” think they know better… I hate that community and it can burn for all I care.” – Participant 435, 25-year-old woman
Potential Benefits: Identified potential benefits of genetic testing		76	“Genetic testing is as harmful or as beneficial as the research question itself [if it is used to] strengthen and grow humanity’s knowledge of the incredible diversity encompassed by human neurodivergence and to better design education or interventions or supports so that they meet the needs of the whole spectrum of child-learners? If so, yay […]” – Participant 120, a 46-year-old woman
Consent: Talked about the consenting process for genetic testing		35	“Genetic testing should only ever be done on consenting individuals who understand the possible risks involved” – Participant 448, a 44-year-old man
Anger: Expressed anger about genetic testing		30	“SO MANY autistic adults are against genetic testing for autism. How on earth is it ethical to steamroll over us?” – Participant 30, a 44-year-old non-binary person
Oppose Medicalization: Opposition to the medicalization of autism		19	“…I worry that people would use the information to try and devise gene therapies with the goal of curing autistic people, when we’re not broken and we’re fine the way we are. I suffer from ableism in society, not because of my autism!” – Participant 403, a 38-year-old woman
Suggestions: Offered concrete suggestions about how resources should be used rather than genetic testing		18	“More research dollars need to be spent on Autistic life-affirming, social, economic, and mental health needs. More researchers need to be Autistic” – Participant 74, a 49-year-old woman
Inconclusiveness: Discussed the inconclusiveness of genetic testing for autism		18	“[…] Another point is that is a ‘genetic’ explanation the only explanation? Consider my mother was [taking] antipsychotic medication during pregnancy until a doctor informed her about the risks medication may have, could it be that the medication may have impacted my brain in some way?” – Participant 204, a 38-year-old woman
Diagnostic Criteria: Critiqued current diagnostic criteria		11	“[…] the autistic community questions the validity of the current diagnostic criteria on multiple grounds, notably: (1) steep selection bias in autism research, namely the overwhelming majority of subjects being white school-aged boys; and (2) over-representation of trauma symptoms and behaviors in the diagnostic criteria, in light of the excess traumas that autistic people face compared to the general population. Genetic research, especially towards genetic testing, is critically premature while these issues remain unaddressed.” – Participant 337, a 31-year-old woman
Other: Expressed unique ideas that were not shared with any other participants		14	“There is not [an] established scientific basis for genetic determinism in the Baron-Cohen model; it is not consistent with what is now known from recent molecular biology” – Participant 331, a 65-year-old man

All of the 15 respondents whose contributions received pro-choice codes were also coded for concerns of eugenics. Participant 166 (a 19-year-old trans man) wrote:

“I support abortion wholeheartedly, but there is a line between ‘I don’t want a child’ and ‘I want a child, just… not this one’ […] A genetic test for autism actively promotes and contributes to eugenics and the idea that all disabled people live bleak, joyless lives and that killing them is a favor.”

Of the 65 respondents whose contributions received codes about a cure, 57 were also coded for concerns of eugenics. Participant 6 wrote:

“I believe that many people desire to cure autism. The community as a whole is scared that all genetic research could lead to eugenics and cure narratives instead of helping Autistic people.”

All 49 people who compared genetic testing for autism to screening for Down Syndrome expressed a concern that genetic testing would lead to eugenics.

Although we did not ask specifically about consent, 35 respondents discussed the consenting process, highlighting who they think should and should not be able to provide consent to a genetic test. Of these, the most common sentiment discussed was that the decision to pursue testing should not be made by a parent (11/35). However, some participants did write about times when a parent or caregiver could consent for an Autistic individual (7/35). Participants also emphasized the need for consent to be elective (6/35), non-coercive (5/35), and informed—more than “[…] handing a patient a Lifelabs slip casually” (4/35) (Participant 241, a 43-year-old woman).

Benefits and suggestions

Seventy-nine respondents identified potential benefits of genetic testing. The most common benefits cited were aiding in the understanding of comorbid conditions (30/79), normalizing autism and showing it is a normal part of human variation (30/79), community-building around specific genetic diagnoses (25/79), and aiding in the speed or validity of a diagnosis (24/79). Twenty-four of the 79 named no specific benefit, saying only that they could imagine potential benefits. Twenty of the 79 expressed that they felt more knowledge was an inherently good thing.

“If we can create a genetic map of autism spectrum, I think this would be very beneficial to future generations. I am me and cannot change that. However I understand that my genetics help make me who I am and think that the more we understand and know about ourselves would be more beneficial to society as a whole. I think this type of testing could also help identify comorbidities that people on the spectrum could suffer from. Such as depression.” (Participant 143, a 38-year-old man)

Relatedly, 18 participants gave specific suggestions about ways that funds and resources could be better spent on other things than genetic testing and research. Common suggestions were to re-focus efforts away from cause-related research to emphasize supports and resources, to focus research on comorbidities rather than autism itself, to develop laws to protect Autistic individuals from discrimination, and to center Autistic researchers in the space.

Correlations and relationships between variables

Among participants who expressed a lack of trust in the institutions performing testing, there was a significant correlation with identifying potential benefits of testing itself (cor = 0.439; p < 2.2e–16) (Fig. 1). This was well-articulated by Participant 24 (a 22-year-old woman):

“Mainly moral and ethical concerns. I am interested in genetics, so of course I think learning and researching the genetics of autism would be awesome. But my concerns lay with the potential for discrimination. If genetic testing just leads to helping people be diagnosed earlier via genetic testing, then good idea. But if someone has a variant not yet known/researched, it would be discriminatory to prevent their diagnosis on the basis of no genetic evidence. There is also the potential for researchers or doctors with eugenic beliefs to abuse the knowledge of genetic variants. The good and bad potentials exist simultaneously. I really like the idea of it, but I also recognize the capacity for harm.”

Fig. 1. Correlations between categories.

The strength of correlation between categories is indicated by intensity of shading. Positive correlations are in red, and negative in purple.

The other significant correlations are artefacts of the way that certain codes were constructed—for example, the strong correlation between diagnosis legitimacy and broad division arose because all participants who were coded as having concerns about diagnosis legitimacy were also coded as having concerns about division broadly.

Finally, given the previous work demonstrating differences between Autistic adults who had had genetic testing, and those who had not (in terms of attitudes towards genetic testing for autism) [7], we further explored our data by examining the frequency of codes among those who had, and had not received genetic testing (Table 3). In this content analysis, there were no significant differences observed between participants who had received genetic testing and those who had not.

Table 3.

Code counts, broken down by whether or not the individual had received genetic testing for autism.

Code	Total count across all participants who answered Count (N = 389)	Individuals who did not have genetic testing for autism Count (N = 369)	Individuals who had genetic testing for autism Count (N = 20)
Broad Negatives	333 (85.6%)	317 (85.9%)	16 (80%)
Eugenics	281 (72.2%)	269 (72.9%)	12 (60%)
Down Syndrome	49 (12.6%)	49 (13.3%)	0 (0%)
Pro-Choice	15 (3.9%)	14 (3.8%)	1 (5%)
Trust	125 (32.1%)	118 (32%)	7 (35%)
Discrimination	66 (17%)	66 (17.9%)	0 (0%)
Cure	65 (16.7%)	64 (17.3%)	1 (5%)
Broad Division	35 (9%)	34 (9.2%)	1 (5%)
Diagnosis Legitimacy	32 (8.2%)	31 (8.4%)	1 (5%)
Disagree with Community Norms	3 (0.8%)	3 (0.8%)	0 (0%)
Potential Benefits	76 (19.5%)	70 (19.0%)	6 (30%)
Consent	35 (9%)	33 (8.9%)	2 (10%)
Anger	30 (7.7%)	29 (7.9%)	1 (5%)
Oppose Medicalization	19 (4.9%)	19 (5.1%)	0 (0%)
Suggestions	18 (4.6%)	17 (4.6%)	1 (5%)
Inconclusiveness	18 (4.6%)	16 (4.3%)	2 (10%)
Diagnostic Criteria	11 (2.8%)	10 (2.7%)	1 (5%)
Other	14 (3.5)	13 (3.5%)	1 (5%)

Discussion

This study adds to a growing body of literature detailing Autistic adults’ opinions on genetic testing for autism. The majority of respondents expressed concerns about the possibility of genetic testing being used to facilitate eugenics.

Our findings are supported by, and in turn reinforce those of the study of Klitzman et al. [6], where participants reported both potential benefits of genetic testing including societal acceptance, earlier diagnosis, and increased availability of services, and potential harms, notably the potential for eugenics, with 6 of 11 interviewees bringing up the subject unprompted [6]. Klitzman was able to ascertain that some of these concerns arose from misunderstandings held by participants about the likelihood of finding monogenic causes for autism in the absence of developmental delay and intellectual disability.

Gallion et al. also examined the attitudes of Autistic adults towards genetic testing for autism. In that study, 19 of 163 participants included in the analysis had received genetic testing, and the authors found that those who had already had genetic testing for autism were more likely to have positive beliefs about the practice than those who had not [7]. In our study it appears that there may be similar differences in perceptions between those who had received genetic testing previously and those who had not. However, as only 20 out of 389 participants fell into this category, the statistical analyses and conclusions that can be drawn are limited. Additional work delving into the specific beliefs and experiences of those who have had genetic testing for autism could shed further light on this topic.

The results of this study are also concordant with findings in other communities where those with a genetic “disability” have embraced it as a cultural or social identity. The Deaf community has been relatively well-studied, and one of the first studies in that population found that many in the community opposed or had concerns about genetic testing [10]. There are commonalities in the perspectives expressed by members of the Deaf community and our respondents: that genetic testing may do more harm than good, that genetic testing would devalue people with the trait, and that genetic testing could be used to “cure” or eliminate the trait [10, 11]. Subsequent publications have found a shift towards a more positive view of genetic testing since this initial study, though participants continue to express mixed opinions about genetic testing [12–14]. It seems likely that a person’s extent of affiliation or cultural identity as a Deaf person impacts both their views on and reasons to pursue genetic testing for deafness [11]. In our population, most respondents reported a preference for identity-first language, suggesting a strong affiliation with autism as identity. However, we did not find this or any other demographic factor to reliably predict responses.

Another similarity that can be drawn between our population and other, better-studied populations is a lack of trust in the medical system [15, 16]. In our population we found a significant correlation between expressing mistrust and identifying potential benefits of genetic testing, which is similar to what has been found in other communities. Many of our participants articulated that they feel society is “not yet” ready for genetic testing for autism due to the potential harms it could bring about. Most of these harms would ostensibly be the result of deep-seated ableism. These worries are not without foundation: understanding of the genetics of deafness has led to controversial laws in the UK disallowing the selection of an embryo with a “serious disability,” which deafness was considered to be [17, 18]. Hereditary deafness can often be attributed to a single-gene cause with a clear inheritance pattern; in contrast, while there are single gene causes of genetic syndromes which can include autism as a feature, there are no single gene causes of isolated autism. Our results echo those of Ellis and Asbury, who note that work is needed to “reconnect” the Autistic and research communities since autism research is particularly fraught with historical examples of abuse [9]. Anti-vaccination movements have often cited an infamous, discredited, and long-withdrawn paper linking vaccines to autism. This speaks to an intense social fear of autism among parents. While the effects of medical ableism on perceptions of Autistic individuals have not been well-studied, ableism is, in many ways, integrated into the foundations of biomedical research.

Some participants wrestled with personal conflicts in their beliefs about genetic testing and their feelings about termination of pregnancy. Fifteen respondents mentioned that they are generally “pro-choice” or support the right to choose to terminate a pregnancy. However, all 15 of these people also expressed concerns about selective termination of Autistic fetuses, and their responses grappled with the nuances of individual choices versus societal patterns.

Challenges and limitations

Analyzing free-text responses presents unique challenges. Since responses are entirely volitional, response length and factual accuracy is not consistent between respondents. As such, responses could be as short as one word or as long as several paragraphs. We dealt with this challenge by using a positivist, counting-based methodology. Using positivist content analysis meant we were analyzing responses for the written content itself—not the underlying themes or meaning of the content. This meant response length and factual accuracy were not factors for how a response was coded.

Text responses are open to interpretation. Notably, respondents may believe incorrect information or use the wrong terms for technical concepts (such as mistaking a microarray for whole genome sequencing). This issue was addressed by the coders achieving consensus on every code rather than achieving an inter-coder reliability early in the process and trusting this reliability throughout. By achieving consensus, both coders had to agree on the content of the text presented.

Participants in this study were recruited online, using an invite link to participate in a survey shared on social media platforms. This means we were only able to reach individuals who can use the internet independently, and who have access to the content of these groups. As was noted in our previous publication, the use of social media groups as platforms for study recruitment may have self-selected for people with stronger opinions on genetic testing. Analyzing the written content of responses has provided insight on the biases that might exist in these groups. Participant 529 (a 34-year-old woman) remarked that “Being vehemently anti genetic testing is a strongly enforced social rule in autism groups”.

A second limitation is that our sample is not generalizable given that it is not representative of the general Autistic public. Our ratio of people who were assigned male at birth to those assigned female at birth is about 1:5. A study in 2022 on the global prevalence of autism found the male to female ratio to be closer to 4:1 [19]. However, our study population does have an increased proportion of gender-diverse individuals, which is an established finding in the Autistic Community [20]. Previous work has primarily studied Autistic children and has rarely collected information on gender identity, instead focusing on (presumptive) sex assigned at birth [19, 21]. Making this distinction is an important step towards engaging and understanding the broader Autistic Community outside of the stereotypical clinical presentation. More voices are needed to characterize the full spectrum of beliefs and opinions of Autistic adults on the issue of genetic testing.

Finally, this analysis was performed on the responses to two questions from our previous survey. It is important to note that one of those questions was phrased in a way that invited responses to focus on negative aspects of testing: “If you have any religious, ethical, or moral concerns with respect to genetic testing for ASD, please elaborate here.” This free text box was provided as a follow up to a Likert scale question where participants were asked to indicate if they strongly disagreed, disagreed, were neutral, agreed, or strongly agreed with the statement “Genetic testing for autism makes me uncomfortable for religious, ethical, or moral reasons”. The survey was designed on the framework of similar surveys presented to parents of Autistic children as described in the initial paper 8. Specifically, the question on moral, ethical, or religious concerns was designed based on a similar question from Cuccaro et al. [22], where they asked parents if “Genetic testing makes me uncomfortable from a moral standpoint”. Notably, in this survey of 408 parents of Autistic children, 88% strongly or somewhat disagreed with that statement, indicating that they “did not find genetic testing morally objectionable”.

In our data set, 60% of Autistic adults surveyed agreed or strongly agreed to the statement that “Genetic testing for autism makes me uncomfortable for religious, ethical, or moral reasons”, 10% remained neutral, and 30% disagreed with the statement. Though the root of the question is the same in the original study (in parents) and in ours (in Autistic adults), the difference between the frequencies of the Likert responses suggests that there may be real differences in perceptions between these two groups. The data we analyzed in this study was from a free text box that was left to give participants space to elaborate on their Likert scale responses - this is an area that – as yet - has not been further investigated in studies of parents.

Future directions

This study was a follow-up to our previous survey investigating the opinions of Autistic adults on genetic testing for autism. The volume that respondents wrote, along with the depth of their responses, indicate that Autistic adults want to express their opinions on topics like genetic testing but have rarely had the opportunity. Future work can continue to engage with Autistic adults.

Despite the generally negative sentiment expressed by most participants, some talked about potential benefits that they could see come from genetic testing and gave suggestions. Many of these focused on future direction for research: focusing on supports and comorbidities, and centering Autistic researchers.

Finally, our study had the limitation of using social media exclusively as a recruitment strategy. While social media is an excellent tool for building community and reaching a large audience quickly, it also has the potential to push users toward groups which reaffirm their ideas. It is possible that Autistic adults who are not active in social media communities online have different perspectives than those in our survey. Future work should endeavor to reach Autistic individuals in other contexts to gather diverse perspectives outside of the confines of social media.

Conclusions

This is the largest study that we are aware of that has investigated the opinions of Autistic adults on genetic testing for autism. Most respondents expressed serious concerns about how genetic testing for autism may be used in the future, citing eugenics, discrimination, and a lack of trust in society as their fears. Despite these concerns, participants also identified potential benefits to testing, and offered suggestions about how resources can be allocated to best align with the wishes of Autistic people. Future work can continue to characterize the opinions of other groups of Autistic adults, and focus on supports, comorbidities, and Autistic leadership.

Author contributions

All authors contributed to the intellectual design, the drafting, revision, and final approval of the manuscript.

Funding

This research did not have funding support.

Data availability

Data are available from the corresponding author on reasonable request.

Code availability

The data generated and analysed during this study are available from the corresponding author on reasonable request.

Competing interests

MH, LB, and JA declare that they have no conflicts of interest.

Ethical approval

This study was approved by the BC Children and Women’s Research Ethics Board (H21-01937). All participants provided informed consent.