A Qualitative Study Exploring the Acceptability of Taking Part in a Large Multicentre RCT of Medication for Anxiety in Autistic Adults (the STRATA Trial) Authors: Alba X. Realpe, Jade Eloise Norris, Ava Lorenc, Leonora Cotton, Zoe Morgan, Aws Sadik, Dheeraj Rai, Nicola Mills, David Carmichael Published: 2026 ABSTRACT Autistic adults experience significant physical and mental health inequities yet remain underrepresented in clinical research, with few randomised controlled trials to guide care. Randomised controlled trials (RCTs) of selective serotonin reuptake inhibitors (SSRIs) are limited, underpowered, and rarely focused on anxiety. Anticipating recruitment challenges in a large RCT (“STRATA”) evaluating sertraline for anxiety in autistic adults, we embedded qualitative research to support recruitment, retention, and monitoring trial acceptability. We organised our findings into the theoretical framework of acceptability (TFA) constructs to assess the acceptability of trial design and delivery for autistic adults. We conducted 64 interviews with autistic adults at different trial stages. Data were analysed thematically and mapped to the seven TFA constructs. Participants considered involvement in a blinded medication RCT acceptable across the TFA domains, which they weighed differently when reflecting on anticipated versus experienced aspects of participation. STRATA was a low-burden, ethically sound, and methodologically coherent study for most participants, who reported minimal trade-offs, potential benefits, and self-efficacy in managing anxiety and research participation. Acceptability of trial participation is dynamic and multidimensional, which can be enhanced by meaningful involvement of autistic people throughout the research cycle, accessible participant information design, and responsive ongoing engagement.Lay AbstractAutistic adults often experience poorer physical and mental health than the general population. Yet they are rarely included in clinical research. There have been very few high-quality studies (RCTs) testing medications for anxiety in this group. Most existing studies are small and focus on other outcomes. They don’t provide clear guidance for care. To help address this gap, the STRATA trial tested whether the medication sertraline (an SSRI) can reduce anxiety in autistic adults. Recruiting participants for such trials can be challenging. We included a qualitative study to better understand what helps or hinders people from joining and staying in the trial. We aimed to explore what aspects of the STRATA trial made it easier or more appealing for autistic adults to take part. We used a framework called the theoretical framework of acceptability (TFA) to define acceptability in this context. We interviewed 64 autistic adults at different stages of the trial, including 2 who chose not to take part. Most participants found the trial acceptable when assessed against the seven aspects of the TFA (i.e., how someone feels about taking part, how much effort is needed, whether taking part fits with a person’s values, how well someone understands the study, what someone may have to give up, whether the study is likely to help, and how confident someone feels about taking part). In summary, they felt positive about taking part. They thought the study was ethical and easy to understand and believed it could benefit them. Many also felt more confident in managing their anxiety and contributing to research. STRATA is one of the largest studies of its kind; 318 autistic adults took part across the United Kingdom and Australia. The trial had a very high retention rate. Ninety-two per cent of participants stayed until the main outcome point, and 87% completed the full 52 weeks. How acceptable clinical trials like STRATA are may change during their course, and researchers need to be responsive. To do this well, researchers should involve autistic people meaningfully throughout the research process and from an early stage. Researchers also need to respect individual communication needs and provide clear and accessible information. These approaches were central to STRATA and supported by other studies. Introduction Trial recruitment can be hindered by various factors, such as attitudes towards research and trial procedures, communication barriers, and perceived value of participation (Hughes-Morley et al., 2015; Rodríguez-Torres et al., 2021), which make a trial more or less acceptable to potential participants and influence retention (Anastasi et al., 2023; McCann et al., 2010). Systematic analysis of inclusion criteria and recruitment procedures in clinical trials indicates that people with intellectual disabilities and/or autism are frequently excluded, either explicitly or via lack of accessible recruitment (Hamilton et al., 2017). Qualitative research plays a critical role in understanding recruitment and retention in clinical trials, especially when barriers may be significant, such as with underserved populations (de Salis et al., 2008; Howard et al., 2009). Autistic adults are an underserved population who experience disparities across various health and mental health outcomes, including early mortality (Croen et al., 2015; Hirvikoski et al., 2016). Despite a clear need to develop evidence-based interventions to support autistic adults (Lord et al., 2022; NHS England, 2023), there is a lack of well-powered randomised controlled trials (RCTs). Although selective serotonin reuptake inhibitors (SSRIs) are used in autistic populations (Jobski et al., 2017; Sadik et al., 2025), there is evidence to suggest they may work differently in autistic people than in nonautistic people (Muller et al., 2016). Only three RCTs of SSRIs in autistic adults have been conducted to date (Buchsbaum et al., 2001; Hollander et al., 2012; McDougle et al., 1996), but they had small samples and a focus on repetitive behaviours rather than anxiety. In response, the STRATA RCT (Rai et al., 2024) was designed to determine the clinical and cost-effectiveness of the SSRI sertraline in reducing symptoms of anxiety in autistic adults. STRATA recruited 318 people aged 18 years or over with a diagnosis of autism and anxiety as measured by GAD-7 (Generalised Anxiety Disorder Assessment, Spitzer et al., 2006). They were recruited from autism diagnostic services grouped in four regional centres in the United Kingdom and one in Australia. Potentially eligible participants completed a preliminary online screening questionnaire and expression of interest form based on broad inclusion/exclusion criteria. If minimal requirements were met, permission to contact the individual’s primary care doctor was sought to check if it would be safe for the patient to potentially take sertraline. If the individual was potentially eligible, they attended an online or telephone appointment with a trial researcher to discuss the trial and provide informed consent. Each participant was in the trial for up to 1 year, with regular safety checks at specified intervals to assess adverse effects, medication dose titration, and anxiety and depressive symptoms. The research team, clinicians, site staff, and participants were all blinded to the allocation of treatment group(s) (sertraline/placebo). Extensive preparation informed the design, with a strong emphasis on coproduction with the autistic community, researchers, and clinicians (Realpe et al., 2025, 9 July). Notably, 87% of participants completed the 52-week follow-up questionnaires, demonstrating a high retention rate. Prior to STRATA, we undertook a qualitative study to explore autistic adults’ views on trial participation based on a hypothetical RCT testing an active medication versus placebo to treat anxiety (Beasant et al., 2023). Autistic adults found the hypothetical trial processes acceptable and linked positive attitudes towards RCT participation to autistic peoples’ heightened sense of fairness, preference for evidence-driven knowledge, and belief in science as a common good (Beasant et al., 2023). However, they also raised concerns that suggested that conducting large-scale RCTs with the autistic population could be challenging, due to factors such as (1) conducting an RCT relies on uncertainty, that is, randomisation and concealed treatment allocation (blinding), which could be difficult for autistic people who often prefer control and predictability (also in Nicolaidis et al., 2020; Robertson et al., 2018); and (2) RCT designs may be incompatible with perceived physical and mental health vulnerability and a history of misunderstanding and exclusion and distrust of health care professionals (Also in Lord et al., 2022). To support STRATA’s ongoing recruitment and retention, we embedded qualitative interviews with autistic adults to understand their views and experiences of trial participation, as well as to monitor the acceptability of trial procedures in real time. Acceptability in health care research often lacks precision because, as a concept, it is context-specific, variable, and subjective (Gooding et al., 2018). Although implementation frameworks such as Proctor’s Implementation Outcomes Framework and the Consolidated Framework for Implementation Research (CFIR) consider acceptability, they do so in the context of multilevel implementation outcomes and determinants (e.g., adoption, appropriateness, and organisational context; Damschroder et al., 2009; Proctor et al., 2011). The theoretical framework of acceptability (TFA) offers a conceptually explicit, multicomponent account of acceptability at the individual level and can be applied prospectively and retrospectively to structure analysis of participants’ experiential appraisals (Sekhon et al., 2017) and is therefore well suited to our focus on the narratives of autistic people in relation to trial participation. The aim of this study was to investigate the acceptability of trial design and delivery that encouraged autistic adults to enrol and remain engaged with STRATA and what lessons can be learnt, including aspects deemed unacceptable. Analyses were also carried out to understand the experiences of the trial intervention (sertraline or placebo), which are reported elsewhere (Norris et al., 2026). Findings may help other researchers optimise the design and conduct of future RCTs for autistic adults and other underserved populations. Method This study is reported using the COREQ guidelines (Tong et al., 2007). The research was part of the qualitative assessments embedded in the STRATA trial (Rai et al., 2024). Qualitative research alongside clinical trials has long been recognised as a valuable approach for gaining in-depth understanding of beliefs and perceptions surrounding key medical events (Lewin et al., 2009; Skivington et al., 2021). Participants Autistic adults eligible to participate in STRATA were invited to a semistructured interview with a researcher about their experiences related to the trial. Interviews took place either during their trial participation or at the 52-week final follow-up appointment (henceforth “exit interviews”). We also invited study participants who withdrew from taking the study medication and people who declined participation in STRATA. Sampling and Data Collection The study received ethics approval from the Oxford B Research Ethics Committee (22/SC/0296). Data collection took place in two phases. In the first phase, interviewees were purposively selected to maximise variation in gender, level of education, presence of intellectual disability, ethnicity, employment status, and time in the STRATA study. Interviews conducted during this first phase focused on trial delivery, with the aim of identifying barriers to recruitment and informing remedial actions to be implemented. As the trial progressed, recruitment targets were being met, and retention was high, with participants consistently completing the 52-week follow-up. A second phase of data collection was therefore introduced, involving exit interviews at the final 52-week appointment. These interviews explored participants’ views and experiences of their continued participation in STRATA, as well as reasons for remaining in the study to the end. In both phases, interviewees were contacted predominantly by email from interviewers and directed to the original STRATA Information Leaflet that contained information about the qualitative assessments embedded in STRATA. If they agreed to take part, they completed an electronic consent form. All interviews were undertaken virtually through interviewees’ preferred method (email, telephone, or videoconferencing) and audio-recorded. Audio files were transcribed, either verbatim or via auto-transcription, checked for accuracy, and anonymised. For email interviews, questions were emailed to interviewees in batches of three or four questions, plus follow-up/clarification questions regarding previous responses. Questions and prompts explored interviewees’ decisions on whether to take part in the trial; their understanding and acceptance of study processes; and, if applicable, of the treatment received. The interview topic guide was initially developed by an experienced trial researcher (NM) with support from the trial team and autistic advisors to monitor potential issues with recruitment or retention at the start of the trial. The guide was used flexibly and adapted as the research progressed to explore developing findings (see Supplementary Files A and B for final versions). Interviews with trial participants and decliners were conducted by two experienced qualitative researchers (AL, AR) and a psychiatry resident doctor (AH) with varied experiences of interviewing autistic people. Exit interviews were undertaken by trial researchers (JN, LC, ZM), who were responsible for follow-up appointments and, therefore, mostly known to the participants. Five interviewers were female and one male; none had an autism diagnosis, and all described themselves as nonautistic. Data Analysis Ongoing trial interviews were initially and independently coded and analysed within NVivo (Lumivero, 2023) by AL, and the 52-week exit interviews by JN using reflexive thematic analysis (Braun & Clarke, 2019). Codes were drawn mainly inductively from the data, with some drawn from the interview questions, and synthesised using familiarisation, coding, generating, and developing themes (Braun & Clarke, 2019). Data were examined for credibility, context, language, negative cases, and rival explanations (Patton, 1999). Another experienced member of the qualitative team (AR) independently analysed a proportion of transcripts to assess the dependability of coding. The first authors (AR, JN) then met to develop a shared coding framework. They applied this framework to both datasets and combined the coding, resulting in two separate code sheets: (1) experiences of RCT trial processes (the focus of the current article) and (2) experiences of taking a medication (placebo or sertraline) as an anxiety treatment in the context of an RCT, reported elsewhere (Norris et al., 2026). Coding consistency was supported by cross‑checking of coded segments and regular calibration meetings, during which the team reviewed code definitions, merged or split overlapping codes, and refined inclusion/exclusion criteria. Disagreements were resolved by consensus; if consensus was not reached within the core analytic team, the issue was referred to the wider team, which included a senior qualitative researcher (NM) and academic psychiatrist specialising in autism (DR). The wider team met regularly to review coding and descriptive findings, agree on further sampling, and discuss theoretical development. During the analysis, we agreed to introduce an adapted version of the TFA (Sekhon et al., 2017) to deepen our understanding of the acceptability of participating in STRATA from the point of view of interviewed participants. We selected the TFA because it proposed a multifaceted construct to define acceptability based on people’s cognitive and emotional reactions to receiving said intervention, which fit the overall nature of the data collected (i.e., people’s cognitive and emotional reactions to trial participation). We sought to map our inductive coding onto the TFA constructs that became higher-order categories in our coding framework. Negative examples and alternative explanations were included in the description of our findings under each of the TFA categories. Themes were then reviewed against the coded data and the full data, named and defined prior to write‑up. The core interpretive material for complex constructs came primarily from the longer interviews conducted during recruitment, while shorter interviews were treated cautiously, using them to corroborate or nuance themes rather than to drive interpretation. However, we needed to balance our approach, as autistic people often tend to provide shorter answers in response to questioning (Norris et al., 2020). This process enhanced analytic transparency and credibility by balancing independent coding, iterative comparison, and team‑based consensus building. Table 1 shows the adapted version of the TFA constructs to fit our assessment of the acceptability of trial participation. Table 1.Applying the Seven Constructs of the Theoretical Framework of Acceptability (TFA) to Trial Participation.ConstructDefinitionAdapted definitionAcceptabilityA multifaceted construct that reflects the extent to which people delivering or receiving a health care intervention consider it to be appropriate, based on anticipated or experienced cognitive and emotional responses to the intervention.Whether autistic people considered taking part in STRATA appropriate based on anticipated or experienced cognitive and emotional responses to trial participation.Affective attitudeHow an individual feels about the intervention.Affective attitude towards trial participation How individuals felt about taking part in STRATA.BurdenThe perceived amount of effort that is required to participate in the intervention.Burden of trial participation Their perceived amount of effort to participate in the trial.EthicalityThe extent to which the intervention has a good fit with an individual’s value system.Ethicality of participating in an RCT The extent to which the trial has a good fit with an individual’s value system.Intervention coherenceThe extent to which the participant understands the intervention and how it works.Intervention coherence of the trial design The extent to which the participant understands the trial and how it works.Opportunity costs (trade-offs)The extent to which benefits, profits, or values must be given up by engaging in the intervention.Opportunity costs of trial participation (emotional and cognitive trade-offs) The extent to which benefits, profits or values must be given up by engaging in trial participation.Perceived effectivenessThe extent to which the intervention is perceived as likely to achieve its purpose.Perceived effectiveness of trial participation The extent to which trial participation is perceived as beneficial and achieving its purpose.Self-efficacyThe participant’s confidence that they can perform the behaviour(s) required to participate in the intervention.Self-efficacy in relation to trial participation The participant’s confidence that they can perform the behaviour required for trial participation.Source: Adapted from Sekhon et al. (2017). Applying the Seven Constructs of the Theoretical Framework of Acceptability (TFA) to Trial Participation. Source: Adapted from Sekhon et al. (2017). Results During the first data collection phase, we invited 64 trial participants and 3 people who had declined participation in STRATA to an interview. Thirty people did not respond to the invitation. Two trial participants declined an interview as it was too burdensome, one scheduled an interview but did not attend, and one offered to be interviewed but was not available before the first phase of data collection finished. Although we planned to interview more people who declined participation, a feature of the trial made it difficult to approach them. Namely, most potential participants accessed the online expression of interest form, but unless the person agreed to submit their personal details, these were completely anonymous at this stage. We interviewed 31 individuals when they were 1–26 weeks into trial participation and 2 who declined to take part in STRATA between February and June 2022. A second phase of data collection was led by trial researchers, who invited participants to an exit interview at their final 52-week follow-up appointment. When exit interviews were introduced, 144 trial participants were still due to complete the trial. Forty could not be approached because their site was not included in the qualitative study, and 20 did not respond to invitation emails. Of the remaining 84 participants, 31 consented to exit interviews between February and October 2024. Participant demographics (N = 64) are presented in Table 2; further details are available in Supplemental Material C. The sample broadly reflects that seen for the main STRATA trial (Rai et al., manuscript in preparation). Most interviews were conducted online using videoconferencing (n = 53), nine by phone, and three by email. The average length of interviews was 16 min (range 5–33 min). Table 2.Summary Demographics for Interviewees.CharacteristicN = 64 a AgeMean = 36.80; SD = 13.22 (range = 19–70)Year of autism diagnosisMedian = 2020 (range = 1999–2023)GenderMale = 33Female = 26Other = 5EthnicityWhite = 57Other = 7Level of qualificationNone = 4GCSEs or equivalent = 8A levels or equivalent = 12Bachelor’s degree = 21HNC/HND/SVQ = 3Postgraduate degree (e.g., MA, MS, MEd, PhD) = 12Professional degree = 2Employment statusFull-time = 27Part-time = 9Self-employed = 5Student = 9Looking for work = 3Unable to work = 6Retired = 1Other = 2Note. SD = standard deviation; GCSEs = General Certificate of Secondary Education; HNC/HND/SVQ = higher national certificate/Higher National Diploma/Scottish Vocational Qualifications; MA = Master of Arts; MS = Master of Science; Med = Master of Education; PhD = Doctor of Philosophy.aOnly gender and ethnicity data were available for decliners (N = 2). Summary Demographics for Interviewees. Note. SD = standard deviation; GCSEs = General Certificate of Secondary Education; HNC/HND/SVQ = higher national certificate/Higher National Diploma/Scottish Vocational Qualifications; MA = Master of Arts; MS = Master of Science; Med = Master of Education; PhD = Doctor of Philosophy. Only gender and ethnicity data were available for decliners (N = 2). Findings Overall, most interviewees who took part in STRATA considered participation in a blinded randomised controlled trial of a medication acceptable when assessed against the seven constructs of the TFA: affective attitude, burden, ethicality, intervention coherence, opportunity costs, perceived effectiveness, and self-efficacy. Although views were predominantly positive, participants described combining and weighing these constructs in different ways depending on whether they were discussing anticipated (e.g., initial interviews) or experienced (e.g., exit interviews) aspects of trial participation. Results are presented under subheadings corresponding to the seven TFA constructs. For each section, we outline facilitators and challenges to acceptability and provide illustrative quotes in Box 1. Box 1.Illustrative Quotes of Acceptability of Trial Participation.3.1 Affective attitude towards trial participation [I feel] Incredibly positive. I am experiencing a reduction in anxiety and . . . whatever the results of that are I am happy to be experiencing this and also I think it’s important to do research. [P13, during trial] I think it was more of a curiosity thing than anything, I did a lot of research into blind studies and things when I was younger, it’s always something I’ve been interested in anyway. [P36, exit interview] Since early teen years, being given various different types of medicine and nobody listened, they didn’t work. When the trial was offered, I’d recently started my ADHD meds and I suppose mentally was in the best place I’ve ever been in. [P54, exit interview] I just wanted to make sure [invitation to the study] was legit. Because anybody could have paid that advert up. But it all seemed above board to me. Especially when I did speak to [researcher] initially. She helped satisfy any curiosities I had about it. It’s just the fact that it was backed by a university and it was happening in Australia as well. So, it seemed like there must be some pretty big funding for it to be happening internationally as well. [P22, during trial] [trial processes] have been so consistent . . . when you suffer from this [anxiety], consistency and things being really manageable and things being really dependable are huge factor you know. So in that respect it’s been like the perfect kind of study environment. [P62, exit interview] I’ve always felt listened to, I never felt uncomfortable, I’ve actually enjoyed the whole process. [P57, exit interview] 3.2 Burden of trial participation It’s been really straightforward so far. It’s been very clearly outlined. Obviously start from the participant information, so again we’ve got all the headers that are in green, anything that’s important or I would assume is important they’re kind of in bold, you’ve got things that are underlined. Again like contact information, it’s got its own blue box and stuff, so all that was clear. And then even in the interviews that I’ve had it’s been very clearly outlined so these will be the next steps, so again, yeah, very clear, can’t really fault it I don’t think. [P3, during trial] I have difficulty understanding my feelings and my emotions, and just having somebody to talk about that, I think was just really helpful. [P18, during trial] I find it a lot easier when it’s not in person because I don’t have to worry about travelling anywhere, I don’t have to worry about arriving to places on time. [P10, during trial] The medication being posted out and there wasn’t too much kind of effort needed to do it successfully, shall we say, you know, the medication’s sent out. We were doing the meetings on Teams. [P62, male, exit interview] [The questionnaires] are easy to do, even on my phone. The format is fine, and frequency too. [P25, during trial] [Questionnaire] was easy to do though sometimes get a little tedious. It kind of feels like you’re being asked the same things over and over again and a lot of the questions I’d want to answer “it depends” but you can’t put that as it’s multiple choice. Also sometimes lists of text and boxes to tick can be a bit hard to read. [P2, during trial] Hesitation around symptom disclosure The only thing is that I’m saying my symptoms etc, but often my symptoms vary throughout the month – typically affected by hormones (very annoying) – so if the questionnaire gets me on the week I’m due my period the results are potentially going to be quite skewed, but I guess that’s hopefully going to get taken into account as I do always make a note about it. [P25, during trial] 3.3 Ethicality of participating in a blinded medication RCT I thought, if I can save anybody from going through that process [of anxiety] by helping in some way, it would kind of make me feel better. [P32, during trial] [the study’s aim] was quite relevant to issues that actually I cared about (. . .) there are studies on autism where it’s just like the bits I don’t care about, like, “how can they communicate better” and stuff like that. And I’m just like “I don’t care if I communicate badly” (. . .) But the way that anxiety and stuff actually does affect people with autism rather than the people around them.[P12, during trial] I think [the study] is a good way in [to start taking antidepressants for anxiety]. Because if you take it via a doctor, they just throw it at you and say: “Try that. See how you get on”. And at least the study, you have a point of contact . . . And somebody’s checking in on you. Which you don’t get anywhere else. [P32, during trial] It’s the bigger [sic] study of its kind in the context of sertraline and autism, so again if I can just do my bit that’s good enough for me. [P3, during trial] You can see on your leaflets you’ve got five people who are autistic on your advisory thing so yeah, I think, that’s obviously a good thing. [P7, during trial] [study results] will filter into NICE guidelines and things like that, whether Sertraline is actually effective for autistic adults who have got anxiety, so again I’m very happy that I came across the study by chance and then able to participate in it. [P3, during trial] [The study] makes absolute sense but I’m afraid I’m very opposed to the, what I see as, the medicalisation of autism. [P16, declined taking part in trial] 3.4 Intervention coherence of the trial design [This study aims] to help target the right medication for the right problem and . . . to also find if autistic people respond differently to the medication. [P11, during trial] I googled sertraline and I read the Wiki page about it, and I was just reading lots of stuff from the NHS pages and just trying to understand more about the drug . . . I wanted to educate myself so I knew what to think about sertraline. [P27, during trial] The need for placebo/blinding is clear, as with most trials, in that there needs to be an identifiable benefit in those taking the medication vs those on placebo to ensure that it’s the medication having the effect rather than other factors. [P25, during trial] The need for placebo/blinding is clear, as with most trials, in that there needs to be an identifiable benefit in those taking the medication vs those on placebo to ensure that it’s the medication having the effect rather than other factors. [P25, during trial] [Not knowing which arm I’m in is] going to bug me for the whole year . . . But I am, I accept it because I decided to do the trial, I knew I’d get one or the other. [P7, during trial] With a lot of autism there’s that need to feel in control, so the fact that I wouldn’t know whether it was the placebo or the sertraline would really play with my mind . . . it’s that need for control, definitely. [P16, declined taking part in trial] 3.5 Opportunity costs of trial participation (emotional and cognitive trade-offs) I was especially worried about taking medication again. The fact that it was mentioned several times that I could stop taking anything or stick to a dose that I was comfortable with was a huge help. [P29, during trial] [Not knowing which arm I’m in is] going to bug me for the whole year . . . But I am, I accept it because I decided to do the trial, I knew I’d get one or the other. [P7, during trial] I feel more anxious but I don’t know if that’s simply because I’m focusing in on it more. [P11, during trial] I’m actually fine with [the uncertainty] because I knew in advance (. . .) I wasn’t going to know until the end of the trial. I’m more curious I think than bothered about it. [P6, during trial] I don’t want to get into the mindset of I’ve got a placebo, it’s not going to work, ‘cos I don’t want that to affect it (. . .) there’s a whole year left. Things can change so I’m just trying to keep balanced and unbiased thoughts and feelings about how the drug’s affecting me. [P27, during trial] You’re given so many different options [in questionnaires], it can be difficult to [answer], because I think some of the things are quite subjective in a way, so it’s like it’s quite a grey area, what I’m trying to say, so it can be difficult to put your finger on the exact sort of answer [P1, during trial] With a lot of autism there’s that need to feel in control, so the fact that I wouldn’t know whether it was the placebo or the sertraline would really play with my mind (. . .) it’s that need for control, definitely. [P16, declined taking part in trial] I feel slightly embarrassed that I couldn’t continue. And a little bit of me is annoyed with myself for not persevering with it. Which is something that I tend to do. So, it’s almost like a little failure that I didn’t carry on. [P32, during trial] 3.6 Perceived effectiveness of trial participation on anxietyI’ve been wanting to try something to get my anxiety under control because the older I get the worse its getting and then with Covid. [P8, during trial] I really enjoyed the introspection and the discussions, and just getting a better understanding of not only me, but the neurology behind it, and all the all the bits and bits and pieces you’ve talked to about how it works, the study itself not the medication. [P48, exit interview][The study] didn’t meet my expectations, it exceeded them, because like I said at the start, I was more going into this as like “let’s see what this does” because I, it sounds harsh. I didn’t expect nothing, so actually going in and having the space to just open and honestly, I really genuinely enjoyed it, it’s helped me in that way. [P42, exit interview]3.6.1 Self-efficacy in relation to trial participationAnd I feel like even just knowing I’m on [the study], taking pills every day gives me a chance to remember that I’m doing something for myself so I feel better about myself by taking part in it. Like I’m being responsible, I’m taking a chance. [P33, during trial] I probably could go year to year without thinking about my physical or mental health until something happens whereas actually having to sit down and go “yeah, well, how was I feeling for the last 2 weeks?” actually helps. Cause it’s not always negative. If you wait until something bad happens before you start thinking “how am I feeling”? or “what am I experiencing” then, you know, you’re always going to see things in a negative light. If you’re just going right, it’s 3 months, I will tick all these boxes or whatever yeah, it can throw up positive things as well. [P53, exit interview] Part of it I was actually thinking earlier was just kind of stubbornness, I want to be able to say that I finished something. Part of it was seeing, you know, if there are any changes or developments. [P59, exit interview] A couple of years down the road, I think I’ll definitely do some sort of blind study again because again it it’s helping. It could help not only me, but thousands, millions of other people [P36, exit interview] Illustrative Quotes of Acceptability of Trial Participation. Affective Attitude Towards Trial Participation Feelings about taking part in the RCT were almost all positive – happy, no regrets, incredibly positive, really good decision, and most would recommend the study to others and would make the same decision again. Furthermore, some interviewees were driven by academic curiosity. They were “fascinated” by, “interested” in, and “curious” about the concept of RCTs, the placebo effect, or scientific research in general. A few participants appeared more neutral in relation to their emotional reactions to trial participation, but we did not observe reports of negative feelings, with the exception of the two people who did not go on to become trial participants, as expected. When asked about their initial reaction to the trial invitation, interviewees mentioned that the study invitation arrived at a time that was right for them, for example, after receiving their autism diagnosis, when needing support, or because they felt empowered to try something to help themselves. They remarked on their satisfaction with the study communication and information, with concerns adequately addressed by the study team, creating an initial positive reaction towards trial participation. As the trial progressed, participants reported enjoying the regular contact with trial researchers and appreciated the supportive and caring trial staff, who communicated clearly without being overwhelming. Overall, participants’ affective attitudes were predominantly positive, characterised by feelings of reassurance, support, and emotional comfort associated with taking part in the trial. Burden of Trial Participation Most participants reported that joining the study was straightforward. Invitations to the study came through accessible channels, such as social media, email, or direct recommendations from clinicians, family, or other trusted sources. The clear and concise written information (e.g., the Participant Information Leaflet), with key points highlighted, helped participants make informed decisions. Interviewees remarked on how the offer of completely remote participation probably allowed them to take part by reducing anxiety, for example, associated with travelling to new environments. Participants identified several features that helped them to remain engaged after joining the study. The study’s flexible and accessible design made it easy for participants to engage, with minimal disruption to their daily lives, facilitated by the participants integrating the study’s processes into their daily routines (Norris et al., 2026). Completing questionnaires was described as straightforward, even on mobile devices. However, some participants reported struggling with thoughts and feelings around side effects of the medication, which are explored in more detail elsewhere (Norris et al., 2026). Participants also noted a mild cognitive burden associated with self-assessment when completing questionnaires. A few participants found the self-report process of completing the study questionnaires a little tedious. A small number of participants hesitated to disclose certain symptoms, worrying these may be misinterpreted as concerning, despite being normal for them. One participant noted that poor interoception (bodily awareness) in autistic individuals added to difficulties in reporting symptoms accurately. Finally, a handful of participants encountered communication challenges. Some were uncomfortable with video or phone calls, disliked speaking with unfamiliar researchers, or found the paperwork cumbersome. Although participants identified some cognitive, practical, and communication-related demands, these were generally described as manageable and did not substantially hinder continued participation for those interviewed. Ethicality of Participating in a Blinded Medication RCT As expected, the perceived moral appropriateness of the trial played an important role in the decision to take part. Most interviewees viewed the study as ethical and legitimate and that it fitted with their individual values. They recognised the need for this research to advance the evidence base, particularly given the prevalence and impact of anxiety in autism. Individuals valued participation in this specific trial because of a desire to reduce anxiety in autistic adults. Many had struggled with anxiety for a long time or were experiencing high anxiety levels that were disrupting multiple areas of their lives. Some described the impact as crippling or consuming my life. Many participants also appreciated that the study involved autistic people in its design, as well as its focus on issues relevant to the autistic community. This inclusivity added credibility and made the study feel more meaningful. Some specifically highlighted how the study addressed anxiety in autistic people, rather than aiming to “fix” communication differences or other aspects of autism they did not see as problematic. Furthermore, trial follow-up appointments allowed discussions about potential side effects in detail, helping them to feel safe participating in the trial. In contrast, the two interviewees who had declined participation in STRATA reported having strong beliefs against taking medications for anxiety. They viewed the use of medication to treat anxiety in autistic people as a form of medicalising autism and argued that research should instead focus on making society more accommodating. Their convictions were strong enough to deter them from participating. A few participants raised concerns about ensuring the inclusion of people with intellectual disabilities or the representation of different gender identities. However, on the whole, these concerns seemed to have been sufficiently addressed during the sign-up process, easing their reservations. Overall, participation in the blinded medication RCT was widely perceived as ethically appropriate and value-congruent, with concerns largely addressed through transparent communication and inclusive study design. Intervention Coherence of the Trial Design Most participants demonstrated a clear understanding of the study’s design and purpose. Many appreciated the potential challenges of recruitment and were motivated to contribute, even when they had concerns about medication. Several reported reading extensively about sertraline and the study itself to make informed decisions. Only a few interviewees relied solely on the information provided by the study team. Several mentioned feeling reassured by thorough explanations from the researchers and detailed participant information. In terms of core RCT processes, most participants were comfortable with the placebo, blinding, and randomisation aspects of the trial. They displayed strong understanding of RCT methodology, including the necessity of having a placebo arm, the 50–50 chance of receiving either placebo or sertraline (randomisation), and the importance of blinding (participants, researchers, and clinicians) to ensure unbiased results. When discussing their experience of being in the trial, some found not knowing their treatment allocation to be frustrating, yet they were able to accept the rationale for this and to tolerate the uncertainty. Several participants even described the ambiguity as fun and exciting. Most participants demonstrated strong intervention coherence, showing clear understanding of the trial’s scientific rationale even when elements such as blinding generated discomfort. Opportunity Costs of Trial Participation (Emotional and Cognitive Trade-Offs) The interviews revealed some trade-offs associated with study participation. First, most interviewees in the initial interviews expressed some concerns about their personal reactions to taking sertraline, including potential side effects, drug interactions, or challenges when stopping the medication. These worries were often linked to past negative experiences, either personal, involving family members, or, in one case, informed by professional knowledge of psychotropic medications. However, most were sufficiently reassured by study researchers or their own support networks, as well as by the flexibility of being able to discontinue the study medication at any time. Second, a few participants who noticed positive effects on their mental health anticipated feeling disappointed or embarrassed if they were in the placebo group. One viewed this possibility negatively, saying they were always getting the short stick. Another anticipated feeling was frustrated at the thought of having invested their time for no direct benefit but still understood the scientific rationale. When discussing which treatment allocation they believed they had received, participants were evenly split between being unsure, thinking they were taking sertraline, and believing they were taking a placebo. Their guesses were largely based on perceived changes in anxiety, side effects (or lack thereof), and prior experiences of medication (discussed in detail in Norris et al., 2026). Some participants spoke about making a deliberate effort not to overthink their involvement or speculate about treatment allocation, to avoid unnecessary anxiety or unintentionally influencing their perceptions of changes in anxiety, as well as the trial results. However, for one of the interviewees who declined to take part in STRATA, blinding of their group allocation created significant anticipated discomfort. They found the lack of control over which treatment they would receive too difficult to accept and ultimately contributed to their decision against participation. Third, for the small number of participants who chose to discontinue the medication while remaining in the study, this decision carried emotional weight. Some described feeling that they had failed or worried about having invested time and effort without completing the treatment, although one participant described finding the choice to discontinue medication straightforward. Finally, some participants described opportunity costs related to emotional demands over time. A small number of participants reported that participation itself occasionally increased their anxiety, as regularly tracking symptoms made them more aware of anxious thoughts and bodily sensations. Participants mentioned the mental effort involved in remembering to complete questionnaires as feeling like a “chore,” while others found selecting responses from fixed-multiple-choice options challenging when they were unsure about how to capture their experiences. While participants described emotional and psychological trade-offs associated with uncertainty, risk, and withdrawal, these opportunity costs were generally viewed as acceptable within the context of trial participation. Perceived Effectiveness of Trial Participation For some participants, perceived effectiveness was linked not only to the medication itself but also to their broader involvement in the study. During the initial interviews, several participants anticipated that participation might offer more support than their previous experiences with antidepressant treatment (e.g., in primary care), where individuals felt they had been prescribed a new medication and left to “see how they get on,” which shaped expectations about what trial participation might offer instead. Participants’ past experiences with antidepressants and other medications (e.g., stimulant-type medications for co-occurring conditions such as attention-deficit/hyperactivity disorder (ADHD) varied widely, from very helpful to no benefit at all. However, these experiences did not deter participation. The trial was viewed as an opportunity to try sertraline and address their own anxiety, which was especially appealing to those who had not found other treatments beneficial. In terms of how participants experienced effectiveness, they described benefits associated with the structured and supportive nature of trial participation. Regular self-monitoring through questionnaires, alongside ongoing contact with the research team, helped some participants to better recognise and manage their anxiety. In particular, safety check appointments with researchers were valued and described as thorough and reassuring, with several participants finding them beneficial as an emotional check-in. Overall, participants’ perceptions of effectiveness were often attributed to the supportive and structured features of taking part in the trial rather than definitive judgements about the medication itself. These experiences contributed to a sense that participation could positively influence anxiety management, independent of treatment allocation. Self-Efficacy in Relation to Trial Participation Most interviewees reported gaining confidence and a better awareness of their mental health through participating in the study. They also felt more capable of managing the uncertainty of their involvement as time went on. Some participants expressed a strong determination to complete the study – not only because they recognised the value of contributing data until the end, but also due to feeling that they had made a commitment to follow through, a trait they often linked to being autistic. Discussion Our findings indicate that participation in STRATA was largely acceptable to participants when evaluated using the TFA. Although participants identified specific challenges and trade-offs associated with trial procedures, including safety concerns about medication for anxiety, these challenges were generally perceived as manageable and outweighed by emotional, ethical, and participation-related benefits. Acceptability was shaped by how participants understood, valued, and engaged with the trial design, highlighting the interrelationship between TFA constructs in practice. Acceptability also emerged as a dynamic process. Anticipated concerns were often addressed early, while new issues arose over the course of participation and were mitigated through responsive changes in study processes (e.g., clear communication and a flexible design). Participation in the trial appeared to strengthen self-efficacy, as participants reported increased confidence in managing both their anxiety and the demands of trial involvement. Overall, these findings support the TFA’s conceptualisation of acceptability as evolving over time, rather than static, in the context of clinical trial participation. Two interviewees declined participation due to views linked to concerns about medicalisation of autism and inadequate societal support. Mistrust of trials is a known barrier among underserved groups, including disabled people (Bodicoat et al., 2021), and may be accentuated in some autistic people who prefer looking at mental health problems through the lens of the social model of disability (Dwyer et al., 2024). Although STRATA targeted an autistic-community priority outcome related to improving anxiety and quality of life rather than, for example, treating autism per se, this framing may not have reached everyone. Anxiety significantly affects autistic adults (Hollocks et al., 2019; Robertson et al., 2018), many of whom are open to medication trials despite initial concerns (Norris et al., 2026). The timing of the STRATA invitation – often soon after an autism diagnosis for many participants – may have encouraged participation, as much recruitment occurred at autism diagnostic services (Realpe et al., 2025, 9 July). With limited access to postdiagnostic support (Norris et al., 2024), research can offer access to new treatments, improved health care, and opportunities to obtain a deeper understanding of autism (National Health Service England, 2021). Altruism, a desire to help other autistic people, and self-learning were also major motivators in this and previous studies (Haas et al., 2016). Furthermore, many autistic adults showed a strong understanding of research processes, which has been reported before (Beasant et al., 2023), and displayed academic curiosity, which may also reflect autistic traits such as attention to detail and hyperfocus (Ross et al., 1999), deliberative and logical decision-making (Lewton et al., 2019), and intellectualising to give meaning (Dachez & Ndobo, 2018). The overall acceptability of the trial may seem surprising given preconceptions about intolerance for uncertainty in autistic people (Boulter et al., 2014; Robertson et al., 2018), yet findings indicate that, at least for populations comparable to one involved in the STRATA, successful recruitment to large, complex trials with long-term follow-up is achievable. However, uncertainty may have been a barrier for those who decided not to take part, although data on this group are limited, as these individuals did not leave contact details to understand their views. Prior research highlights the clinician-patient relationship in shaping satisfaction and perceived effectiveness of antidepressants, especially at treatment initiation (NHS England, 2023; Norris et al., 2026). Our findings suggest that frequent contact with trial researchers fostered a therapeutic alliance, with participants reporting paying more attention to and controlling their anxiety, which appeared to improve their self-efficacy regardless of treatment allocation. As in many medication trials, participants expressed concerns about side effects, which are common barriers to study participation in the general population (Sheridan et al., 2020) and autistic people (Beasant et al., 2023). STRATA allowed participants to explore and reflect on their beliefs before deciding to take part, as well as being discussed in subsequent appointments. A few interviewees noted difficulties with questionnaires, particularly when they felt multiple-choice options did not capture their experiences (Nicolaidis et al., 2020). Such challenges may reflect limitations of standardised questionnaires in this population. STRATA involved several stages, including initial expression of interest; a check by the individual’s primary care doctor to ensure it was safe for them to take the study intervention and take part in the trial, a baseline appointment, eight follow-up appointments, daily medication, and completing multiple questionnaires over 52 weeks (Realpe et al., 2025, 9 July). Despite this higher burden than typical RCTs and the complexity involved in its delivery and the need to safeguard participants while maintaining rigour (Realpe et al., 2025, 9 July), participants generally found the trial accessible and engaging. Qualitative findings suggest that the intensive preparation and meaningful coproduction made STRATA autism-friendly, enhancing recruitment and retention. Findings support evidence that involving communities in meaningful coproduction can help overcome most research barriers (Crocker et al., 2018; Nicolaidis et al., 2011). Finally, we found that applying the TFA framework (Sekhon et al., 2017) to our inductively coded results helped to systematically define trial acceptability. Furthermore, applying the TFA contributed to explaining some individuals’ calculations around decisions associated with recruitment and retention in an RCT such as STRATA, as well as the gap between anticipatory and experienced aspects of acceptability. We can see the potential of this theory-informed, participant-centred approach to identify target behaviours (e.g., participant burden and study coherence) that can be modified so trial processes become more acceptable to potential participants in future trials (Coffey et al., 2022; Newlands et al., 2022). Strengths and Limitations Key strengths of this study included a large and varied sample of autistic adults and its analytical rigour, which was enhanced through multiple coders, team discussions, and interviews with both independent researchers and those known to participants. A few features of our work limit the conclusions we can draw about the acceptability of trial participation. The TFA was applied during analysis to provide a structured definition of acceptability, rather than informing the study design. Some constructs may therefore have been explored less directly, as the framework provided a structured analytic lens, supported by detailed team discussion to enhance rigour. The interviews were conducted via different modalities (e.g., verbal vs written text), potentially producing differences due to interactional dynamics. However, we prioritised a flexible approach to data collection to fit participants’ communication preferences and support inclusion of people who may not be comfortable speaking directly (e.g., phone/video call) with unknown researchers. There was limited data from decliners, which restricts our conclusions about nonparticipation. Despite purposive sampling, our participant group, derived from the host trial, predominantly consisted of White, employed autistic people without intellectual disabilities, and therefore closely reflected the trial sample. Although unintended, this mirrors long-standing patterns in autism research, which has overrepresented White, socio-economically advantaged, and geographically well-resourced populations (Cascio et al., 2021). Furthermore, ethnic minorities and socioeconomically disadvantaged people experienced barriers to diagnosis and service access (O’Nions et al., 2023; Roman-Urrestarazu et al., 2021; Sadik et al., 2025) and are therefore underrepresented among those invited to participate. These shortfalls reflect a broader problem across clinical trials (Peters et al., 2023) and have motivated inclusion-focused guidance such as NIHR’s INCLUDE initiative in the United Kingdom (Bodicoat et al., 2021; National Institute for Health Research, 2020). A further limitation concerns interviewer familiarity. Some exit interviews were conducted by researchers who already knew participants, while others were conducted by researchers with no prior contact. The consequences of this familiarity are uncertain. On one hand, participants may have felt less inclined to share negative feedback with a researcher they knew. On the other hand, rapport built over a 52-week trial period may have supported openness, enabling participants to provide a candid and nuanced account that they might have offered to an unfamiliar interviewer. The trial consistently emphasised trust and encouraged participants throughout to share any concerns openly. This context may have mitigated potential bias and supported the integrity of the data collected. Conclusion Our findings suggest that acceptability of participation in clinical trials with and for autistic adults is multifaceted and dynamic. To enhance participation acceptability, studies with this population should prioritise meaningful involvement of autistic people throughout the research cycle. Recommendations for future trials include coproduction and creating accessible information and materials at the trial design stage and further considerations for ongoing involvement and respecting individual contact and communication preferences, as demonstrated in STRATA and others (Cashin et al., 2025; Emerson et al., 2023). Future trials should include an intersectionality lens to be able to capture intersecting identities, involving more inclusive and diverse samples within the autistic community. Integrating qualitative research alongside RCTs can further deepen our understanding of more inclusive research practices. Supplemental Material sj-docx-1-aut-10.1177_13623613261466306 – Supplemental material for A Qualitative Study Exploring the Acceptability of Taking Part in a Large Multicentre RCT of Medication for Anxiety in Autistic Adults (the STRATA Trial) Supplemental material, sj-docx-1-aut-10.1177_13623613261466306 for A Qualitative Study Exploring the Acceptability of Taking Part in a Large Multicentre RCT of Medication for Anxiety in Autistic Adults (the STRATA Trial) by Alba X. Realpe, Jade Eloise Norris, Ava Lorenc, Leonora Cotton, Zoe Morgan, Aws Sadik, Dheeraj Rai and Nicola Mills in Autism