Introduction
Autistic people face elevated mental health conditions, including increased prevalence of depression and anxiety, as well as psychosis, compared to non-autistic people (Lai Reference Lai2023; Lai et al. Reference Lai, Kassee, Besney, Bonato, Hull, Mandy, Szatmari and Ameis2019). These co-occurring mental health conditions reduce quality of life (Mason et al. Reference Mason, McConachie, Garland, Petrou, Rodgers and Parr2018, Reference Mason, Mackintosh, McConachie, Rodgers, Finch and Parr2019) and increase suicide risk (Hirvikoski et al. Reference Hirvikoski, Mittendorfer-Rutz, Boman, Larsson, Lichtenstein and Bolte2016; Newell et al. Reference Newell, Phillips, Jones, Townsend, Richards and Cassidy2023). Consequently, there is an urgent need for accessible mental health services tailored to the needs of autistic people.
Autistic people experience significant barriers to accessing healthcare, including mental health services (Adams & Young, Reference Adams and Young2020; Camm-Crosbie et al. Reference Camm-Crosbie, Bradley, Shaw, Baron-Cohen and Cassidy2019; Crane et al. Reference Crane, Davidson, Prosser and Pellicano2019; Doherty et al. Reference Doherty, Atherton, Boland, Hastings, Hives, Hood, James-Jenkinson, Leavey, Randell, Reed, Taggart, Wilson and Chauhan2020; Malik-Soni et al. Reference Malik-Soni, Shaker, Luck, Mullin, Wiley, Lewis, Fuentes and Frazier2022). These barriers arise at multiple levels (Calleja et al. Reference Calleja, Islam, Kingsley and McDonald2020). Healthcare settings are often inaccessible, owing to patient-provider communication challenges, sensory differences, and a lack of autistic-informed care (Bradshaw et al. Reference Bradshaw, Pellicano, van Driel and Urbanowicz2019, Reference Bradshaw, Pickett, van Driel, Brooker and Urbanowicz2021; Doherty et al. Reference Doherty, Atherton, Boland, Hastings, Hives, Hood, James-Jenkinson, Leavey, Randell, Reed, Taggart, Wilson and Chauhan2020; Micai et al. Reference Micai, Ciaramella, Salvitti, Fulceri, Fatta, Poustka, Diehm, Iskrov, Stefanov, Guillon, Roge, Staines, Sweeney, Boilson, Leosdottir, Saemundsen, Moilanen, Ebeling, Yliherva and Schendel2021). These barriers are further compounded by service shortages and challenges navigating complex healthcare systems (Crane et al. Reference Crane, Davidson, Prosser and Pellicano2019; Gilmore et al. Reference Gilmore, Longo, Krantz, Radford and Hand2022; Malik-Soni et al. Reference Malik-Soni, Shaker, Luck, Mullin, Wiley, Lewis, Fuentes and Frazier2022), as well as limited training and knowledge about, and increased stigma of autism, among healthcare professionals (Bradshaw et al. Reference Bradshaw, Pellicano, van Driel and Urbanowicz2019; Micai et al. Reference Micai, Ciaramella, Salvitti, Fulceri, Fatta, Poustka, Diehm, Iskrov, Stefanov, Guillon, Roge, Staines, Sweeney, Boilson, Leosdottir, Saemundsen, Moilanen, Ebeling, Yliherva and Schendel2021).
Clinicians’ lack of knowledge also extends to the mental health needs of autistic people. Consequently, autistic adults report limited confidence in discussing mental health concerns, but also that when they do share them, their concerns are all-too-often dismissed, leading to inaccurate diagnosis and treatment options and adverse health outcomes (Au-Yeung et al. Reference Au-Yeung, Bradley, Robertson, Shaw, Baron-Cohen and Cassidy2019; Bradshaw et al. Reference Bradshaw, Pellicano, van Driel and Urbanowicz2019; Crane et al. Reference Crane, Davidson, Prosser and Pellicano2019). Furthermore, clinicians report low confidence treating autistic people (Brookman-Frazee et al. Reference Brookman-Frazee, Drahota and Stadnick2012; Lipinski et al. Reference Lipinski, Blanke, Suenkel and Dziobek2019, Reference Lipinski, Boegl, Blanke, Suenkel and Dziobek2022). Lack of professional knowledge and/or training therefore represents a major barrier to mental health services.
While providers report a lack of autism-specific training (Malik-Soni et al. Reference Malik-Soni, Shaker, Luck, Mullin, Wiley, Lewis, Fuentes and Frazier2022), studies of general practitioners (GPs) and psychiatrists in the UK have shown good knowledge of autism (Crane et al. Reference Crane, Davidson, Prosser and Pellicano2019; Unigwe et al. Reference Unigwe, Buckley, Crane, Kenny, Remington and Pellicano2017). For UK psychiatrists, greater knowledge of autism, as well as training and personal connections to autism, were correlated with greater confidence in working with autistic people (Crane et al. Reference Crane, Davidson, Prosser and Pellicano2019). Yet they also reported systemic and autism-specific factors that challenged their ability to provide expert care and support for autistic people, especially in mental health services (Crane et al. Reference Crane, Davidson, Prosser and Pellicano2019; Unigwe et al. Reference Unigwe, Buckley, Crane, Kenny, Remington and Pellicano2017). Improvements in specialist service provision, including the need for co-designed, accessible, and person-centred services, have been recommended (Crane et al. Reference Crane, Davidson, Prosser and Pellicano2019; Unigwe et al. Reference Unigwe, Buckley, Crane, Kenny, Remington and Pellicano2017).
In Ireland, mental health policy acknowledges the challenges autistic people face in accessing mental health services (A Vision for Change, 2006). Existing policy recommends that autistic people with co-occurring mental health conditions should have access to stepped mental health care, integrated care approaches and autism-specific training for mental health professionals. Yet, there are ongoing challenges in accessing mental health services for autistic people across the lifespan and a need for mandatory autism-related training for all healthcare staff (Final Report of the Joint Committee on Autism, 2023). To date, there has been no systematic approach to determining the autism-related training needs of psychiatrists in Ireland, nor any attempt to gather their perspectives on barriers and facilitators to mental health service provision. The current research therefore focused specifically on psychiatrists working in Irish mental health services to determine their (i) current level of autism knowledge; (ii) confidence in providing mental health services to autistic children and adults; and (iii) perceived barriers and facilitators to such provision.
Methods
Design
We used an online survey (adapted from Crane et al. Reference Crane, Davidson, Prosser and Pellicano2019) to gather quantitative and qualitative data from psychiatrists in Ireland. There were four sections: (1) Demographics; (2) Knowledge of Autism Scale; (3) Self-Efficacy; (4) Barriers and Facilitators to working with autistic patients.
Psychiatrists working in mental health services in Ireland in addition to other mental health professionals were eligible to participate. The results for psychiatrists only are presented here due to low response rates from other mental health professionals and lack of information about their professional background. Given these constraints, other mental health professionals were excluded due to difficulties in drawing valid conclusions from their survey responses.
Psychiatrists were recruited via the membership of the College of Psychiatrists of Ireland, is the sole body recognised for training doctors to become specialists in psychiatry, with the approval of the College Management Team. Members are psychiatrists in Basic Specialist Training (BST), Higher Specialist Training (HST) as well as consultant psychiatry roles across various psychiatry sub-specialties. A link to the survey was distributed to the College of Psychiatrists of Ireland mailing lists. The mailing lists contained contact information for 1056 members of the college. In addition, the survey link was posted to a WhatsApp group of senior registrars training in psychiatry in Ireland and members of the College of Psychiatrists of Ireland, (n = 167).
Participants
Overall, 168 psychiatrists and mental health practitioners responded. Respondents were largely consultant psychiatrists, but also included basic and higher specialist trainees (Table 1). Mental health practitioners (n = 28) were excluded in this analysis due to an inconsistent response from professional training bodies and difficulty identifying the scope of practice of respondents. (Supplementary Table 1). 140 psychiatrists completed sections (1) and (2), 128 completed sections (1), (2) and (3), and 90 respondents completed all four sections. Around two-thirds of participants’ caseloads consisted of adult patients (>18 years), and most provided care to those with average IQ or mild intellectual disability (ID) reflecting the organisation of mental health services in Ireland. While we do not know how many participants were autistic themselves (as our survey did not intend to collect personal health information from respondents), around half reported a personal connection to autism (i.e. they reported personal or lived experience of autism e.g. family member, friend or close acquaintance).
Participant characteristics (n = 140)
*Due to concerns about anonymity, the research ethics committee requested removal of gender and workplace variables from the survey and required that the age variable was gathered in age bands (<30y, 30–50y, >50y).
Materials
The survey comprised four sections (supp.). Section 1 (five questions) collected demographic information including age (<30 years, 30 – 50 years, >50 years), current role (Consultant, HST, BST, Independent Practitioner, Mental Health Practitioner), caseload age (0 – 18 years, 18 – 65 years, 18+ years, >65 years), caseload IQ (average/above average IQ, mild ID, moderate ID, severe ID, profound ID), and lived experience of autism (yes/no).
Section 2 presented 22 true-false statements assessing autism knowledge, with a score of ‘1’ assigned for each correct answer (max. 22; higher scores indicating higher autism knowledge). Following Unigwe et al. (Reference Unigwe, Buckley, Crane, Kenny, Remington and Pellicano2017), knowledge scores were adjusted for chance responding. The scale showed poor internal consistency (Cronbach’s α = 0.57, C.I. = 0.23 – 0.74) in the current sample.
Section 3 consisted of 19 statements assessing confidence when working with autistic patients on a scale of ‘1’ (not at all confident) to ‘10’ (extremely confident). Item scores were summed to yield a mean self-efficacy score; higher scores indicate higher confidence in working with autistic patients. The scale showed excellent internal consistency (α = 0.96, C.I. = 0.94 – 0.97).
Section 4 posed open-ended questions on perceived barriers and facilitators when working with autistic patients.
Procedure
The survey was open for responses between April 2021 and April 2022.
Data analysis
Quantitative analysis
Descriptive analysis was used to summarise data on respondents’ knowledge of autism and self-efficacy (Tables 2 and 3). Differences between groups of psychiatrists were analysed with two-sided, two-sample independent t-tests. To identify the explanatory variables underlying variation in perceived self-efficacy scores, we conducted a stepwise hierarchical, linear regression. Five models were constructed using a forward selection method in which variables were entered one at a time (Table 4).
Variables used in linear regression models
Number of correct responses to items on the knowledge of autism scale (n = 140)
Mean response scores on self-efficacy scale (n = 128)
IQ group, caseload age and current role were entered as categorical variables, personal/lived experience of autism as a binary variable, and knowledge score as a continuous variable. Multicollinearity was tested using a variance inflation factor threshold of >3 and diagnostic plots were used to assess the linear model assumptions. P-values were corrected using Bonferroni correction for five tests (p < 0.01). Finally, partial F-tests were performed to assess whether each model significantly explained more variance in self-efficacy scores than the prior model.
Qualitative analysis
Our qualitative analysis was based on Section 4 of our survey, where we posed open-ended questions on how participants felt about their own practice with their autistic patients, perceived barriers and facilitators for accessing diagnostic or mental health services and perceived training needs (see Appendix 1). To analyse participants’ responses, we used reflexive thematic analysis (Braun & Clarke, Reference Braun and Clarke2021) within an essentialist framework. The analysis was informed by the authors’ experience and training in psychology and/or psychiatry; and a neurodiversity lens to autism research and practice (Chapman & Botha, Reference Chapman and Botha2023; Pellicano & den Houting Reference Pellicano and den Houting2022). An inductive, bottom-up approach was used to identify patterned meanings within the dataset. LC and EP began by immersing themselves in the data, applying codes to each response. All authors met multiple times to discuss initial codes and decide on the final themes and subthemes. Analysis was iterative and reflexive (Braun & Clarke Reference Braun and Clarke2021).
Results
We identified three themes and associated subthemes (Fig. 1), all of which focused on the improvements psychiatrists felt were needed to ensure their autistic patients receive timely and effective care.
Thematic map.
Knowledge of autism
Respondents scored highly on autism knowledge (M scaled knowledge score = 20.01; 91%, SD = 1.91, 9%; range = 7.33 –22.00, 33 – 100%) (Table 4). There were no significant differences in knowledge scores between consultants and HST/BST (t[130.76] = 0.32, p = 0.75) or between psychiatrists >50 years and those <50 years, (t[108.1] = 0.53, p = 0.60). Similarly, scores did not differ significantly between respondents with and without a personal/lived experience of autism (t[155.51] = 1.44, p = 0.15).
Self-efficacy
Respondents’ self-efficacy varied widely within and across questions asked (Table 3). Consultants’ self-efficacy scores did not differ significantly from those of their HST/BST colleagues (t[126], p = 0.45) and there was no significant difference between psychiatrists >50 years and <50 years (t[126] = 0.61, p = 0.54).
Results of the hierarchical multiple, linear regression analysis showed that caseloads of 0–18 years and Moderate/Severe/Profound IQ significantly predicted greater self-efficacy scores in all models (Table 5). Current role (HST) and knowledge score were associated with self-efficacy scores, but not significantly after Bonferroni correction. Model 5 explained the most variance in perceived self-efficacy scores (F 119,118 = 6.31, p = 0.01).
Results of hierarchical multiple regression analysis investigating factors contributing to perceived self-efficacy scores amongst psychiatrists (n = 128)
Qualitative results
While some psychiatrists felt that progress was being made, most conveyed a strong sense of frustration about the ‘massive deficit in care for these patients and their families’. As one experienced psychiatrist put it: ‘I am sad after 45 years of service to know that people continue to experience great pain in accessing services’.
We identified three themes and associated subthemes (Fig. 1).
Theme 1. Nurturing strong, trusting therapeutic relationships
‘Having a good therapeutic relationship with the patient’ was taken as a given. Yet, despite suggestions that professionals need to treat autistic patients ‘the same way I interact with other people’, developing such relationships was felt to demand deeper care and attention. Respondents felt that the first step was generating a therapeutic space supportive of autistic experience (subtheme 1.1). This included engendering ‘a calm, uninterrupted environment’, and ‘promoting predictability’ in the structure of therapeutic sessions and in easing potential ‘discomfort’ around new settings and unfamiliar people. Integral to this supportive space was encouraging ‘open’, ‘honest’ two-way communication, where respondents adjust their own language so that it was ‘direct’, ‘clear and unambiguous’, as well as adapting to the ’style of individual patients’ ‘using a range of methods’, modalities (especially ‘visual supports’) and media (’I often use drawings/doodles with young people I know do not like direct questions or eye contact’) to ‘allow them express their needs’. One respondent felt that ‘changes that would benefit autistic patients would actually make for a much better work environment for other patients and staff’.
Being attentive to these issues and making the necessary adjustments demanded ‘patience and plenty of time’ (subtheme 1.2), which could be challenging within ‘busy clinical services’. Indeed, it was felt to be ‘almost impossible to give patients the time they need’. Yet, ‘taking the extra time to get to know the patient and their families’, was perceived as essential, especially for ‘identifying goals that are important to the person and their family’.
Respondents called for ‘a non-judgmental and empathetic approach’ (subtheme 1.3) – ‘listening attentively’ and being open to learning from their autistic patients. This approach was felt to be especially important for groups ‘not on the radar’, whose needs could be ‘complex’ and often overlooked (e.g., adults, women and girls, gender non-conforming patients). Some insisted that, since autism is a form of ‘neurodiversity, not an illness’, autistic differences should be respected. Some respondents emphasised redressing power imbalances within the therapeutic relationship, by ‘involving them in the decision-making around their care’. Working in this way was perceived to result in ‘high levels of mutual trust and understanding’.
Theme 2. Building capacity to provide the most effective support
Respondents described feeling ‘underprepared and lacking in confidence’ when working with their autistic patients; As one respondent explained: ‘I would like to be better equipped to help this population’ (subtheme 2.1). This lack of knowledge and experience was felt to extend to ‘health, education and social support staff’, who showed ‘limited team expertise regarding understanding communication/emotional needs of those with autism’. They wanted to see ’specific knowledge and training in’, and ‘proper exposure to autism’ in their multidisciplinary team members. Such experience was felt to ensure they and their multidisciplinary colleagues could prevent autistic people from ‘being misdiagnosed… and risking further mismanagement’. Respondents mentioned a need for staff to ‘offer intervention, not just medication’.
Many described not having ‘received any formal training for autistic patients’ and being ‘unaware of any specific offers coming from our training body’. They mentioned wanting training on autism ‘in both medical school and by the [College of Psychiatrists of Ireland]’ to ‘gain further knowledge and practical skills for clinical assessment’, ranging from ‘immersive training’ and ‘case-based discussions in peer groups’, to ‘regular collaborative sessions between adult and mental health and intellectual disability services’. Moreover, they wanted to ‘learn from Autistic adults/children’ (subtheme 2.2). Many commented on learning through informal means, by ‘talking to peers who know this space better than me’, while others wanted formal training ‘led by the Autistic community itself’.
Theme 3. Adapting pathways, processes and infrastructures to better meet patient needs
Respondents felt that training alone was not enough: ‘additional training is nice, but you can’t train yourself out of staff shortages’. Our respondents overwhelmingly emphasised the prevailing barriers – at a system level – to providing effective care for autistic patients:
There are significant human rights violations where people have autism and challenging behaviours. Due to lack of disability services, people get sent into mental health units initially with possible symptoms of mental illness. Once assessed and no mental illness found or illness resolved, they get left in acute psychiatric units. You can arrange meetings, case conferences, funding applications, referrals to lots of places but, at the end of the day, nothing happens. No-one cares. It’s awful… You feel complicit in a system that you can’t fix.
Respondents were clear that more funding is desperately needed for greater access to existing services (subtheme 3.1). Some bemoaned the absence of services, which meant patients and their families were ‘being shunted from Billy to Jack… nobody to take ownership’. Others reported severely ‘overstretched services’, including ‘child development services that are woefully limited’. As one respondent said, ‘services only provide assessment and no therapeutic input – I would fill out lots of very long referral letters, but I don’t believe any of those kids have/will ever be seen’. Autistic adult services were felt to be the most ‘chronically under-resourced’, with ‘accessing private pathways frequently the only option available to adults’. They wanted more funding to address ‘low staffing levels’, and exceedingly long waiting lists, which were taking years.
Respondents also wanted to see ‘a nationally consistent approach to assessment/diagnosis/support so that autistic people aren’t subject to a postcode lottery’. They spoke of ‘disconnected and fragmented services’, which were repeatedly described as being ‘confusing’, ‘for both diagnosis and treatment planning’, with ‘no signposting of where our patients and families can get support’. They wanted to address the silos and ‘barriers between Mental Health and Disability’ and called for greater ‘clarity of care pathways”’and ‘inter-agency collaboration’ (subtheme 3.2), including ‘better relationships between disability teams’ and mental health services and ‘liaisons with schools and rehabilitation units’, ‘with fully staffed multidisciplinary teams’ and ‘support from specialists in autism’. The otherwise ‘patchy’ provision and lack of ‘joined-up services’ were felt not only to be ‘a shameful waste’, but also to have damaging effects for their patients.
Respondents called for a dedicated, ‘appropriately-resourced specialist mental health service’ for autistic people (subtheme 3.3). They wanted specific ‘investment in neurodevelopmental services’ – a ‘pathway outside of generic mental health services with specialist occupational therapy, psychology, family therapy and social work inputs for individuals with autism throughout the life cycle’. They felt that such specialist services would ‘provide experienced staff’ and ‘allow for assessment and initial management and helping link people to community resources’.
Crucially, respondents wanted to see the ‘inclusion of persons and families in policy making and development of services’ (subtheme 3.4) – especially for initiatives designed to improve community-based supports and services, such as the provision of key workers with strong ties with the autistic person and their family; ‘advocates or peer support workers’; ‘community and home supports’ post-diagnosis; ‘better public health initiatives to improve awareness of autism’ in workplaces, schools and communities; and community navigators ‘to help direct clinicians/patients/families to how to get someone assessed, referral pathways, what appropriate interventions they should be getting’.
Discussion
Main findings
We investigated Irish psychiatrists’ autism knowledge, self-efficacy, and perceived barriers and facilitators to providing mental health services to support them. Autism knowledge was high irrespective of clinical role or age of respondent in our sample, similar to knowledge scores in UK psychiatrists, despite a higher proportion of non-consultant psychiatrists here compared with the UK study (Crane et al. Reference Crane, Davidson, Prosser and Pellicano2019). In contrast, prior studies have reported variable associations between knowledge of autism and clinician age or years of experience in other types of healthcare professionals, rather than psychiatrists alone and (Zerbo et al., Reference Zerbo2015; Hayat et al., Reference Hayat2019; Atun-Einy & Ben-Sasson, Reference Atun-Einy and Ben-Sasson2018). It is possible that local contextual factors may play a greater role in autism knowledge than clinician age or experience alone.
Notably, Irish psychiatrists had a similar level of personal experience of autism to that reported by the UK psychiatrists. Similar to the UK psychiatrists, self-efficacy scores varied, with lower scores in this study observed in response to questions about referral processes and service pathways. Psychiatrists with caseloads of children and adolescents, and individuals with intellectual disability, reported higher perceived self-efficacy. Unlike UK psychiatrists (Crane et al. Reference Crane, Davidson, Prosser and Pellicano2019), self-efficacy was not associated with greater knowledge, experiences and training but, instead, was linked to knowledge of referral processes and service pathways. This could be related to a lack of available services in Ireland, especially for autistic adults without intellectual disabilities (National Disability Authority, 2017). These findings were corroborated by the qualitative data, where respondents highlighted key ways to enhance existing and new services to provide the most effective support for autistic people and their families. Next, we discuss three main areas of development based on our qualitative findings.
Capacity and training among mental health professionals
Psychiatrists reported feeling ill-equipped to recognise and assess mental health difficulties in autistic children and adults and thought this applied also to members of multidisciplinary mental health teams. A need for autism-related health professional training in community mental health is reflected in Irish mental health policy (Government of Ireland, 2020). Elsewhere, various approaches have been proposed to enhance autism-related training such as developing evidence-informed autism programs in medical and healthcare curricula and professional education and training co-created with autistic people (Gilmore et al. Reference Gilmore, Longo, Krantz, Radford and Hand2022; Malik-Soni et al. Reference Malik-Soni, Shaker, Luck, Mullin, Wiley, Lewis, Fuentes and Frazier2022). In the UK, the Autism Champions Network is a co-produced initiative that partners staff (autism champions) and lived-experience experts to enhance knowledge and understanding of autism. It has been positively evaluated (Kirby & Payne, Reference Kirby and Payne2023) and is precisely the sort of training that our respondents called for. A systematic plan to evaluate and implement co-produced – or even autistic-led – training for psychiatrists and mental health professionals should address psychiatrists’ capacity and self-efficacy in managing the mental health of their autistic patients.
Improving mental health service provision for autistic children and adults
Our respondents also identified multiple system-level barriers to providing effective care for autistic patients, including under-resourcing, gaps in services provision, an absence of transparent processes and pathways and fragmentation and siloing of mental health services. They highlighted the potential harms associated with poorly integrated services and felt they contributed to poorer longer-term mental health outcomes. These perspectives echo the findings from prior studies (Anderson et al. Reference Anderson, Sosnowy, Kuo and Shattuck2018; Crane et al. Reference Crane, Davidson, Prosser and Pellicano2019; Gilmore et al. Reference Gilmore, Longo, Krantz, Radford and Hand2022). Similar barriers were also identified in a recent public consultation to inform a national autism strategy in Ireland (Government of Ireland, 2023).
Integrated stepped care approaches, as recommended in clinical guidelines and Irish mental health policy (National Institute for health Clinical Excellence 2012; Sharing the Vision, 2020), can reduce fragmentation and barriers to care, and are especially critical given high prevalences of co-occurring mental and physical health conditions for autistic people (Lai et al. Reference Lai, Kassee, Besney, Bonato, Hull, Mandy, Szatmari and Ameis2019). Psychiatrists in this study suggested broadening the remit of the National Clinical Program for ADHD in adults to encompass several neurodevelopmental conditions including autism. Since ADHD and autism frequently co-occur, a combined mental health pathway could provide more consistent care, although more research regarding the design and effectiveness of such models of mental healthcare, in collaboration with community members, is required.
New models of care will take time and resources to develop. In the interim, steps can be taken by individual clinicians and services. Tools, such as the Autism Healthcare Accommodations Tool, are evidenced to be acceptable and feasible and may reduce barriers to care, improving healthcare efficacy and satisfaction with physician communication (Nicolaidis et al. Reference Nicolaidis, Raymaker, McDonald, Kapp, Weiner, Ashkenazy, Gerrity, Kripke, Platt and Baggs2016). Perhaps even more critical, however, is the need to cultivate stronger therapeutic relationships between psychiatrists and their autistic patients. Autistic adults have repeatedly reported that their concerns and perspectives are often dismissed or not taken seriously by healthcare professionals (Camm-Crosbie et al. Reference Camm-Crosbie, Bradley, Shaw, Baron-Cohen and Cassidy2019). Taking time to listen to, and learn from, autistic patients, would be one important way to develop mutual trust and understanding and, ultimately, more effective care – precisely as our respondents suggested.
Co-creation and neurodiversity-affirmative care as the essential foundation to care
Psychiatrists strongly identified the need for co-creation and neuro-affirmative care. Neuro-affirmative care (Chapman & Botha, Reference Chapman and Botha2023) is rooted in neurodiversity principles that advocate for the acceptance of neurological differences rather than deficits or disorders that need to be fixed, as well as greater consideration of the broader social and environmental context as one key source of a person’s challenges (Kapp Reference Kapp2020; Pellicano & den Houting, Reference Pellicano and den Houting2022; Walker & Raymaker, Reference Walker and Raymaker2020). The neurodiversity perspective emerged more strongly amongst our participating psychiatrists compared with previous perspectives (Crane et al. Reference Crane, Davidson, Prosser and Pellicano2019). This finding could reflect a growing appreciation of neurodiversity in addition to a growing movement to engage patients and families in co-producing mental health services (Hawke et al. Reference Hawke, Natasha, Bastidas-Bilbao and Rodak2024). Our respondents considered co-designing services and policy with autistic people towards neuroaffirmative care as critical.
Developing a neuro-affirmative framework for health services in Ireland has recently been recommended in a draft Autism Innovation Strategy (Autism Innovation Strategy - Analysis of Initial Public Consultation Submissions, 2023). Applying the principles of the neurodiversity paradigm requires a shift in fundamental assumptions about autism, alignment with the autistic community’s priorities and involving autistic people in research and practice (Pellicano & den Houting, Reference Pellicano and den Houting2022). Applying these in mental healthcare must account for the duality of the medical and neurodiversity viewpoints (Lai et al. Reference Lai, Anagnostou, Wiznitzer, Allison and Baron-Cohen2020). To this end, Lai et al., proposed three pillars: maximising potential, minimising barriers, and optimising the person-environment fit (Lai et al. Reference Lai, Anagnostou, Wiznitzer, Allison and Baron-Cohen2020). Translating principles for neuro-affirmative care into an implementation plan within mental health services will require clear actionable steps, ensuring resources and responsibilities are clearly allocated and establishing timelines and metrics for measuring progress. The recommendations of the Autism Innovation Strategy, while positive, will require research and evaluation to match the ambition and to ensure success.
Future recommendations
This study identified critical systemic and professional challenges with providing mental health care for autistic people in Ireland from psychiatrists’ perspectives. Several steps can be taken to overcome these challenges, creating more inclusive, accessible and effective mental health services for autistic people in Ireland (Table 6).
Recommendations towards inclusive, accessible and effective mental health services for autistic people in Ireland
These recommendations echo others’ suggestions for reducing barriers to healthcare (Malik-Soni et al. Reference Malik-Soni, Shaker, Luck, Mullin, Wiley, Lewis, Fuentes and Frazier2022) and mental health care (Gilmore et al. Reference Gilmore, Longo, Krantz, Radford and Hand2022) in other contexts. The overwhelming evidence is that autistic people have elevated prevalence of co-occurring mental health conditions and increased risk of suicidal and non-suicidal self-harm over the longer term (Hirvikoski et al. Reference Hirvikoski, Mittendorfer-Rutz, Boman, Larsson, Lichtenstein and Bolte2016; Lai et al. Reference Lai, Kassee, Besney, Bonato, Hull, Mandy, Szatmari and Ameis2019). It is therefore essential that mental health services are equipped and appropriately resourced to deliver inclusive, evidenced-based care to autistic people across the lifespan.
Conclusion
This study found that Irish psychiatrists possessed adequate knowledge of autism but low levels of self-efficacy, particularly related to pathways to care. Multi-system barriers to care were identified and psychiatrists strongly endorsed investment in improved mental health services that were neuroaffirmative and accessible, as well as noting a clear need for enhanced training and professional development. As per our recommendations, future research is required to address the systemic and professional barriers to providing mental health care to autistic people.
Limitations
Our study has several limitations. The sample of 140 psychiatrists may not fully represent the perspectives of all psychiatrists in Ireland given the potential for self-selection bias among survey respondents. The cross-sectional survey design may not capture changes over time. The Knowledge of Autism Scale showed poor internal consistency, potentially affecting reliability and the self-efficacy scale is subjective and may not accurately reflect actual clinical competence. Gender and workplace variables were removed as recommended by the ethics committee limiting exploration of the influence of these factors on knowledge and self-efficacy. The perspectives of other mental health professionals, such as psychologists, occupational therapists, and speech and language therapists, were not included. Furthermore, it was not possible to tease apart the differences in views or the barriers experienced between child and youth vs adult psychiatrists which could inform understanding of specific areas of service design and delivery in more depth. All these issues should be addressed in future studies. The findings are specific to the Irish context and may not be directly applicable to other healthcare systems and training programs.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/ipm.2025.9.
Financial support
This research received no specific grant from any funding agency, commercial or not-for-profit sectors.
Competing interests
Mr. Adam Harris is the Chief Executive Officer of AsIAm, an autism charity in Ireland. The remaining co-authors declare no conflicts of interest.
Ethical standards
Ethical approval was obtained via the Research Ethics Committee of the School of Medicine, Trinity College Dublin on 26/03/2021. The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional committee on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008. The authors assert that ethical approval for publication of this original paper has been provided by their local Ethics Committee.’
Open access
