Neurodivergence, intersectionality, and eating disorders: a lived experience-led narrative review

EDs are psychiatric disorders with significant negative consequences on psychosocial wellbeing and physical health [ 360 ]. Despite this, mechanisms underlying ED risk, development, maintenance, and relapse remain poorly understood, irrespective of neurotype [ 19 , 361 ]. While manualised psycho-behavioural mental health approaches 9 are often promoted as first-line ‘evidence-based’ treatments for most mental illnesses, including EDs [see 18 , 19 ], there are several factors that should be considered when assessing the generalisability of these claims [see 362 ]. Concerns have been raised regarding high risks of publication bias [ 363 – 365 ], selective reporting [ 366 , 367 ], undisclosed conflicts of interest [ 368 – 370 ], under-reporting of adverse effects [ 371 – 373 ], lack of long-term follow-up studies [ 19 , 374 ], and lack of participatory research [ 50 , 51 , 375 , 376 ]. These non-trivial research quality concerns justify a cautious approach to generalising research results to real-life individualised circumstances. Indeed, psycho-behavioural treatments that are considered the ‘gold standard’ for EDs only lead to less than half of people with EDs reaching long-term recovery [ 18 , 19 , 377 ]. There are overarching factors involved in low reported rates of long-term recovery in the context of EDs. Diagnosis and treatment of EDs are often delayed, which negatively impacts on illness progression (e.g., increased risk of chronicity) [ 378 ]. Indeed, according to an Australian study [ 379 ], the average delay between onset of ED symptoms and treatment-seeking is 5.28 years. In the United States of America, treatment delays of 2.5 years for AN and up to 6 years for BED have also been reported [ 380 ]. There are a variety of factors underlying such delays in identifying EDs and accessing professional support. For example, individuals with EDs may experience weight stigma and/or racial biases from healthcare professionals [ 381 ], diagnostic overshadowing [ 63 , 382 ], fear of change, shame and guilt due to stigma surrounding EDs, and/or feeling as though they are ‘not sick enough’ to be deserving of care [ 20 , 383 ]. Additionally, individuals with EDs may experience difficulties accessing appropriate services due to multiple referrals [ 384 ], workforce shortages [ 385 ], and limited financial resources [ 386 , 387 ]. Even once engaged, some people with EDs self-discharge prematurely from services because of treatments and/or healthcare professionals not meeting their support needs [ 20 , 53 , 63 , 384 , 387 ] or due to experiencing dismissive attitudes or a lack of compassion, understanding, and validation from clinicians 10 [ 20 , 53 , 63 , 388 – 390 ]. With emerging evidence showing that NDI have poorer ED treatment experiences and outcomes compared to neurotypical individuals [ 42 , 47 , 58 , 142 , 391 – 394 ], it is imperative to consider why traditional ED interventions may be less beneficial, and possibly harmful, for many NDI who receive them. Insight into both intervention efficacy and risk of harm is essential to the practice of evidence-based medicine; as noted by Junqueira et al. [ 395 , p. 216], “treatment decisions rely on accurate knowledge of both efficacy and harms.” However, adverse effects related to psycho-behavioural treatments for mental ill-health frequently lack systematic assessment and transparent reporting in research, subsequently skewing evidence regarding risk–benefit ratio [ 371 , 372 , 396 – 398 ]. For example, in a recent systematic meta-review of meta-analyses of ED treatments, adverse events were only acknowledged in the context of pharmacology [ 399 ]. A 2023 report looking at early intervention programs for EDs also found that none of the 14 clinical trials identified evaluated harms [ 400 ]. Furthermore, Linardon et al. [ 401 ] conducted a systematic review and meta-analysis to assess the efficacy of CBT for treating EDs, where they deemed CBT to be efficacious despite omitting to discuss measures of short- and long-term harms. Iatrogenesis, that is, harm resulting from healthcare practices, may not manifest immediately and be delayed [ 371 ], in which case it would only be captured in long-term follow-up studies. Late-onset iatrogenic harm is therefore missed by the lack of long-term follow-up studies in ED research [see 19 ]. Subsequently, lack of reported iatrogenic harm in the literature is mistakenly interpreted as evidence of no harm [ 373 ], leading mental health practitioners under-resourced to recognise and address iatrogenesis from psycho-behavioural treatments [see 372 ], particularly in the context of neurodivergence [ 20 , 47 , 62 , 402 – 404 ]. There is an increasing number of community resources where NDI with lived experience of EDs report having experienced iatrogenic harm from ED treatments, including psychological trauma [ 54 – 57 , 405 ]. Unfortunately, such narratives are seldom reflected in published mental health clinical trials and meta-analyses. Biases undervaluing qualitative, phenomenological, and lived experience-led research have been suggested to contribute to such narratives not being adequately discussed in mental health research [ 406 , 407 ]. However, there is growing recognition that psycho-behavioural treatments can be harmful to NDI [ 62 , 402 , 408 ], including in the context of EDs [ 20 , 53 , 58 ]. The indiscriminate use of a narrow range of manualised therapies, assessment tools, and public health campaigns for different demographic groups 11 is problematic, as each group experiences unique underlying factors in the development of EDs and may require different, sometimes conflicting or opposing, forms of support and formulations to recover [ 10 , 20 , 53 , 409 – 411 ]. As outlined in section six, NDI, as well as those with other minority and/or marginalised identities, have been found to have divergent feeding, eating, and ED experiences. However, hegemonic conceptualisations of feeding, eating, and EDs do not incorporate such diversity of lived experiences, a limitation that impacts all stages of ED care [ 10 , 20 , 53 , 387 ]. Evidence-based medicine requires that clinicians consider research evidence, clinical expertise, and patient preferences when deciding on a course of action [ 412 ]. ED clinical guidelines identify a narrow range of 'evidence-based' ED interventions, many of which are manualised, whilst also acknowledging the important role of person-centred approaches to ED care [ 413 , 414 ]. Importantly, the legacy of hegemonic conceptualisation of EDs places clinicians who are attempting to translate research evidence into clinical practice in a double bind: on the one hand, clinicians must practice in accordance with ED clinical guidelines which identify a narrow range of suitable treatment approaches, typically manualised, that they can choose from, whilst simultaneously being asked to practice person-centred care that, if followed, is not likely to fit neatly into the prescribed formulation and treatment plan laid out in the approved manualised interventions. Personalised ED care would be a more realistic goal if our understanding of EDs and recovery was informed by evidence derived from diverse and representative samples who had been afforded epistemic agency. Traditionally, however, people experiencing mental ill-health have been excluded from leadership roles in mental health research and treatment design due to “presumed irrelevance or lack of expertise” [ 415 , p. 1064]. People with EDs, too, have been denied epistemic agency as deemed ‘untrustworthy’ narrators of their own experiences [ 416 , 417 ]. Similar denial of epistemic authority has also been perpetrated against autistic people [ 418 – 420 ]. It is unsurprising, then, that experiences of epistemic injustice have been reported by autistic people with EDs [see 10 , 68 ]. As Elwyn [ 20 , p. 5] puts it, “[c]linicians’ conceptualisations of ED in treatment often directly erased and invalidated the meaning of my experiences, and these experiences reinforced my AN.” We argue that this historical silencing of lived experience voices has contributed to the dominance of a narrow range of 'evidence-based' ED treatments despite their lacklustre success rates [see 18 , 19 , 377 ], while limiting clinicians' ability to deliver affirming and personalised care. Ultimately, we propose that the development of affirming, innovative, and inclusive conceptualisations of, and interventions for, EDs has likely been constrained by consistently failing to recognise, acquire, and utilise the wisdom contained in first-person phenomenological accounts [see 421 ]. In addition to aiming for weight restoration, as indicated, nutritional rehabilitation interventions typically aim to ‘normalise’ eating by encouraging an intake of food and fluids that is nutritionally adequate, spread out in regular increments across the day, is comprised of a variety of foods and fluids, and that the individual is able to eat socially and be spontaneous or flexible with their eating and food-related cognitions. These nutrition rehabilitation targets are captured in the RAVES model (an acronym representing regular eating, adequacy, variety, eating socially and spontaneity), a clinical tool that is commonly used by dietitians to guide nutritional rehabilitation interventions [ 422 ]. Whilst the intervention targets identified in this clinical tool appear logical when framing eating through a neuronormative lens, lived experience-informed conceptualisations of adaptive neurodivergent feeding and eating experiences highlight areas of difference, such as preference for sameness [ 209 ], sensory-related food variety constraints [ 9 , 10 , 61 , 154 ], and preference to eat alone [ 188 ], that are important for clinicians and services to consider when working towards neurodiversity-affirming, person centred ED care. Although there are certainly going to be NDI for whom some or all of these intervention targets are acceptable and achievable, in order to avoid indiscriminately applying the tool and risking unintentionally invalidating authentic and adaptive neurodivergent feeding and eating support needs and preferences, it is essential that clinicians are aware of how supporting adaptive neurodivergent feeding and eating experiences may require an adapted or alternative approach [ 154 ]. As has been noted previously, NDI frequently report experiencing divergent sensory processing patterns that influence their eating [ 9 , 10 , 42 , 58 , 73 , 154 ]. With some evidence suggesting that sensory habituation can be attenuated amongst autistic individuals [ 210 , 211 ], it is prudent to consider that food-related exposure-based techniques, which are often used in tandem with psycho-behavioural ED treatments such as CBT [ 423 ], carry a risk of causing sensory overwhelm, distress, or even sensory trauma for NDI when they do not take individual sensory processing differences into account and/or do not respect an individual’s agency and right to self-determination during exposures [see 212 ]. Traditional ED interventions for adults (e.g., CBT) are largely consistent in targeting factors presumed to maintain the ED such as over-evaluation of weight and/or shape, cognitive features such as perfectionism and cognitive inflexibility/intolerance of uncertainty, emotional and interpersonal difficulties, and self-esteem [see 424 , 425 ]. We can thus begin to hypothesise why traditional ED treatments may be less beneficial and acceptable to NDI, in comparison to their neurotypical peers, when we consider how well the conceptualisation of EDs that informs the above traditional frameworks fits with what we now know about NDI ED experiences. Taking autism and restrictive eating as an example, Brede et al.’s [ 61 ] model of autism-specific mechanisms underlying restrictive EDs points to specific autism-related differences, such as need for predictability, intense interests, and emotional and sensory processing differences. These mechanisms are proposed to have both direct and indirect impacts on restrictive eating behaviours in autistic individuals [ 61 ]. This conceptualisation sits counter to that of traditional, psycho-behavioural, treatments, which assume that many of these difficulties are the result, at least in part, of starvation and/or ED psychopathology, and therefore focus on “teaching the patient that their emotional and cognitive reactions to food and eating are inaccurate” [ 425 ]. Unfortunately, this neuronormative understanding of EDs fails to recognise that, for autistic individuals, these cognitive and emotional processing differences frequently are part of their autistic-based lived experiences, and therefore not amenable to change through weight restoration or cognitive-behavioural interventions [ 154 ]. When traditional interventions fail to understand, and work with [ 10 ], these differences, they risk invalidating patients’ authentic neurodivergent experiences, may encourage masking 12 in treatment, and may result in iatrogenesis [see 20 , 47 , 54 – 57 ]. NDA practice promotes the notion that interventions should not target authentic neurodivergent features for eradication or ‘normalisation’ [ 426 – 429 ]; rather, neurodivergent features and associated support needs should be understood and accommodated, with the intervention’s focus redirected to supporting adaptive and self-advocacy skills [ 105 , 429 – 431 ]. This will be discussed in further detail in Sect. " What does neurodiversity-affirming eating disorder care entail? ". According to Curnow et al.’s [ 432 ] umbrella review investigating the overall suitability of currently available mental health therapies in the context of autism, no existing mental health framework can claim to be NDA; a conclusion that has also been highlighted in the ED context [ 10 , 20 , 47 , 58 ]. There are two main factors playing into this: namely, the lack of representation of NDI in mental health research and the lack of non-tokenistic participatory research [see 53 , 63 , 421 , 433 ]. The neurodivergent community, and other marginalised communities, 13 are not adequately represented in ED research [ 52 , 53 , 409 , 421 , 433 , 434 ]. The lack of neurodivergent representation is not limited to EDs; for example, Jubenville-Wood et al. [ 375 , p. 2] noted that “limited research exists that examines the experiences of autistic individuals in psychotherapy.” However, as highlighted by Le Cunff et al. [ 435 , p. 2], “inclusive research practice characterized by the involvement of NDI in meaningful roles has been described as a ‘requirement of excellence’ in neurodiversity research.” The neglect of intersectionality in ED research further extends to co-occurring mental health conditions, despite a majority of those with EDs also having another psychiatric condition [ 436 ]. This is particularly relevant in the context of neurodivergence as NDI experience mental ill-health (e.g., depression, anxiety, post-traumatic stress disorder) at significantly higher rates compared to neurotypical individuals [ 437 – 441 ]. Participatory frameworks are rarely utilised in ED research [ 50 – 53 , 63 , 421 , 433 ] or in mental health research more broadly [ 406 , 407 , 442 – 445 ]. Inclusive participatory research is “research being carried out ‘with’ or ‘by’ members of the public rather than ‘to’, ‘about’ or ‘for’ them” [ 446 , p. 1] and has the potential to improve overall mental health outcomes for NDI [ 376 , 444 , 447 ], including in relation to EDs [ 52 , 63 , 67 , 421 , 433 ]. Indeed, participatory research is suggested to facilitate the translation of findings into real-world contexts and ensure that the work yields relevant and meaningful benefits for the community [ 446 – 449 ]. Hence, a greater focus on participatory research that takes an intersectional lens is indicated to better understand the connections between neurodivergence and EDs and improve ED prevention efforts and treatment experiences for NDI [ 10 , 52 , 53 , 67 ]. A new United Kingdom-based research project, the Eating Disorder and Autism Collaborative (EDAC), aims to support co-produced and lived experience-led research collaborations between the autism and ED research fields [ 433 ]. EDAC seeks to embed an NDA culture within ED research [ 421 , 433 ], and to this end, engaged in online workshops and discussions to inform the development of a framework for engaging in ethical participatory ED research with autistic people [ 421 ]. The primary theme identified involved the importance of sharing power and avoiding traditionally held power imbalances and inequalities, such as by involving community members from research inception to reduce the influence of neuronormative biases [ 421 ]. Other key aspects of ethical co-production included ensuring accessibility (e.g., using clear, consistent, and transparent language), accommodating communication preferences and needs (e.g., speaking, writing), ensuring the research environment is a safe place for autistic collaborators to engage (e.g., avoid using dehumanising language or ableist terminology [see 1 , 2 , 4 , 6 ]), and promoting capacity-building within the community 14 as well as autistic researchers with ED lived experience. Furthermore, the Autism Cooperative Research Centre (Autism CRC), based in Australia, produced a guideline for conducting general participatory research with autistic people [ 450 ]. Autism CRC has also implemented a program that teaches research skills to members of the autistic community and skills for engaging in inclusive participatory research to researchers to promote meaningful collaborations [ 451 ]. While EDAC’s [ 421 ] and Autism CRC’s [ 450 ] participatory research guidelines have been developed with the autistic community in mind, there is potential to adapt and expand their scope to accommodate and facilitate the inclusion of a wider range of neurotypes and lived experience communities. Current psycho-behavioural frameworks for treating EDs are often broadly promoted as effective and ‘evidence-based’, despite meta-analytic evidence consistently demonstrating only less than half of people who access these treatments experience long-term recovery, irrespective of neurotype. There are several factors suggested to contribute to overall low recovery rates, such as delayed diagnosis and care as well as workforce shortages. However, evidence shows that autistic people experience poorer treatment outcomes compared to neurotypical people; unfortunately, there is a paucity of empirical evidence for other types of neurodivergence. Issues relating to lack of long-term follow-up data, limited assessment and reporting of iatrogenesis, undisclosed conflicts of interest, and protocol and publication biases, have been identified as factors to consider when interpreting efficacy claims. More specifically, participatory research methods are seldom used in ED research and NDI are rarely represented or taken into consideration in treatments’ design and implementation, as are other marginalised communities, thus significantly undermining the validity of ‘evidence-based’ mental healthcare for a more diverse range of people. Furthermore, ED research rarely takes a holistic lens that acknowledges intersectional factors, as outlined in section six, needing to be considered in treatment formulations. Such epistemic injustice and power imbalances contribute to oversimplifications in manualised treatment frameworks that do not account for the fact that different people and communities have different, and sometimes conflicting, support needs. We also acknowledge that more research is needed to confidently determine whether some components of traditional treatments may be helpful for some NDI.

The concept of NDA care is grounded in the principles of the neurodiversity paradigm, which views neurodiversity as a natural and valuable form of human diversity, rejecting the idea that there is a singular ‘normal’ personality and neurocognitive profile that all humans should universally aspire to and emulate [ 23 – 26 , 38 , 40 , 41 ]. An NDA mental healthcare framework, therefore, is a holistic and person-centred framework rooted in the principles of phenomenology, epistemic justice, and cultural humility [ 105 , 426 , 452 – 454 ]. NDA ED care requires that ED clinicians engage in ongoing self-reflection to challenge biases and neuronormative assumptions, social norms, 15 beliefs, and goals that further perpetuate the pathologisation and systemic oppression of NDI [see 20 , 63 , 154 ]. Crucially, NDA ED care necessitates clinicians be able to differentiate neurodivergent traits from ED symptoms, understand the underlying reasons and adaptive mechanisms of neurodivergent traits, and validate and accommodate the individualised support needs of NDI, such as with regards to sensory processing, communication, and executive functioning, rather than attempt to ‘fix’ these at the expense of overall wellbeing and self-determination [ 154 , 455 , 456 ]. To the authors’ knowledge, there is no validated psychometric instrument specifically designed for the assessment of EDs in NDI. However, there are assessment tools have been developed to support the identification of body image disturbances, as well as feeding and eating differences and difficulties amongst autistic individuals: Swedish Eating Assessment for Autism Spectrum Disorders [ 457 ], Body Appreciation Scale 2 for Autistic Adults [ 122 ], Autism Eating Questionnaire [ 458 ], and Brief Autism Mealtime Inventory Assessment [ 459 ]. Suggestions have also been made to tailor ED treatments for NDI [ 10 , 460 – 463 ]. However, except for the Body Appreciation Scale 2 for Autistic Adults [ 122 ], none of the adaptations to either identification or treatment discussed in this section have been explicitly identified by their authors as having been developed through lived experience-led research. It is also important to note that while care adaptations discussed below have been developed with the intention of improving treatment outcomes for NDI, this does not necessarily guarantee that they will be experienced as NDA by NDI or helpful in practice [see 154 , 463 ]. Empirical evidence led by NDI with ED lived experience is required to provide insight on whether these treatment adaptations are experienced as helpful. Overall, treatment adaptations that have been suggested as offering potential benefit for autistic individuals include recognising that autism and AN are interlinked; upskilling clinicians in their ability to disentangle behaviours that are autism-related from those that are ED-related and in their understanding of how being autistic influences cognition; offering person-centred supports that may be distinct from what is typically offered to neurotypical individuals (e.g., support with daily tasks, attention, sensory processing, routines, and social connectedness concerns); treatment being more flexible and individualised [ 10 , 68 ]; and tailoring ED services (including inpatient settings) to better meet the needs of NDI [ 10 , 47 , 58 ]. An ED inpatient care pathway tailored for autistic people, the PEACE Pathway, has been developed and implemented in the United Kingdom [ 462 ]. The PEACE Pathway outlines the need for individualised sensory processing accommodations and includes training for clinicians to assist in making autism-specific adjustments to ED care and psychoeducation [ 464 ]. A preliminary evaluation of the PEACE pathway found that it reduced treatment duration and resulted in admission cost savings [ 464 ]. Two adjunctive manualised inpatient interventions have been implemented and evaluated comparing their impact in individuals with and without autistic features; Cognitive Remediation Therapy (CRT) and Cognitive Remediation Emotional Skills Training (CREST) [ 465 – 469 ]. CRT targets cognitive flexibility and central coherence and has been adapted for AN in both individual and group formats [ 465 ]. CREST, also available in both individual and group formats, targets social-emotional challenges in AN, supporting individuals to identify and describe their emotions and subsequently communicate their needs in a safe way [ 469 ]. The impact of autistic features on the effectiveness of CRT has been examined in both group [ 465 ] and individual formats [ 466 ]. Whilst brief group CRT has demonstrated effectiveness in individuals with low autism features, it does not appear to support improvements in targeted outcome measures for individuals with high autism features [ 465 ]. Dandil et al. [ 466 ] found that individual CRT may support improvements in cognitive flexibility, but not central coherence, regardless of the presence of autism features, in a small naturalistic sample of adults receiving inpatient treatment for AN. Adamson et al. [ 467 ] compared whether the presence of self-reported autistic features in people receiving CREST, either in a group or individual format as an adjunctive intervention to their inpatient treatment for AN, impacted self-reported alexithymia, social anhedonia, and motivation scores. In this study, group format CREST was found to enhance motivation scores, but had no impact on alexithymia or social anhedonia [ 467 ]. In individual format, CREST appeared to support an improvement in alexithymia and confidence but there were no significant impacts on social anhedonia or motivation [ 467 ]. Across both formats, the presence of self-reported autism features did not appear to differentially influence intervention outcomes [ 467 ]. However, across studies, results are heterogeneous, with Saure et al.’s [ 468 ] review concluding that individuals with AN with high autistic traits appear to benefit less from CRT (as well as from CREST and CBT) compared to those with low autistic traits. Whilst it does appear that there is some evidence to support feasibility of individual CRT in individuals with more autistic features who have AN, there is a clear need for more research to confirm the feasibility, efficacy, and acceptability of CRT in this specific cohort, and in NDI with EDs more widely. Drawing on available research and clinical experience of supporting autistic children and young people with AN, Loomes and Bryant-Waugh [ 461 ] identified a range of autism-specific modifications for family-based interventions for AN. The suggested adaptations target four domains: sensory processing, cognitive and behavioural factors, difficulties with social communication and relationships, and emotional processing [ 461 ]. The authors recognised that the suggested adaptations are conceptual and thus require empirical assessment before their acceptability and efficacy can be determined [ 461 ]. A single case report detailed adaptations to CBT to meet the needs of an individual with ID experiencing difficulties with binge eating [ 470 ]. Adaptations included a focus on regularity and predictability of appointments, the use of images to aid with understanding, adapted use of language, and sensory considerations [ 470 ]. The authors stated that the participant’s self-regulation associated with loss of control eating improved as a result of the intervention [ 470 ]. An NDA approach to mental healthcare is a human rights-based approach rooted in the social model of disability and existential-humanistic psychology [ 426 , 427 , 452 , 471 , 472 ]. As such, NDA mental healthcare is inherently gender-affirming and trauma-informed [ 426 , 452 , 456 ]. This requires the rejection of compliance and exposure-based behavioural approaches (e.g., operant conditioning, contingency management) that are imposed rather than self-determined and client-led, as these have been associated with iatrogenesis, including learned helplessness and psychological trauma [ 290 , 452 , 473 – 476 ]. Instead, the priority must be on supporting self-determination, self-advocacy, and self-efficacy, and incorporating a person-centred, compassionate, and strengths-based approach to care that is informed by lived experience expertise [ 418 , 426 , 477 ]. It is imperative that all NDA mental health interventions aim to support the development of a positive and authentic neurodivergent identity [ 105 ], as this has been associated with benefits to overall psychosocial wellbeing and increased self-compassion [ 427 , 452 , 455 , 478 ]. The benefits of a positive neurodivergent identity have also been noted in relation to supporting ED recovery specifically [ 10 , 20 , 53 , 61 , 63 ]. Such an approach echoes the recent call to action released in a joint statement by the United Nations and the World Health Organization urging mental health practitioners and researchers to focus on the social determinants of mental health, epistemic justice, harm reduction, and human rights-based mental healthcare [ 479 ]. Founding members of Autistic Doctors International have created SPACE, an autistic-led framework for autism-specific accommodations aimed at improving accessibility of healthcare services for autistic individuals [ 480 ]. The SPACE framework identifies key reasonable adjustments across core support needs for autistic people, such as sensory, predictability, acceptance, communication, and empathy, as well as across three main domains of experience: physical space, processing space, and emotional space [ 480 ]. As noted in section six, NDI have unique sensory, communication, cognitive, and emotional needs that have been identified as important to understand, validate, and accommodate as part of NDA ED care, and as such, SPACE has potential utility as a tool to support more inclusive service delivery. For example, some autistic people may be non-speaking, while others may be partially speaking (e.g., situational mutism), and allowing communication in ways that feel comfortable is likely to improve trust between patient and clinician [ 154 , 480 ]. Additionally, adapting the consulting room to meet the sensory needs of autistic patients may also assist in building trust, self-regulation, and safety (e.g., noise, temperature, smells, lights) [ 480 ]. Research exploring general mental healthcare has highlighted empirical evidence supporting the benefits of sensory-based approaches in the general population, where the physical environment is tailored to the sensory processing characteristics of an individual [ 481 – 485 ]. Sensory-based approaches can include strategies such as adjusting the room’s lighting and/or colours, providing access to weighted blankets, fidget tools, rocking chairs, and/or noise-cancelling headsets to reduce sensory stimulation [ 485 ]. Sensory-based mental health approaches have been shown to create a sense of safety and control, increase self-regulation, reduce distress and anxiety, improve self-perception, and help with stabilising (or ‘grounding’) acute emotional arousal [ 483 – 485 ]. There is also emerging evidence suggesting that sensory-based approaches in inpatient mental health facilities may contribute to improvements in autistic individuals’ wellbeing and self-regulation [ 486 , 487 ], giving further weight to the argument in support of ED services attending to the unique sensory needs of NDI. NDA care involves respecting and validating different personality and neurocognitive profiles instead of de facto pathologising divergence and trying to ‘fix’ differences rather than celebrating diversity and accommodating specific support needs. Promoting self-determination, self-advocacy, and a positive, authentic, sense of self is critical to NDA care. In the context of EDs, NDA care requires that clinicians engage in epistemic justice and remain curious towards the different ways NDI engage with eating as well as self-perception and expression. Indeed, distinguishing between ED symptoms and adaptive neurodivergent traits is key to NDA ED care. Additionally, while further empirical research is required, the SPACE framework and sensory-based approaches may be useful in the context of ED care for NDI.

Autistic people and those with attention deficit hyperactivity disorder have a high risk of developing an eating disorder. Other forms of neurodivergence, such as giftedness, intellectual disability, obsessive–compulsive disorder, psychosis, and Tourette’s syndrome, have also been associated with disordered eating. Compared to the general eating disorder population, neurodivergent people experience eating disorder symptoms over a longer period of time and commonly experience poorer eating disorder-related treatment outcomes. Yet, there is limited research which is led by neurodivergent people with lived or living experience of eating disorders. However, it is essential to incorporate lived experience expertise in research and treatment design and implementation as this could improve understanding of the many complex and varied underlying eating disorder risk factors which may otherwise guide necessary treatment adaptation(s). Based on the perspectives of lived experience researchers and/or clinicians, this paper discusses the various mechanisms underlying disordered eating and/or eating disorder risk in the context of neurodivergence to inform future developments in eating disorder-related research and clinical practice.

The associations and psychopathological mechanisms underlying eating disorders in the context of neurodivergence remain understudied and poorly understood, as illustrated by limited lived experience-led, transdiagnostic, and intersectional research in this area. With this field of inquiry still in its infancy, this paper does not seek to provide a comprehensive, systematic, review of the literature. Instead, our aim is to provide a narrative review serving as a conceptual, lived experience-led, foundation for future empirical work. In this view, this review builds on the work of Adams et al. [ 9 ] investigating the underlying mechanisms linking autism and eating disorders and Field et al.'s [ 10 ] exploration of autism-specific eating disorder treatment adaptations. However, given the documented overlapping and transdiagnostic nature of neurodivergence [ 11 – 13 ], our work diverges from Adams et al. [ 9 ] and Field et al. [ 10 ] as we take a wider lens to investigate neurodivergence using a neurodiversity-affirming theoretical framework. Indeed, our paper presents a holistic overview that considers intersectionality by succinctly capturing the complexity of interconnected socio-cultural and biological factors regarding neurodivergence and eating disorder risk from the perspectives of the authors’ broad professional and personal lived experience.

A narrative review methodology was selected to provide a broad descriptive framework for this topic, suitable to the overall limited status of literature published to date. In addition, the review is intentionally informed by the lived experiences of the authorship team, all of whom are ED researchers and/or clinicians, providing insights with the aim of promoting inclusive research and clinical frameworks. For section five, two databases (PubMed and Google Scholar) were searched (11/03/2024 and 07/06/2024) for prevalence data specific to the co-occurrence discussed (e.g., autism/EDs, ADHD/EDs). Search terms involved a type of neurodivergence (e.g., autism, ADHD, Tourette’s syndrome, giftedness) alongside ‘eating disorder’ or ‘feeding disorder.’ The most recent systematic reviews and meta-analyses were selected: here, if no systematic reviews and meta-analyses were identified, the most recent studies were discussed. However, there were instances where no recent studies assessing prevalence were found, in which case it was highlighted that research evidence is limited. 1 Section six provides an analysis of factors associated with ED risk and maintenance in relation to neurodivergence. These factors were identified from an examination of both community resources [ 54 – 57 ] and ED research (especially qualitative) investigating NDI’s lived experiences of eating, body image, and EDs [ 10 , 20 , 47 , 52 , 53 , 58 – 73 ]. Co-authors contributed to themes that matched areas of their lived experience, interest, and/or academic expertise. Sensory processing and executive functioning differences are discussed first due to their core overarching nature across neurodivergence. Individual characteristics, such as gender identity, sexual orientation, and giftedness follow. Intersectional factors like systemic discrimination and chronic illnesses are then investigated. Section seven consolidates the evidence and positionality to identify current gaps in research and practice while section eight provides suggestions for NDA ED care. Furthermore, all co-authors contributed to revision and editing of this narrative review. It is critical to acknowledge the role that being neurodivergent researchers with lived experience of a range of feeding and eating differences and disorders, ED treatments, and a variety of life experiences that include trauma, iatrogenic harm, and interpersonal victimisation, had on our interpretations of the literature and subsequent discussion [see 74 ]. Additionally, all co-authors identify with a research orientation aligned with the principles of the neurodiversity paradigm [see 23 , 24 , 38 , 40 ], which encourages the adoption of a critical lens to publications that pathologise neurodivergence, draw on neuronormative frames of reference when interpreting results, and/or perpetuate epistemic injustice against NDI.

Gillberg [ 75 ] was the first to suggest an association between autism and EDs. Throughout the 1990s and early 2000s, sporadic studies have discussed the co-occurrence of EDs and neurodivergence; for example, Speranza et al. [ 76 ] found an overlap between obsessive–compulsive traits and anorexia nervosa (AN), Gravestock [ 77 ] highlighted an association between ID and binge eating disorder (BED), and Surman et al. [ 78 ] posited a link between ADHD and bulimia nervosa (BN). There has been an acceleration of research on this topic in the past decade, consistently reporting high co-occurrence rates between autism and ADHD with various ED diagnoses 2 [ 42 , 43 ]. Overall, there is a high degree of overlap and co-occurrence between many forms of neurodivergence 3 [ 38 , 79 , 80 ]; however, transdiagnostic research regarding forms of neurodivergence other than autism and ADHD in relation to EDs is limited. As such, the evidence below is presented as a high-level summary within the most investigated forms of neurodivergence in relation to EDs. 4 Despite the increased attention toward EDs in autistic and ADHD populations, there are limitations to estimating co-occurring prevalence rates. Indeed, there is accumulating evidence showing NDI assigned female at birth (AFAB) being erroneously diagnosed with borderline personality disorder [ 81 – 90 ]. One leading argument for such under- and/or misdiagnosis is gender-biased diagnostic instruments and practice [ 91 – 95 ]. Research also suggests that autism and ADHD are underdiagnosed in racially marginalised individuals, due, in part, to Western ethnocentrism embedded in diagnostic measures [ 96 , 97 ]. Moreover, research shows that autism and ADHD are more likely to be misdiagnosed as conduct disorder, antisocial personality disorder, and/or oppositional defiant disorder in racially marginalised communities [ 98 – 104 ]. In addition, many individuals experience barriers to accessing an autism and/or ADHD diagnosis (e.g., cost, wait times, fear of not being believed) [ 105 ], and significant delays averaging over two years in securing an assessment have been reported [ 87 ], including in Australia [ 106 ] and the United States of America [ 107 ]. Misdiagnoses and delayed diagnoses are particularly worrying considering that undiagnosed autism and/or ADHD is associated with many risks and negative outcomes related to psychosocial well-being and mental health [ 105 , 108 ], such as suicide, self-harm [ 109 ], and EDs [ 110 ]. Therefore, there have been calls for greater acceptance of self-diagnosis [ 105 , 111 , 112 ]. We also acknowledge that there is a lack of consensus on how autism prevalence should be assessed in research contexts (e.g., self-report psychometric instruments versus diagnostic tools) due to predictive validity variations. Some of the research assessing autism prevalence in EDs uses self-report psychometric instruments (e.g., Autism Quotient) [ 113 ] while other studies use diagnostic tools such as semi-structured or structured interviewing techniques (e.g., Autism Diagnostic Observation Schedule, ADOS) [ 114 ]. There have been some concerns raised by the use of self-report instruments, as opposed to diagnostic tools, for assessing autism prevalence in ED populations (e.g., risk of overestimation). However, it is worth noting that research assessing the prevalence of autism in EDs using the second version of the ADOS (sensitivity: 83–91% and specificity: 86–94% over all modules) [ 115 ] consistently shows overrepresentation of autism in AN (20–30%) [ 116 – 118 ]. Additionally, with emerging evidence suggesting that the second version of the ADOS fails to fully account for the autistic features and lived experiences of AFABs, leading to autism under-diagnosis in this demographic [ 119 – 121 ], it is possible that it may also underestimate autism prevalence when applied in ED populations. Relatedly, psychometric instruments commonly used for assessing eating psychopathology and body image-related constructs have not been tailored to, nor validated for, NDI [see 71 , 122 ]. Therefore, it is possible that ED measures predominantly used today, such as the Eating Disorder Examination Questionnaire or the Eating Attitudes Test [ 123 ], do not accurately capture EDs and body image disturbances as experienced by NDI [see 7 , 122 ]. There is also a paucity of research investigating how the presence of co-occurring neurodivergent neurotypes influences ED presentations, despite evidenced overlaps between many forms of neurodivergence [ 11 – 13 , 80 ]. As such, there is a significant knowledge gap regarding ED manifestations and course of illness in multiply neurodivergent populations. 5 Systematic and meta-analytic reviews have consistently highlighted a strong link between autism and AN: Huke et al. [ 124 ] found the prevalence of autism among those with AN varying between 8 and 37%, while Boltri and Sapuppo [ 117 ] identified a prevalence range between 8.8 and 24.5%. Additionally, between 8.2 and 54.8% of children with ARFID are autistic [ 42 ], while 21% of autistic people are deemed at risk of developing ARFID based on a genome-wide association study [ 125 ]. Furthermore, undiagnosed ARFID in autistic children may represent a risk factor for malnutrition and death if left undiagnosed and untreated [ 126 ], highlighting the urgent need for systematic ED screening, including ARFID, in autistic children. Between 14 and 36% of autistic children also show pica behaviours [ 127 , 128 ]. Although research has so far focused mostly on dietary restriction in relation to autism, there is emerging evidence linking autistic traits with all forms of disordered eating, including bingeing behaviours [ 129 , 130 ]. There is a significant association between ADHD and EDs. A recent review, which included five cross-sectional studies, found that 31.37% of children and adolescent patients with an ADHD diagnosis presented with BED, 19.23% with BN, and 9.38% with AN [ 43 ]. Additionally, a meta-analysis found that the prevalence of ADHD in adults with EDs was 3–16.2% for AN, 9–34.9% for BN, and 19.8% for BED [ 131 ]. Here, ADHD traits were also found to predict binge eating and bulimic symptoms' severity, even after controlling for anxiety and depression [ 131 ]. Moreover, in a study involving 1,165 adults diagnosed with an ED, Svedlund et al. [ 48 ] found that 37% of individuals with AN and 35% of those with BN scored above the cut-off on the Adult ADHD Self-Report Scale [ 132 ]. While prevalence has not been established, research suggests that children with ADHD present an increased risk of loss of control of eating compared to children without ADHD [ 43 , 133 ]. A high incidence of ADHD has also been observed in adolescents [ 134 ] and adults [ 135 ] seeking weight management surgery. There is limited research investigating ID in relation to EDs, in part driven by the lack of adapted ED assessment tools in the context of ID [see 136 ]. It has been estimated that 27% of people with ID had an ED, with BED being the most prevalent [ 137 ]. However, reported ED prevalence rates tend to be lower for people with ID living in the community (1–19%) compared to those who reside in care institutions (3–42%) [ 77 ]. Eating and feeding problems are common amongst children and adults with an ID. In a study of children with ID, Gal et al. [ 44 ] found that 97% of participants had at least one eating or feeding problem (i.e., food refusal, food selectivity, dysphagia, rumination, binge eating). Similarly, for adults with ID, a systematic review highlighted a high incidence of swallowing difficulties (e.g., aspiration, dysphagia) contributing to malnutrition [ 138 ]. There is limited data regarding the co-occurrence of Tourette’s syndrome and EDs. However, one study estimated that the lifetime prevalence for EDs was 2% for those with Tourette’s, with females (7%) more at risk than males (0.3%) [ 139 ]. Furthermore, studies have highlighted a high incidence of disordered eating and feeding challenges for children with Tourette’s syndrome [ 45 , 46 ]. For OCD, meta-analytic prevalence estimates report co-occurrence of OCD in EDs being 13.9–18%, with greater frequency in AN [ 140 , 141 ]. In comparison to neurotypical individuals experiencing EDs, individuals with OCD, or those with elevated OCD traits, are at greater risk of more severe ED symptomatology and poorer treatment outcomes [ 142 ]. Research on the co-occurrence rates of psychotic spectrum disorders and EDs is also limited; however, higher levels of co-occurrence between these two diagnostic groups have been indicated compared to the general population [ 143 ]. Indeed, a systematic review showed higher rates of disordered eating compared to controls in individuals with schizophrenia, with disordered eating behaviours observed in 10–41.5% of patients [ 144 ]. Moreover, transient psychosis occurs in EDs secondary to malnutrition, electrolyte imbalance, metabolic disturbance, and re-nutrition [ 145 , 146 ], and the incidence of transient psychosis among ED patients has been estimated to be 10–15% [ 146 ]. Lastly, although some studies have suggested a link between giftedness and EDs, with gifted individuals thought to be at a high risk of developing EDs [ 147 – 153 ], there is no prevalence data currently available. While clear associations between autism, ADHD, and EDs have been highlighted, more investigations are needed with regards to other forms of neurodivergence as well as research into EDs in the context of co-occurring neurodivergent neurotypes. Furthermore, estimating prevalence remains challenging due to issues such as NDI being misdiagnosed with personality disorders and lack of ED-focused psychometric instruments tailored to, and validated for, NDI.

While our work did not entail a comprehensive systematic review of the literature due to the nascent nature of this specific area of inquiry, this narrative review allows for scoping of the literature available to date whilst providing a lived experience-led framework for discussion and an initial conceptual map for future empirical work. Additionally, this review can only draw from the strength of the evidence that is available. Although an extensive narrative review, research is still limited, and this can only be improved with funding and support. Moreover, most of the literature to date focuses on either autism or ADHD in relation to EDs. Hence, there is limited data relevant to multiply neurodivergence to draw upon. This paper was written exclusively by NDI with lived or living experience of EDs, which may influence reflexivity. However, the importance of lived experience-led mental health research has been consistently highlighted, particularly with regards to translation and implementation.

While evidence demonstrates overall lifetime prevalence of eating disorders (EDs) ranging between 0.74 and 2.2% for males and 2.58–8.4% for females [ 14 ], there are growing concerns over the increasing prevalence of EDs, particularly for children and adolescents [ 15 ]. Indeed, a national surveillance study conducted in Australia showed that the rate of children aged 5–13 years diagnosed with an ED had doubled compared to ten years ago [ 16 ]. However, pathophysiological mechanisms underlying the development and maintenance of ED behaviours and cognitions remain poorly understood [ 17 ]. Furthermore, less than half of people with EDs who undergo ‘evidence-based’ treatments make a full, long-term, recovery [ 18 ]. The high relapse and mortality rates observed for EDs, even for those who can access treatment [ 19 ], have prompted an increased recognition of the need for more lived experience-led research [ 20 ] and person-centred care [ 21 , 22 ]. Neurodiversity refers to the diversity of human minds that naturally exists in our society [ 23 ]. Neurodivergence refers to individuals who have a neurocognitive profile that diverges from the majority (i.e., atypical), whereas neurotypical people are those whose neurocognitive profile is considered aligned with the majority (i.e., typical) [ 24 ]. There are many forms of neurodivergence, such as autism, attention deficit hyperactivity disorder (ADHD), Tourette’s syndrome, giftedness, dyslexia, dyscalculia, dyspraxia, and intellectual disability (ID) [ 11 , 25 ]. Furthermore, while neurodivergence is commonly associated with autism and ADHD, other neurocognitive variations, such as obsessive–compulsive disorder (OCD), schizophrenia, and psychotic spectrum disorders, can also be regarded as part of the neurodivergent umbrella [ 26 – 28 ]. Individuals with OCD, psychosis, and/or schizophrenia may experience differences in sensory and perceptual or spatial processing [ 29 – 32 ], different modes of inner speech (e.g., intrusive thoughts, voice-hearing experiences) [ 33 ], social-emotional processing [ 34 ], movement, and expression [ 28 , 35 ]. Neuronormativity relies on the assumptions and beliefs which uphold neurotypical privilege—an ideologically oppressive framework that views some personality and neurocognitive traits as inherently superior to and more valuable than others [ 36 , 37 ]. The neurodiversity paradigm, however, rejects the assumption of there being a singular ‘normal’ personality or neurocognitive profile and opposes the pathology paradigm, whereby those who do not fit within the narrow socially constructed understanding of ‘normal’ are perceived as deficient, disordered, and in dire need of fixing or curing [ 38 – 40 ]. In this view, the neurodiversity paradigm argues that difference should not be regarded as inherently pathological and that the diversity of human minds should be accepted and accommodated [ 40 , 41 ]. Research has highlighted high co-occurrence rates between autism, ADHD, and EDs [ 42 , 43 ]. Preliminary research has also associated other forms of neurodivergence, such as ID [ 44 ] and Tourette’s syndrome [ 45 , 46 ], with disordered eating. Yet, the mechanisms underlying links between neurodivergence and EDs remain understudied and poorly understood; for instance, there is limited research on the intersectional nature of ED risk factors for neurodivergent individuals (NDI). Subsequently, NDI are shown to experience poorer ED treatment outcomes compared to neurotypical people [ 47 , 48 ]. Although there have been calls to include lived experience expertise at the core of mental health research [ 49 ], participatory and phenomenological approaches are rarely utilised in the context of EDs [ 50 – 53 ]. With all co-authors being neurodivergent and having lived or living experience of an ED, our narrative review aims to provide a lived experience-based conceptual map to serve as a starting point for future empirical work. A high-level summary of the co-occurring prevalence rates between neurodivergence and EDs is first provided, followed by an investigation of key factors underlying ED risk for NDI. Additionally, a critical appraisal of current ED research and treatments is provided before recommendations for neurodiversity-affirming (NDA) ED care are discussed.

Prevalence data highlights that autistic people and those with ADHD are at an increased risk of developing EDs compared to those who are neurotypical. The limited research on other forms of neurodivergence and eating psychopathology has found them to be associated with high incidences of disordered eating, highlighting a need for more research. It is possible that the real co-occurrence rates between neurodivergence and EDs are higher as issues such as the underdiagnosis or misdiagnosis of autism and ADHD in AFABs and racially marginalised populations, lack of tailored psychometric instruments, and limited research despite documented overlaps between various forms of neurodivergence potentially lead to underestimation of true prevalence rates. Given the overrepresentation of neurodivergence in ED populations, it is ideal for ED clinicians to promptly screen for autism and ADHD to improve ED care individuation. There are a wide range of interwoven factors underlying ED risk for NDI, from atypical sensory processing, executive functioning, social communication, emotional processing, and a higher likelihood of experiencing chronic illnesses and systemic discrimination. Therefore, it is important that future investigations into ED risk in the context of neurodivergence take an intersectional approach. Furthermore, transdiagnostic research factoring in the overlaps between different neurotypes is also warranted given that co-occurring forms of neurodivergence appear to have an impact on ED presentations and the nature of support needs. Future empirical work should be participatory, non-tokenistic, and lived experience-informed, and acknowledge power structures associated with discrimination and trauma to better achieve epistemic justice and better translate to cultural humility in ED care.

All co-authors are neurodivergent and have lived experience of an eating disorder. We are researchers and/or clinicians specialising in the field of eating disorders. We also occupy a wide range of other intersecting marginalised identities across gender, sexuality, body size, class, and neurotype; for example, some of us identify as gender diverse (e.g., transgender, gender non-binary) and/or live in larger bodies. However, we acknowledge that there remain many experiences we cannot speak to as we identify as white, none of us has an intellectual disability, and we are all living in Western countries (Australia and the United States of America). Our broad collective personal and professional experiences have informed our investigation of intersectional factors related to disordered eating and/or eating disorder risk, critical appraisal of current eating disorder research and therapies, and considerations for neurodiversity-affirming eating disorder care. We take on a neurodiversity-affirming theoretical framework. Therefore, we purposefully use non-pathologising and non-stigmatising terminology in relation to neurodivergence. For example, we use terms such as ‘traits’, ‘features’, ‘characteristics’, ‘support needs’, ‘differences’, ‘atypical’, ‘challenges’, or ‘difficulties’ rather than ‘abnormal’, ‘symptoms’, ‘deficits’, ‘impairments’, ‘tantrums’, ‘attention seeking’, ‘dramatic’, ‘treatment resistant’, or ‘non-compliant’ [ 1 – 3 ]. We also use identity-first language (i.e., autistic person) in relation to autism as research shows this is the preference of English-speaking autistic individuals [ 2 , 4 ]. Additionally, we refrain from using the terms ‘low/high functioning’ as these perpetuate erroneous and monolithic stereotypes—that is, autism being a linear construct [ 5 , 6 ]—and fail to capture the dynamic and fluctuating nature of disability [ 7 , 8 ].

The reasons underlying NDI being at a high risk of disordered eating and/or EDs are varied, complex, and multifaceted, and encompass a wide range of interconnected environmental, psychosocial, and biological factors [ 154 ]. NDI experience the world differently in a myriad of ways, as illustrated by differences in sensory processing, communication and socialising (e.g., the double empathy problem), thinking and learning styles (e.g., monotropism, inertia), gender identity, sexual orientation, sex diversity, and executive functioning [ 154 – 156 ]. Together, these differences may contribute, either directly or indirectly, to a wide array of pathways potentially contributing to the development of disordered eating, EDs, and/or body image disturbances for NDI [ 9 , 10 , 20 , 53 , 63 , 154 , 157 , 158 ]. Sensory processing differences encompassing exteroception (related to external stimuli such as sounds or tastes), interoception (related to internal stimuli such as thirst, pain, or hunger), and alexithymia (the ability to recognise and/or describe affective states and associated physiological sensations) are core overarching characteristics associated with neurodivergence [ 29 , 159 – 166 ]. Collectively, atypical sensory processing patterns are linked to EDs and body image disturbances, irrespective of neurotype [ 167 , 168 ]. Additionally, research shows that atypical sensory processing, as experienced by autistic people more specifically, is strongly associated with disordered eating [ 169 – 171 ]. Exteroception and interoception influence cognitive processes such as goal-directed behaviours (e.g., eating, drinking), motor control and coordination, memory, learning, motivation, self-perception, and emotional responsiveness, and are tied to neurobiological systems, particularly thalamo-cortical connectivity [ 172 – 174 ]. The thalamus is located at the centre of the brain and has historically been perceived as a passive relay centre for sensory information [ 175 ]. However, a growing body of evidence shows that the thalamus plays a more active and important role in cognitive functions than previously thought, including memory and attentional control [ 176 , 177 ], executive functioning [ 178 ], and emotional processing (e.g., thalamo-limbic connectivity) [ 179 ]. Furthermore, emerging neuroscience research suggests that the thalamus is involved in a wide range of psychiatric conditions (e.g., depression, anxiety, bipolar disorder, post-traumatic stress disorder, substance abuse) [ 180 – 182 ], including EDs [ 183 – 185 ]. Atypical sensory processing likely underlies differences in eating processes and habits as well as vulnerability to ED development [see 186 ]. There are different ways in which atypical sensory processing may lead to eating difficulties and/or disorders. Exteroception may contribute to dietary restriction resulting from aversions, whereas sensory seeking and associated cravings tied to specific tastes or textures may underlie bingeing [ 167 ]. Heightened exteroception (e.g., hyperacusis, misophonia) may also contribute to the preference for eating alone to avoid sensory overwhelm [ 187 , 188 ]. Moreover, interoception may impact eating through the modulation of hunger and satiety; for instance, feeling full after consuming small amounts of food may underlie restriction, while delayed satiety may promote bingeing and/or eating in the absence of hunger [ 167 ]. Interoception further plays an important role in body image and has been associated with body image disturbances [ 157 , 189 , 190 ]. Alexithymia refers to difficulties regarding the identification and description of one’s own emotions and feelings; as noted by Gramaglia et al. [ 191 , p. 1], “alexithymic individuals usually show a paucity of words to describe their affective status and find it difficult distinguishing feelings from physical sensations.” As alexithymia relates to an individual’s capacity for appraising and distinguishing between affective and non-affective states (e.g., sadness/hunger), it is strongly associated with interoception [ 192 – 194 ] and emotional processing (e.g., difficulties navigating instances of emotional arousal) [ 195 , 196 ]. Furthermore, research shows that alexithymia is involved in EDs [ 73 , 197 – 199 ]; however, the relationship between alexithymia and EDs is complex and multifaceted, encompassing inter-related sensory processing, emotional awareness, and psychosocial domains [ 200 – 202 ]. Alexithymia is linked to neurodivergence, including OCD, ADHD, and autism [ 203 – 207 ]. Relatedly, research suggests that alexithymia mediates, at least partially, the high co-occurrence between EDs and neurodivergence; for example, El Archi et al. [ 208 ] found alexithymia to mediate the relationship between binge eating and ADHD, while Vuillier et al. [ 72 , p. 1] found that alexithymia “could partially explain the prevalence of ED in autistic people.” Eating a restricted range of foods is often framed as an ED symptom. While this phenomenon may reflect psychopathology for some, qualitative research shows that, for autistic people, it may be an adaptive and self-regulatory mechanism to prevent sensory distress, cognitive overwhelm, and mealtime anxiety [ 68 , 158 , 209 ]. Indeed, research suggests that autistic people do not habituate to sensory stimuli in the same ways as non-autistic people [ 210 , 211 ]. In the context of such reduced habituation, coercing autistic people to eat aversive foods may be experienced as traumatic and unintentionally exacerbate feeding difficulties (e.g., food phobia) [ 212 ]. It is therefore critical to take the different ways NDI experience eating and feeding from a sensory perspective into account for ED identification, care, and recovery [ 9 , 10 , 20 , 47 ]. Executive functioning differences (e.g., task initiation, memory, demand avoidance) are influenced by sensory processing patterns [ 213 ] and are observed in many NDI [ 214 ]. Executive functioning differences are also linked to EDs [ 215 , 216 ]; for example, executive functioning differences may contribute to difficulties related to remembering to eat, grocery shopping, meal preparation, consistently following a meal plan, and/or choosing what to eat [ 68 , 154 , 167 ]. Such difficulties may promote restrict-binge cycles, whereby forgetting to eat or being unable to organise meal preparation leads to prolonged periods of time without eating, which may be followed by compensatory bingeing episodes [ 154 ]. Additionally, given that many autistic people have a literal, or concrete, thinking style [ 24 , 217 ], dichotomous portrayals of ‘healthy’ or ‘good’ versus ‘unhealthy’ or ‘bad’ foods associated with diet culture and public health messages may contribute to fixations towards foods considered ‘healthy’ (e.g., orthorexia) [ 61 , 154 , 218 ]. People with a diverse sex, sexuality, and/or gender identity are more likely to be autistic and/or have ADHD [ 219 – 225 ] and face a high risk of EDs and body image disturbances [ 226 , 227 ]. Factors contributing to the increased risk of EDs among neurotypical sex, sexuality, and gender diverse people include gender dysphoria, internalised cis-heterosexist ideals, and minority stress [ 186 , 226 , 228 – 233 ]. In contrast, identity pride, community connectedness, and gender-affirming healthcare can be protective factors [ 186 , 226 , 233 , 234 ]. Trans people experience disproportionate rates of EDs compared to cisgender individuals [ 230 – 232 , 235 , 236 ]. EDs, in this context, may function as a way to manage gender dysphoria; for instance, restrictive eating can delay the onset of puberty or minimise the appearance of secondary sex characteristics [ 63 , 230 – 232 ]. ED risk for trans people is further exacerbated by minority stressors, including discrimination and social exclusion, which can lead to heightened body dissatisfaction and increased engagement in ED behaviours [ 237 ]. Additionally, high levels of gender dysphoria are observed in the autistic population, partially mediated by atypical sensory processing, contributing to greater difficulties in coping with gender incongruence and body image disturbances [ 221 – 224 , 238 , 239 ]. Variations of sex characteristics (VSC)—otherwise referred to as intersex—are differences in innate sex characteristics that diverge from social and medical norms [ 240 ]. In a study of 1022 people with VSC, 8.8% had a diagnosis of autism, 9.1% had high autistic traits, and 4.1% screened positive for ADHD [ 221 ]. Higher autistic trait scores were associated with lower self-esteem in participants with Turner Syndrome, Klinefelter Syndrome, and female participants with 46XY variations with and without androgenisation [ 221 ]. Additionally, higher ADHD trait scores were associated with poorer body image in participants with Turner Syndrome, female participants with 46XY variations with and without androgenisation, and male participants with 46XY variations [ 221 ]. ED diagnoses were reported in 11.1% of the overall sample (3.7 times higher than the general population); however, stratified data of ED diagnoses for autistic intersex people and those with ADHD were not provided [ 221 ]. Intersectionality plays a crucial role in understanding elevated ED risk for NDI who are also members of the Lesbian, Gay, Bisexual, Transgender, Queer, Intersex, and Asexual (LGBTQIA +) community, highlighting the need for targeted interventions and support systems that address the unique intersections of neurodivergent and LGBTQIA + identities [ 219 , 227 ]; for example, by acknowledging that dietary restriction may stem from different motivations in this population (e.g., sensory and gender dysphoria-based), underscoring the importance of NDA ED care also being gender-affirming. Giftedness, which refers to outstanding intellectual and/or creative potential [ 241 ], may represent a risk for ED development through perfectionism, low self-esteem, striving to achieve social acceptance, and avoiding ostracization and rejection [ 242 – 244 ]. Obsessive perfectionism may develop in gifted people within the context of interpersonal trauma and marginalisation, wherein an individual pursues creative and/or intellectual achievements as a survival process to protect autonomy, identity and individuality, and to anesthetise psychological pain [ 245 – 247 ]. High academic and/or creative achievement in the context of giftedness may be an asset to accepting uniqueness and social difference; however, this may become a burden when an individual's emotional vulnerability and associated need for support are not recognised and addressed [ 248 ]. Twice-exceptional people 6 may also experience psychosocial vulnerability and imposter syndrome due to asynchrony—referring to disparities between one’s chronological and developmental trajectories, belongingness with peers, and expectations [ 242 ]. Perfectionism and disordered eating may be an attempt to cope with the psychosocial distress [see 249 ] associated with such asynchrony [ 153 , 242 , 250 ]. Twice-exceptionality may also present a social and identity threat due to the experience of being ‘twice different’ [ 251 ]; thereby, disordered eating may serve as a social survival strategy to fit in with peers through striving toward societal body ideals [ 242 ] and alleviate the existential angst that may accompany fears of failure and/or imposter syndrome [ 252 ]. NDI commonly experience systemic oppression and discrimination across varying settings and contexts (e.g., education, healthcare, employment) [ 253 – 260 ]. Such forms of systemic discrimination increase the likelihood of unemployment [ 261 ], houselessness [ 262 ], and food and nutrition insecurity [ 263 , 264 ], therefore contributing to psychosocial distress and mental ill-health [ 265 ], including EDs [ 266 ]. Indeed, high rates of houselessness and unemployment have been reported for NDI [ 261 , 263 , 267 – 270 ]. The risk for autistic people to be food insecure is four times greater compared to non-autistic people [ 263 ]. Moreover, a systematic review highlighted that between 10 and 40% of adults with ID experienced houselessness [ 271 ]. Food insecurity is a known ED risk factor, particularly for BN [ 272 – 274 ]. Rejection and stigmatisation are common experiences for NDI, and negatively impact self-esteem and psychosocial wellbeing [ 275 – 277 ]. Rejection and stigmatisation are often rooted in deficit-based and dehumanising stereotypes, such as that autistic people “lack empathy” and “social skills” [ 278 – 281 ] or that those with ADHD are “lazy” and “attention seeking” [ 282 , 283 ]. Resultantly, NDI are prone to internalise such stigma and may hide their authentic selves to avoid being rejected and bullied—a phenomenon referred to as masking or camouflaging [ 284 ]. Masking is associated with loss of sense of self and identity confusion [ 285 ], which are known risk factors for EDs [ 286 ]. Furthermore, substantive evidence supports the link between masking and poor mental health outcomes in autistic people, such as anxiety, self-harm, suicide [ 287 – 290 ], and EDs [ 60 , 61 ]. As Ruth [ 56 ] puts it: “ I felt alienated, inadequate, lonely and confused. I would return from school as a ball of pent-up energy and emotion. It was at this time that I began to experience daily episodes of detached but frantic eating, terminated only by the physical impossibility of continuing. Prolonged and repeated experiences of social rejection left me vulnerable to seeking acceptance by whatever means necessary. I had already received the message from numerous external sources that being in a smaller body was somehow preferable. As my body grew in response to the regular binges, I began to receive negative feedback, and encouragement to alter my body size through dieting. ” “ I felt alienated, inadequate, lonely and confused. I would return from school as a ball of pent-up energy and emotion. It was at this time that I began to experience daily episodes of detached but frantic eating, terminated only by the physical impossibility of continuing. Prolonged and repeated experiences of social rejection left me vulnerable to seeking acceptance by whatever means necessary. I had already received the message from numerous external sources that being in a smaller body was somehow preferable. As my body grew in response to the regular binges, I began to receive negative feedback, and encouragement to alter my body size through dieting. ” NDI have been found to experience higher rates of interpersonal victimisation, including domestic abuse and sexual violence, compared to neurotypical individuals [ 291 – 294 ]. For example, nine out of ten autistic women [ 295 ] or those with an ID [ 296 ] report having been sexually assaulted, which is three times more than neurotypical women. Additionally, a systematic review suggested that those with ADHD were more likely to have experienced childhood sexual abuse compared to those without ADHD [ 297 ]. Such experiences of victimisation are highly relevant as interpersonal trauma, including sexual abuse, has been found to substantially increase the risk of EDs [ 266 , 298 – 300 ]. There is emerging evidence suggesting that autistic people experience post-traumatic stress triggered by factors not currently conceptualised within psychiatric nosology as inherently traumatic [ 301 – 304 ]; for example, some autistic people may experience post-traumatic stress resulting from repeated exposures to aversive sensory input (e.g., sounds, smells, tastes, textures) [ 212 , 279 ]. It is estimated that up to 60% of autistic people suffer from post-traumatic stress disorder in their lifetime [ 304 ], with many thought to be misdiagnosed (e.g., borderline personality disorder) and not receiving the trauma-informed care they need [ 83 , 84 , 86 , 303 ]. Since post-traumatic stress disorder is a known contributor to the development of EDs [ 305 ], the delayed identification and delivery of compassionate treatment of trauma may therefore exacerbate the vulnerability towards developing EDs for autistic people. Intersecting experiences of marginalisation and structural oppression (i.e., racism and ethnocentrism, ableism, gender binarism, sexism, cis-heterosexism, and sizeism) may compound ED risk [ 300 , 306 , 307 ]. Indeed, in Nelson et al.'s [ 300 ] longitudinal study involving 331 undergraduate students, different forms of discrimination such as racism, weight stigma, and sexism all predicted body image dissatisfaction and disordered eating. The patterns of disordered eating were different depending on the type(s) of discrimination experienced [ 300 ], highlighting the importance of considering the influence of intersecting factors of identity, oppressive power structures, and trauma experiences in ED development and recovery, irrespective of neurotype [ 21 ]. NDI experience chronic medical illnesses at higher rates than neurotypicals [ 308 – 310 ]. The range of such medical conditions includes, but is not limited to, Ehlers-Danlos syndrome, postural tachycardia syndrome [ 311 ], type 2 diabetes [ 312 ], polycystic ovarian syndrome [ 313 ], endometriosis [ 314 ], asthma, dermatitis, food allergies [ 315 , 316 ], gut microbiome dysbiosis [ 317 , 318 ], and gastrointestinal disorders [ 319 , 320 ]. There is emerging evidence suggesting a bidirectional relationship between the aforementioned conditions and EDs; indeed, experiencing one of the conditions listed above may increase the risk of developing an ED later in life, while having an ED may increase the risk of developing one of the aforementioned conditions later in life [ 321 – 326 ]. Furthermore, there is growing evidence from the field of psychoneuroimmunology suggesting that psychological stressors, such as systemic discrimination, interpersonal victimisation, and childhood trauma, may be a key factor in immune and gut-brain dysregulation [ 327 – 329 ]. The link between psychological stress and gut-brain and immune dysregulation may be rooted in the mechanism whereby stress activates corticotropin releasing hormone (CRH), which in turn triggers the degranulation of mast cells and subsequent release of histamine as well as agents weakening the blood–brain barrier 7 [ 330 – 333 ]. Indeed, research points to bidirectional relationships between the immune system (e.g., mast cells, histamine, pro-inflammatory cytokines), the central nervous system [ 334 – 337 ], the endocrine system [ 338 , 339 ], and the digestive system [ 340 – 342 ]. Additionally, mast cells and histamine have been associated with a wide range of psychiatric (e.g., depression, anxiety) and neurological conditions (e.g., dementia, epilepsy, Parkinson’s disease) [ 343 – 346 ], including EDs [ 324 , 347 – 349 ]. This onnection may be mediated by histamine playing an overarching role in neurotransmission [ 334 – 338 ], including with regards to dopamine and serotonin [ 350 ], particularly in the thalamic region (i.e., sensory system) [ 351 , 352 ]. The interdependence between sensory processing and the immune system was further illustrated by Marques et al. [ 353 , p. 1], “[i]mmune and inflammatory responses share the core property of sensing, defining the immune system also as a sensory system.” Moreover, histamine influences glucose and lipid metabolism [ 354 ], leptin activity 8 [ 355 ], and control of energy expenditure (i.e., homeostasis), including propensity towards exercising [ 356 – 358 ]. As noted by Keeler et al. [ 359 , p. 1], “infections, autoimmune diseases, and stress-related alterations in the immune system influence brain circuits and neurotransmitters involved in appetite and mood regulation.” There are many intersectional factors underlying NDI being at a high risk of experiencing disordered eating and/or developing EDs. This underscores the importance of investigating ED risk through a holistic, intersectional, and lived experience-led lens. Taking these intertwined factors into account in treatment formulation is also of great importance to improve ED care individuation.