This journal entry has been adapted from two essays I wrote on the topic. A long one, but an interesting subject for many individuals...

Please do not plagiarise my work.

Human-written, AI-free!

Autism research has increased substantially over recent decades, while consistently demonstrating a higher prevalence in males than females, alongside a rise in diagnoses (Thurm & Swedo, 2012). That being said, growing literature highlights the under- recognition, misdiagnosis, and late diagnosis of autism in females, alongside increased co-occurring mental health issues. Prevalence estimates and diagnostic outcomes remain complicated by evolving diagnostic criteria and gender bias.

Gender bias

Growing evidence demonstrates that autism is presented differently in females compared to stereotypical males. Furthermore, girls are diagnosed later than boys, averaging at the age of 14. Research suggests that females tend to exhibit fewer repetitive behaviours and have higher cognitive abilities, resulting in less intense symptoms than males (G. Russell et al., 2021). This does not indicate ‘higher functioning’ because they experience additional challenges.

Halladay, A.K. et al (2015) hypothesise ‘Female protective effect’ [FPE], which suggests that females are ‘protected against some of the symptoms’ of ASD, in which they exceed the clinical diagnosis. For example, females often have higher language abilities, less aggression and hyperactivity than males; the ‘presentation of symptoms may be misinterpreted, and accurate diagnosis may be delayed.’ In addition to differences in development between the sexes, meaning females do not typically show signs until adolescence, whereas boys often present earlier, as toddlers.

Additionally, girls are more prone to mask and have higher social skills.

Why this is important: Late diagnosis can decrease quality of life and increase dysfunction and issues, such as difficulties maintaining work and relationships as adults. Therefore, early diagnosis, accompanied by the provision of resources and services for lifelong support, is paramount for individuals with ASD (Halladay, A.K. et al., 2015). A vast amount of research demonstrates a lack of communication within services, long waiting lists, and the burdensome costs for private diagnosis, complicating the system and diagnosis for females.

Inconsistencies in the diagnosis procedure in the UK

Diagnostic terminology and conceptualisations of autism have undergone substantial changes over time (Russel et al., 2022).

The Adult Psychiatric Morbidity Survey (APMS) is widely regarded as a source for estimating autism prevalence in the general population in England (NHS, 2025a). The APMS assessment has utilised a multi-stage assessment structure since 1993 (NHS, 2025a), which may no longer adequately reflect diagnostic variations, since autistic understanding and conceptualisation have evolved considerably over time (Greaves-Lord et al., 2022; Russel et al., 2022; King et al., 2014). Moreover, concerns have been raised over the ADOS (Autism Diagnostic Observation Schedule) threshold being set too high, insufficiently sensitive to detect less stereotypical presentations (NHS, 2025a) and possibly distorting prevalence estimates in women. Additionally, the ADOS has been criticised for prioritising traditionally male behavioural presentations, partially reducing diagnostic sensitivity and contributing to systematic under-identification of women (NHS, 2025a; Cruz et al., 2024; Rea et al., 2023; Yu et al., 2026).

Why this is a problem: These limitations have significant implications for mental health outcomes among autistic women. Delayed or missed diagnosis may increase vulnerability to co-occurring mental health conditions through prolonged masking, unmet support needs, social invalidation, and inappropriate treatment pathways (Belcher et al., 2023). Consequently, some mental health difficulties may not reflect co-occurring pathology, but systematic failures within diagnostic and support systems, rather than by autism itself, additionally causing trauma.

The female phenotype: subtle signs, elevated mental illness, and misdiagnosis

Belcher et al. (2023) proposed the female phenotype theory (FPT), suggesting that autistic women may present with more subtle signs or less stereotypical characteristics than autistic men. Results demonstrated that probably autistic women (showing signs of ASD but not officially diagnosed) were more likely to receive a diagnosis of borderline personality disorder (BPD), while diagnosed autistic women demonstrated elevated rates of mental health conditions than seen in men (Belcher et al., 2023).

Misdiagnosis: Borderline Personality Disorder (BPD)

Fitzgerald (2005) proposed a potential link between autism and the misdiagnosis of BPD, identifying substantial overlap in traits including social functioning, interpersonal relationships, identity disturbances, suicidal ideation, feelings of emptiness, anger, and paranoia, alongside impulsivity. Therefore, some individuals may be misdiagnosed with BPD when they are autistic.

Furthermore, Ryden and Hetta’s (2008) findings demonstrated that individuals reported significantly more suicide attempts, lower global functioning, poorer self-image and increased psychological severity compared with the non-autistic patients (Ryden and Hetta, 2008), highlighting the substantial mental health burden with overlapping presentations and associated trauma.

Why this is a problem: Overlapping features associated with conditions such as BPD (ICD-11) hold stigma, contribute to diagnostic confusion and delayed autism recognition. Tamilson et al. (2024) highlight the negative consequences of BPD misdiagnosis among autistic women. Participants reported experiences of stigma and inadequate treatment from clinicians, alongside significant barriers to obtaining an accurate autism diagnosis.

Why this is important: These findings suggest that co-occurring mental health conditions may complicate the identification of autism in women, particularly if characteristics overlap with other conditions. Consequently, elevated rates of diagnosis, like BPD, may partly reflect diagnostic misinterpretation or overshadowing, rather than true prevalence, raising concerns regarding the validity and gender sensitivity of current diagnostic criteria. Importantly, receiving an autism diagnosis improved psychological well-being and self-understanding for many participants (Tamilson et al. 2024).

Trauma and camouflaging

Further complicating differential diagnosis, trauma has been proposed as a mediating factor contributing to overlapping presentations. Dell’Osso et al. (2018) suggest that autistic individuals may develop maladaptive coping strategies in response to adverse experiences, such as bullying, social exclusion, and chronic invalidation, which may resemble BPD symptomology. This aligns with growing evidence indicating that Complex Post-traumatic stress disorder (C-PTSD) shares substantial overlap with BPD traits, like emotional dysregulation, interpersonal instability and disturbances in self-concept (Karatzias et al., 2023; Sarr et al., 2024; Buuren, Hoekert & Sizoo., 2021).

The effect of Camouflaging

Baldwin and Costley (2016) identified elevated rates of co-occurring psychological difficulties among autistic women engaging in masking behaviours, including isolation, emotional exhaustion, stress, and psychological distress. The authors emphasised that mental health concerns may be particularly significant among autistic individuals who do not conform to stereotypical presentations of autism, resulting in unmet support needs and delayed recognition.

Bargiela et al. (2016) suggested that prolonged masking was associated with heightened anxiety, emotional exhaustion, and identity confusion. Participants additionally reported significant barriers to healthcare access, partly to not appearing ‘typically autistic’ and engaging in behaviours intended to conform to non-typical social expectations.

Hull et al. (2017) identified difficulties relating to self-perception.

Hull et al. (2020), a subsequent systematic review and meta-analysis, evidence suggests that individuals presenting with a female autism phenotype, engaging in camouflaging, are significantly more likely to remain undiagnosed or experience delayed diagnosis (Hull et al., 2020).

Collectively, these findings suggest that masking may contribute to misinterpretation of autistic traits but also function as a chronic psychological stressor that intensifies vulnerability to co-occurring mental health conditions, including depression, identity disturbances, emotional exhaustion, and suicidal ideation.

External factors:

Grzeszak and Pisula (2025) suggest that external stressors, including stigma, social pressures, and discrimination, significantly contribute to poor mental health outcomes and disadvantages.

Yau, et al, (2023) found that autistic women experience increased risks of bulling, exploitation, abuse, exclusion, and lack of support, contributing to negative self-perception and psychological distress, impacting global functioning and mental health.

These findings may suggest that co-occurring mental health difficulties may be strongly influenced by environmental and social stressors, rather than autism alone.

Why this is important: Gelli and Marczak (2023) reported that delayed autism diagnosis was associated with poorer mental health outcomes, while accurate diagnosis improved self-understanding and well-being. Therefore, it’s imperative to provide proper training for health care providers, since support, belonging, and self-acceptance enhance well-being and foster a sense of self and identity (Yau et al, 2023).

Mental health

Substantial evidence demonstrates elevated rates of co-occurring mental health conditions among autistic individuals, particularly for women diagnosed later in life (Yall-Smith, 2008; Lacroix et al, 2026).

Up to 70% of autistic individuals have at least one co-occurring mental health condition, significantly exceeding prevalent rates in the general population (Martini et al., 2022).

Autistic individuals are seven times more likely to die by suicide (NHS: National Health Service, 2025b).

Only 22% reportedly receive appropriate support (Mandy, 2022).

Autistic individuals have increased experiences of discrimination, social disadvantage, and poorer health outcomes.

Autistic women are reported to experience increased rates of self-harm, suicidality, hospitalisation and poor quality of life, compared to both autistic men and non-autistic women. (Mason, et al., 2018; Martini et al., 2022; Lacroix et al., 2026). This highlights the disproportionate burden of mental health difficulties, particularly since there are fewer interventions, resources, and awareness for this cohort (Clearly et al., 2023).

Eating disorders are significantly elevated in autistic women, with 70% having atypical eating patterns, and up to 37% having a distinct style of restrictive anorexia. These are associated with poorer treatment outcomes and clinical complexity (Mayers & Zickgraf, 2019; Westwood & Tchanturia, 2017; Brede et al., 2020). This is concerning since eating disorders have the highest rate of mortality for mental health conditions (Auger et al, 2021).

In conclusion, the impact of seeing past a male-dominated perspective can improve girls’ and women’s lives as they deserve. Reducing the emotional distress of females, while encouraging autistics in the workforce by providing reasonable adjustments, reduces benefits and healthcare costs, improving society.

Phoenix85

Header images from Google. and post images from UnSplash, provided by Square space website.

This blog post is also published on DigiPsych.

Autism spectrum disorder: The rise in diagnosis, the female autism phenotype, and substantial issues