Autism

Autism has been described as an ‘extreme male brain’ (Baron-Cohen Reference Baron-Cohen2002) and has been traditionally viewed as a male-dominated condition. The difference in prevalence by sex is not as marked as previously thought: the ratio is as low as 3:1 male:female in populations that have not been referred for asessment, compared with 4:1 in a referred sample (Loomes 2017). The rate of increase in diagnoses in recent years has been greater for females than males, with closer sex ratios in adults (Russell 2022). The gender ratio in clinical samples reduces markedly with age, from 5.5:1 in younger children to closer to 2:1 in adults (Rutherford 2016). These figures suggest that many autistic females are going undetected. There are several plausible explanations for these differences.

The first possibility is that there is truly a sex difference in prevalence. The ‘female protective effect’ theory purports that females need a higher genetic or aetiological load to manifest the phenotype (Wigdor 2022). Although this may be the case, biases in diagnostic practices cannot be ignored (Brickhill Reference Brickhill, Atherton and Piovesan2023). Females display less severe externalising problems than their male peers, experience fewer difficulties in the school environment and so are less prone to being referred for assessment (Mandy 2012).

Second, the female phenotype of autism is not yet fully understood and diagnostic tools and constructs, which were validated in largely male samples, do not capture it adequately, nor are they sensitive to gender differences (Cruz Reference Cruz, Zubizarreta and Costa2025). Biases in these tools have been exposed, where females need more severe symptom expression to reach a threshold for diagnosis (Dworzynski Reference Dworzynski, Ronald and Bolton2012), and previous research relies on these scales, thus compounding the male bias (Ratto 2018). There is a pressing need to develop instruments that are more sensitive to symptom expression in women and girls (Brickhill Reference Brickhill, Atherton and Piovesan2023). This clinical ascertainment bias contributes to many high-functioning women and girls being missed or diagnosed late (Lai 2015).

Third, the diagnosis of autism relies on two core components: repetitive and restrictive behaviours (RRBs) and social communication deficits (World Health Organization 2019/2021), both of which manifest differently in females and so contribute to underdiagnosis. Autistic females show fewer RRBs and their focus may overlap with neurotypical interests, such as fashion, celebrities or music (Bourson Reference Bourson and Prevost2024), and so may not be appreciated as an RRB. Social communication deficits may be obscured by masking or camouflaging – strategies used by autistic people to present themselves in a neurotypical manner, with females employing these strategies to a greater extent than males, potentially because females recognise the importance of social relationships more (Cook Reference Cook, Hull and Crane2021). Masking is exhausting and is associated with poorer mental health outcomes in both sexes (Hull 2021; Cruz Reference Cruz, Zubizarreta and Costa2025).

Fourth, reproductive events may affect the psychological health of autistic women to a greater extent than non-autistic women, although the literature is still limited: autistic women experience greater levels of perinatal depression and anxiety (Hampton 2022), some autistic women find menstruation distressing due to difficulties with emotion regulation and sensory challenges (Steward 2018), and autistic women experience the menopausal transition as more challenging as it may destabilise their ability to mask their autistic symptoms, exacerbate sensory difficulties and lead to greater deterioration in their mental health (Moseley 2020).

Attention-deficit hyperactivity disorder

There are similar patterns in the sex and gender disparities in attention-deficit hyperactivity disorder (ADHD). Females are diagnosed later than males, with the gender balance in diagnosis shifting considerably from childhood (10:1 male:female) to adulthood (2.73:1 male:female) (Williamson 2015). Adult women with ADHD self-report moreimpairment from their ADHD symptoms (Williamson 2015).

Referral, diagnosis and treatment are largely influenced by levels of hyperactivity and impulsivity rather than inattention, resulting in many girls and women being overlooked by parents or teachers as they display less hyperactivity and less severe externalising problems than their male counterparts (Mowlem 2019; Young 2020). In clinical settings, girls with ADHD exhibit comparable levels of hyperactivity to boys with ADHD but more inattentive symptoms and internalising problems (Young 2020; Loyer Carbonneau 2021). Ratings scales may not reflect the symptom profile in females and, where available, sex-specific cut-offs should be used along with greater reliance on collateral history (Young 2020; Loyer Carbonneau 2021). The use of clinical samples in ADHD research, which is already subject to referral and reporter bias, has resulted in a gender data gap for this disorder. Transitional stages, such as the move to university, employment or the postpartum and matrescence, can reveal otherwise unnoticed ADHD symptoms in women (Young 2020).

In line with ADHD-affected females experiencing more internalising problems than men, they suffer greater rates of comorbid mood, anxiety and eating disorders (Gershon 2002). Men with ADHD are more often diagnosed with antisocial personality and conduct disorders compared with women with ADHD (Williamson 2015). However, both sexes with ADHD have a greater rate of substance use compared with their unaffected counterparts. Additionally, women and girls with ADHD are at an increased risk of unplanned pregnancy compared with unaffected peers (Young 2020).

Sex hormones (oestrogen or progesterone) also have an impact on symptom trajectory and treatment response across various stages of women’s reproductive life, including menarche, the menstrual cycle, the postpartum and menopause (Roberts 2018; Young 2020). Premenstrual exacerbation has been observed in ADHD (Nolan 2022). Doses of stimulant medication may need to be adjusted in hypo-oestrogenic states, such as in the perimenstrual phase of the cycle or postmenopause, to maintain stable treatment response (Roberts 2018).

Schizophrenia

The male to female prevalence ratio for schizophrenia is 1.1:1 (GBD 2019 Mental Disorders Collaborators 2022). Men are diagnosed at an earlier age, with a sharp peak in late adolescence, whereas women have a more level incidence rate and a peak in midlife that may be related to menopause (formal studies using menopausal staging rather than simply age have yet to be completed) (Hafner 2019; Sommer 2020). Because of the later age at onset in women, they benefit from more time to amass social and educational advantages compared with men and initially do better than their male counterparts, although they lose this advantage several years into their illness (Sommer 2020). After midlife, women have a worse illness course than men (Riecher-Rossler 2018). Symptomatically, females experience more positive and fewer negative symptoms than males (World Health Organization 2024a) but suffer more affective symptoms (Sommer 2020).

The female reproductive hormone oestrogen is believed to be responsible for the gender disparities observed in schizophrenia and appears to have a protective effect against psychosis (Riecher-Rossler 2018). Hypothalamic–pituitary–gonadal axis dysfunction has been observed in those with psychosis independent of antipsychotic medication side-effects (Riecher-Rossler 2018; Brand Reference Brand, de Boer and Sommer2021). Oestrogen has neuroprotective properties and increases sensitivity to and bioavailability of dopamine (Brand Reference Brand, de Boer and Sommer2021; Culbert Reference Culbert, Thakkar and Klump2022). Symptoms of psychosis can fluctuate across the menstrual cycle and states of oestrogen withdrawal, both physiological and pharmacological, can trigger psychosis (Reilly 2020; Nolan 2022). The most dramatic example of this is the extraordinarily high incidence of psychosis in the postpartum relative to the entire lifespan (Kendell 1987). Higher psychiatric admissions for psychosis have also been observed during phases of the menstrual cycle when oestrogen levels are lowest (i.e. the late luteal and menstrual phases) (Reilly 2020) but it is unclear whether this is a subgroup of women driving these increased rates or an effect seen in the majority of women.

There are important sex differences in the treatment effects of antipsychotics and this evolves across the female life cycle. There is a reduced antipsychotic response postmenopause due to reduction in endogenous oestrogen (Brand Reference Brand, de Boer and Sommer2021). Early promising evidence indicates that oestradiol may be an effective add-on treatment for women with schizophrenia as they approach menopause (Reilly 2020; Brand Reference Brand, de Boer and Sommer2021) and there is also emerging evidence of a benefit of adjuvant selective oestrogen receptor modulators (SERMs) such as raloxifene, which act on the brain and bones but not the uterus or breast tissue (Brand Reference Brand, de Boer and Sommer2021). Females are more prone to antipsychotic side-effects, specifically metabolic effects and hyperprolactinaemia (Ercis Reference Ercis, Sanchez-Ruiz and Webb2024). Hyperprolactinaemia, which exerts negative feedback on the hypothalamic–pituitary–gonadal axis, reduces oestrogen levels, and therefore prolactin-sparing antipsychotics should be favoured for reproductive age women to maximise treatment efficacy and avoid the hypo-oestrogenic effects on bone and cardiovascular health (Brand Reference Brand, de Boer and Sommer2021). Prolactin-raising antipsychotics may also contribute to an increased risk of breast cancer observed in women with schizophrenia (Leung 2022). The risk of cardiovascular disease increases for women around menopause, underscoring the importance of proactive metabolic monitoring and management of cardiovascular risk factors such as smoking (Seeman 2012). Aside from the effect of oestrogen on dopamine transmission, the physical symptoms of menopause, such as sleep disturbance and hot flushes, could provoke a relapse of psychotic symptoms (Riecher-Rossler 2017).

There is a pressing need to tailor treatment of psychosis in women according to their reproductive hormone status, particularly low oestrogen states such as the perimenstrual and postmenopausal periods. Although the evidence is not at the point yet of recommending adjunctive oestrogen to treat schizophrenia, hormone replacement therapy (HRT) for women with schizophrenia is potentially more important than for the general population and they should be counselled effectively on this within the psychiatric setting (Brand Reference Brand, de Boer and Dazzan2022).

Mood disorders

Women are twice as likely to develop depression in their lifetime compared with men (Salk 2017). Reduced help-seeking and underdiagnosis in men may partly explain disparities in prevalence, as certain features of ‘male-typical’ depression, i.e. aggression, violence and substance use, may obscure underlying mood symptoms (Call Reference Call and Shafer2018). It is promising that treatment-seeking among males for common mental health conditions such as depression has improved in recent years in the UK, and is now similar to that in females (Clery Reference Clery, Morris, Wilson, S, S, T and S2025). Differing incidence between sexes emerges early during puberty, and is likely due to a convergence of biological and psychosocial factors (Salk 2017).

From a biological perspective, female reproductive hormones (oestrogen and progesterone) have potent effects on the brain and interact with many neurotransmitter systems (Deecher Reference Deecher, Andree and Sloan2008). The female brain is subjected to fluctuating hormone levels that it must adapt to during each menstrual cycle, and it experiences more erratic changes at both the beginning and the end of the reproductive life cycle (i.e. menarche and perimenopause) (Deecher Reference Deecher, Andree and Sloan2008).

Psychosocial stressors also play a part. From puberty, the emergence of body dissatisfaction, compounded by greater social media use, affects girls to a greater extent than boys (Kelly 2018). Women are also more frequently exposed to serious adversity, including childhood sexual abuse and domestic violence (Kuehner 2017), with nearly triple the risk for developing depression in those who have experienced intimate partner violence (Chandan Reference Chandan, Thomas and Bradbury-Jones2020). (We explore body dissatisfaction and serious adversity in the sections on eating disorders and violence/trauma below.) Women live with the stress of wider societal inequalities to varying degrees and this includes gender discrimination and the gender pay, promotion and pension gaps, which has implications for their economic security. The burden of caring responsibilities also disproportionately lies with women, which affects both their susceptibility to depression and their ability to seek treatment.

The sex ratio for bipolar disorder is balanced between males and females, although males are more prone to manic episodes and females more to depressive, hypomanic and mixed episodes (Diflorio Reference Diflorio and Jones2010; World Health Organization 2024a). As is the case for mental disorders more widely, men are more likely to suffer comorbid substance use disorders, which is a barrier to accessing appropriate treatment (Diflorio Reference Diflorio and Jones2010; World Health Organization 2024a). The most significant sex/gender considerations in bipolar disorder are the impact of reproductive events, i.e. the perinatal period, the menstrual cycle and menopause. The perinatal period is a critical time for onset of new mood disorders or relapse of pre-existing ones, owing to both biological and psychosocial factors, but this will not be discussed here.

Anxiety and obsessive–compulsive disorders

Women have double the lifetime prevalence of generalised anxiety disorder (GAD), panic disorder, agoraphobia and specific phobia (World Health Organization 2024a). One-third of women will meet criteria for any anxiety disorder at some point in their life, in contrast to 22% of men (McLean 2011). Comorbidities follow similar trends to the prevalence of mental disorders more widely, with females more prone to comorbid affective symptoms and males more prone to comorbid substance use disorders (McLean 2011), which, as mentioned above, has important implications for men accessing appropriate services. Intimate partner violence doubles risk of developing anxiety disorder (Chandan Reference Chandan, Thomas and Bradbury-Jones2020).

The sex ratio for obsessive–compulsive disorder (OCD) is similar, with lifetime prevalence of 1.5% for females and 1% for males (Fawcett Reference Fawcett, Power and Fawcett2020). The nature of OCD symptoms differs between sexes, with symmetry and taboo intrusive thoughts being more common in males, and females being more affected by symptoms revolving around cleaning and contamination (World Health Organization 2024a).

Premenstrual exacerbation of anxiety disorders, particularly panic disorder, can occur in a subset of women with these diagnoses (Nolan 2022) and there is some evidence of premenstrual worsening of OCD but evidence remains limited (Green 2022). Postpartum onset or exacerbation of OCD is common – compulsions frequently revolve around contamination or checking the baby, often to the detriment of sleep; ego-dystonic intrusive thoughts of causing harm to the baby are also common and are extremely distressing for new parents, who may be fearful of disclosing these to a healthcare professional (Hudepohl 2022).

Violence, trauma and stress-related disorders

The diagnosis of post-traumatic stress disorder (PTSD) has historically been associated with male experiences of trauma, originating from descriptions of soldiers returning from the Vietnam War and earlier accounts of ‘war neurosis’ and ‘shell shock’ of the First World War (Crocq Reference Crocq and Crocq2000). In an English population survey, PTSD was more prevalent in females (6.1% of women v. 5% of men), which should be unsurprising, given that 37.4% of females reported a lifetime traumatic event, compared with 32% of males (Wilson 2025). Females are also more likely to present with complex PTSD (cPTSD), where core PTSD symptoms are accompanied by difficulties with ‘affect regulation, persistent negative beliefs about oneself, and persistent difficulties in sustaining relationships’ (World Health Organization 2024a).

Sex differences in PTSD prevalence may arise from both psychosocial and biological factors: the distribution of traumatic events experienced by women differs from that for men. Females are more likely to be the victim of domestic and sexual violence, with mostly male perpetrators of this violence (Office for National Statistics 2024a) and 19% of women experience birth as psychologically traumatic (Xu 2025). The female stress response has also been described as differing from males – with females tending towards the ‘tend and befriend’ response to traumatic events, mediated by oxytocin, rather than the typically recognised ‘fight or flight’ response (Taylor 2006).

Psychiatric care can cause iatrogenic retraumatisation, particularly where use of restrictive practices is high (Sweeney 2018). Those with severe mental illness are more likely to have been victims of gender-based, domestic or childhood physical or sexual abuse (Fisher Reference Fisher, Morgan and Dazzan2009; Kessler 2010) and so it is essential that mental health and wider health services deliver trauma-informed care.

Eating disorders and body dysmorphia

Eating disorders are typically perceived in both the wider population and among healthcare professionals as conditions that affect predominantly women. Traditional thinness-oriented eating disorders (anorexia nervosa and bulimia nervosa) occur in 8.4% of women and 2.2% of men and disordered eating is increasing at a faster rate in males (Galmiche Reference Galmiche, Déchelotte and Lambert2019), with higher prevalence among sexual and gender minorities (Thapliyal 2014). Of course, owing to these sex ratios and the high mortality rates associated with anorexia, it is mostly women who die from this condition (Krug 2025).

Cultural gender norms are central to the pathology of eating disorders. The Western idealised female body is thin and the idealised male body is muscular (mesomorphic), and so body dissatisfaction and subsequent disordered eating behaviour inherently differ between genders. Male participants are poorly represented in eating disorder research (Thapliyal 2014), with resultant diagnostic classification being based on typical female symptoms and commonly used scales for symptom severity being developed using female patient populations (Murray 2016).

Muscle dysmorphia, a form of body dysmorphia classified under ‘obsessive–compulsive and related disorders’ in ICD-11 (World Health Organization 2019/2021), centres on an overvaluation of weight and shape (idealising the mesomorphic body type) and body image disturbance (a belief that one’s own body is too thin), with extreme behaviours to control weight and shape (such as excessive weight training, protein consumption and anabolic steroid use). The prevalence of muscle dysmorphia in men is in line with eating disorders in women and the associated behaviours can pose as great a medical and psychological harm to the patient as those of traditional eating disorders (Murray 2016).

The perception of eating disorders as female illnesses is a major barrier to them being recognised by male patients, their families and professionals, and hinders seeking and receiving treatment (Thapliyal 2014; Murray 2016). Services and treatments that are designed for thinness-oriented eating disorders fail to meet the specific needs of males who are affected by muscularity-oriented psychopathology (Thapliyal 2014; Bomben Reference Bomben, Robertson and Allan2022). Delayed recognition of the problem, along with internalised stigma due to perception of eating disorders as a ‘female problem’, prevent men from accessing timely treatment (Bomben Reference Bomben, Robertson and Allan2022).

Substance use disorders

Males have higher rates of comorbid substance use disorders across the spectrum of mental disorders (World Health Organization 2024a), although the historical gender gap in prevalence is closing (McHugh 2018). There are biological differences in the effect of substances between the sexes, such as where females experience higher levels of intoxication than their male counterparts for the same quantity of alcohol ingested and greater levels of physical harm (McHugh 2018). The ‘telescoping’ effect, where females develop dependence at a faster rate than males, has been observed for alcohol and opiates (Towers 2023).

Women with substance use disorders engage in treatment at lower rates than men (Greenfield 2007). Fear of losing custody of their children or being unable to attend either out-patient or residential treatment programmes owing to lack of childcare are frequently described barriers to seeking treatment (Adams Reference Adams, Ginapp and Price2021). Parenting responsibilities can be a strong motivator to address substance use, even if this is mandated by child protection agencies or the courts (Elms Reference Elms, Link and Newman2018). Women with substance use disorders are more at risk for gender-based or sexual violence, either in the past or present (The European Monitoring Centre for Drugs and Drug Addiction 2023) and the majority have experienced childhood trauma (Covington Reference Covington, Burke and Keaton2008). The combination of unsupportive co-parenting relationships and the threat of losing custody of their children means they are at greater risk of further trauma without appropriate support and services that are responsive to their needs.

Treatment programmes are male oriented and women experience significant barriers in accessing treatment and maintaining recovery – mixed-sex services, which may erode their sense of safety because of potential exposure to volatile male participants or ex-partners, the lack of childcare or child-friendly services that women can bring their children with them, and societal and self-stigma of being a woman or mother with a substance use disorder (Elms Reference Elms, Link and Newman2018). Gender-responsive addiction interventions that are adapted to their specific needs have the potential to improve treatment outcomes for women compared with mixed-sex programmes (Johnstone 2023). It is crucial that these are integrated with other services, including those for mental and physical health, and most importantly are trauma-informed in their delivery (Covington Reference Covington, Burke and Keaton2008; McHugh 2018).