Guidelines to Assist in Autism Diagnostic Assessments with Females: A Delphi Study

Abstract

Purpose: While growing evidence highlights the challenges faced by autistic females during diagnostic assessments, it remains unclear how well this research is applied in clinical practice. This study aimed to gather consensus among clinicians on recommendations for autism assessments to improve diagnostic validity for cis-gendered females.

Method: A three-round Delphi study was conducted with registered healthcare professionals experienced in conducting autism diagnostic assessments with females.

Results: Thematic analysis of Round 1 responses identified five themes: Attributes of the Assessor, Post-Clinical Qualification Education and Learning, Diagnostic Tools, Questions Asked, and Sources of Information. These themes produced 28 statements evaluated in Rounds 2 and 3. Eighteen reached 80% consensus in the final round. Experts emphasized the important of assessors having strong knowledge of research into autism in women, co-occurring conditions, other developmental conditions, and masking. Ongoing professional development on autism in women was considered essential and it was recommended that new assessors shadow female assessments. Experts agreed the ADOS-2 is not always useful for females, whereas screening tools can provide valuable narrative insights, though these need to be interpreted carefully due to potential male bias. Experts supported including more prompts to explore subtle differences in social communication and repetitive patterns of behaviour and stressed the importance of masking behaviours and experiences across different relationships. Masking should also be considered when interpreting informant information.

Conclusions: Findings reveal inconsistencies in how clinicians consider sex-related factors in autism assessments, highlighting the need for formal guidance. The next steps will be to develop this guidance, incorporating research evidence, the perspectives of autistic women and girls, and Delphi results. This aims to improve the experience, and accuracy of diagnosis for girls and women undergoing autism assessment, increasing consistency across services, and supporting clinicians.

Community Brief

Why is this an important issue?

Research tells us that women and girls find it difficult to get a diagnosis of autism. It can take a long time for them to get a diagnosis, and they may be wrongly diagnosed with another condition. Autism can look different in men and women, and professionals might not be able to pick up on these differences. Women can also mask (or hide) their autism from others during their assessment.

When a person has an assessment for autism, professionals in the UK follow the instructions of the NHS and the National Institute for Health and Care Excellence. However, these do not give professionals advice on how to assess women.

What was the purpose of this study?

This study looked at what professionals think is important for assessing women for autism. In this study, we were thinking about cis-gender autistic females (that is, a person whose gender identity matches the sex assigned to them at birth).

What did the researchers do?

Participants in this study were 34 professionals, such as Psychologists, Speech and Language Therapists, and Psychiatrists in the UK. They all had experience doing autism assessments.

We used a Delphi method. First, we asked these professionals how they assess women for autism. By looking for themes in their answers, we developed 18 statements that reflected each theme. Next, we gave these statements to the same professionals and asked them how much they agreed or disagreed with each statement. We asked professionals to do this twice to see whether they had changed their mind. We found the percentage of how much participants agreed with each statement. If this percentage was more than 80%, meaning most professionals agreed with it, it is an important part of autism assessments with women. This helps us to identify what best practice ‘should’ and ‘could’ involve.

What were the results and conclusions of the study?

There were differences between professionals in how they assess women for autism. This tells us that professionals need more guidance. The professionals agreed that they need to have a strong knowledge of the research into autism in women, masking, and other diagnoses that may be present. Screening tools, or questionnaires, that look at autistic traits were thought to be useful. Screening tools for other diagnoses were not recommended. Professionals want to ask questions about masking and camouflaging. Professionals think they should ask more questions and ask for more detail when assessing women. They did not think it would be helpful to ask questions about periods and menopause.

What is new or controversial about these findings?

These findings help us to identify what professionals think should be happening in an autism assessment with women.

What are potential weaknesses in the study?

Participants dropped out this study between stages. This is a small group of people, so we cannot say for sure that these are the views of most professionals in the UK. The professionals that took part were mostly of White ethnicity. This could lead to bias in the results.

How will these findings help autistic adults now or in the future?

The next steps will be the developing a document with guidance made up from other research, the voices of autistic women, and the results of this study. We hope this will make autism assessments of women more consistent and effective.

Introduction

Autism is a neurodevelopmental condition characterized by difficulties with social communication and interactions and restricted, repetitive patterns of behavior or interests.¹ A recent systematic review of the global prevalence of autism found a median prevalence of 100 autistic individuals per 10,000, with males consistently outnumbering females.² Although this suggests that autism occurs more frequently in males, there is some evidence that clinical presentations of autism differ between males and females, potentially affecting prevalence estimates.

Literature indicates that autistic women have fewer social impairments^3-5 and restricted and repetitive interests than males.^6,7 However, findings of behavioral differences can be inconsistent.⁸ For example, a meta-analysis of 20 studies examining gender^a differences in autistic symptoms found few differences in symptom severity between males and females.⁹ Autistic males and females showed similar symptom severity in communication and social behavior. Females did show less restricted interests/behaviors and stereotypical behaviors than males. Research in this area is ongoing and is not yet substantial enough to warrant revisions to the diagnostic criteria for autism in females.

Nevertheless, there remain concerns that autistic females are being missed, misdiagnosed, or delayed in receiving a diagnosis.¹⁰ Autistic women are less likely to receive a diagnosis of autism than men with similar levels of autistic traits¹¹ and are more likely to have been misdiagnosed with other conditions such as personality disorders.¹² Differences in the presentation of autism in females may be nuanced leading to an underestimation in assessments. For instance, restricted interests in females may simply be different in nature rather than in frequency, such as being in relational topics, which may be considered age and gender appropriate.¹³ It has been suggested that for females to meet the diagnostic threshold, their observable characteristics must be exaggerated.⁸ An understanding of the nuanced presentations between sexes^b and genders would be beneficial for equitable diagnosis.¹⁴

There is also the possibility of camouflaging (also termed “masking” and “social mimicry”) during the assessment, which could lead to misdiagnosis or missed diagnoses.¹⁵ Camouflaging, which refers to behaviors or strategies that hide or mask an individual’s autism from others,¹⁶ has been suggested to be a component of the female presentation of autism and is associated with missed or late diagnoses.^6,17,18 A systematic review found substantial evidence for gender differences in camouflaging with females reporting camouflaging more than males.¹⁹ Others have suggested that both men and women engage in camouflaging but for distinct reasons. For example, women were more likely to camouflage to serve a conventional purpose in a workplace or educational setting than males, whereas males and females used camouflaging to ease social interactions to a similar extent.²⁰ If females are more likely to use camouflaging,^13,15 autistic females may not come to the attention of clinicians and may fail to meet diagnostic thresholds. For example, Milner et al.²¹ found self-reported camouflaging scores were related to age at diagnosis, and that this relationship was stronger in female participants compared with males. Lockwood Estrin, Milner, Spain, Happé, and Colvert⁸ suggest that a tool examining aspects of camouflaging may be beneficial for diagnostic assessments.

Diagnoses may also be influenced by male bias in the diagnostic criteria, assessment process, and clinician knowledge.⁸ Giarelli, Wiggins, Rice, et al.²² discussed interpreting bias in autism assessments, where the observed experiences differ from the expected behaviors dependent on sex. Cridland, Jones, Captuti, and Magee²³ highlighted reports of health care professionals being reluctant to diagnose a female as autistic and a lack of awareness of autism in females due to a perceived higher incidence of autism in males. Freeman and Grigoriadis²⁴ found that in Australia, clinicians are not consistently receiving training in assessing females for autism and lack confidence in this area. In addition, Cumin, Palaez, and Mottron³ reported that experienced clinicians found it difficult to differentiate autism from trauma and borderline personality disorder. These clinicians used individually developed techniques to reveal signs indicative of autism in women that were not captured by standardized instruments. These factors may contribute to the health inequalities experienced by autistic individuals assigned female at birth that stem from multiple layers of these barriers.²⁵

The National Institute for Health and Care Excellence (NICE) provides guidance for clinicians in the UK on conducting autism diagnostic assessments.^26,27 The guidelines outline questions to ask, methods to use, and differential diagnoses to consider. There is no specific guidance provided regarding gender or sex differences in the presentation or assessment of autism. Harper, Smith, Happé et al.²⁸ highlighted the need for health and social care strategies to address specific challenges women and girls face in autism diagnostic pathways. However, this is not currently reflected in the national framework for autism assessment pathways.²⁹

Changes or adaptations to diagnostic assessments have been suggested by those in the field. For example, Lai and Szatmari³⁰ suggest that a good awareness of sex and gender impacts is essential, and that camouflaging should be explored during an assessment. The space in which assessments take place has also been identified as important to allow individuals to “unmask.”³¹ Following a narrative review, Cook, Hull, and Mandy³² state that guidelines do not need to be overhauled but rather fine-tuned. They recommend that clinicians need to be aware of the diagnostic bias against autistic women and girls as well as the possibility of camouflaging and diagnostic overshadowing. The authors also recommend that clinicians assess for camouflaging and be flexible when looking for behavioral exemplars of diagnostic features, considering the influences of sex and/or gender on how autism manifests and is experienced.

This study aims to add to the recommendations of Cook et al.³² and Lai et al.²⁵ by systematically exploring the recommendations of experienced clinicians for conducting diagnostic assessments with females using the Delphi method. In this study, we consider cis-gender females and hereafter use the terms “female,” “women,” and “girls” for conciseness. We recognize that gender exists on a spectrum, with a diverse range of identities going beyond the traditional male–female binary. In addition, we acknowledge that autistic individuals are more likely to identify as gender diverse, further highlighting the complexity of the intersection between gender and autism.^33,34 We hope that our research serves as a stepping stone for further exploration into sex and gender differences in autism and contributes to making autism assessments more inclusive and reflective of the diverse experiences of all autistic individuals in the future.

Methodology

Design

Between March and June 2024, a three-round Delphi study was conducted among a panel of professionals with experience assessing females for autism in the United Kingdom. Delphi studies are conducted to gather the views of experts to establish consensus about what best practice “should or could” constitute,³⁵ especially where there is contradictory or insufficient evidence.³⁶ Delphi studies use an iterative group-based, mixed-methods approach used for gathering expert opinions on a complex topic.³⁵ A panel of experts answer questionnaires, and their responses are summarized and presented back to the panel. The process is repeated for multiple rounds, allowing participants to reflect on and revise their opinions based on the group’s responses, to meet a consensus. Due to the scarcity of knowledge about what adaptations clinicians are making and would recommend in autism diagnostic assessments, the Delphi technique was considered the most appropriate method to answer the research question.

Research aims

This study aims to gather consensus on what clinicians would recommend for diagnostic assessments in females to improve diagnostic validity.

Procedure and analysis

The procedure was informed by Sekayi and Kennedy’s³⁷ guidance for qualitative Delphi methodologies and the Conducting and Reporting Delphi Studies (CREDES) guidance.³⁶ Details of the analysis are provided alongside the procedure for clarity and conciseness. When reflecting on the positionality of the research team, all authors identify as White females. Three authors have received the appropriate training in the United Kingdom and have conducted autism assessments for many years, including assessing females. We acknowledge that our positionality may have influenced our methodological choices and our decision to focus on cis-gendered females in this project.

Survey development

To inform the Delphi study, four autistic women engaged in a focus group to gather insight into what they feel are important considerations for the diagnostic process. The views of autistic women may differ from those of experts in relation to what diagnostic considerations are relevant and are equally important as the views of clinicians. The focus group participants were asked questions about their experience of the autism assessment, such as what they wanted the assessor to know about them and what questions they thought were important to be asked, as well as a question about what features of autism, specific to women, were or should have been assessed. The focus group transcript was analyzed by two members of the research team, using Braun and Clarke’s method of thematic analysis.³⁸ More details on the methodology and findings of the focus group can be found in Trundle et al.³⁹

In addition, a brief literature search was conducted using search terms relevant to the project aims (e.g., “autistic women,” “autistic females,” and “diagnostic assessments”). Databases searched included Ovid, EMBASE, and Google Scholar. Themes from the resulting relevant literature were informally explored alongside themes from the focus group jointly with all team members. Key themes were transformed into questions that would capture the experts’ experience in that area. Some questions were identified by the authors through our clinical experience, insight, and consideration of the project aims. This produced open-ended questions that related to four main categories: the assessment context, clinician attributes, assessment tools, and questions asked during the assessment (see Supplementary Data).

For each Delphi round, the participants were emailed a link to the survey, which was created on Microsoft Forms.

Delphi surveys

Round 1 used 18 open-ended questions. Data on participant gender, age, professional discipline, and training experience were also collected. The data collection period lasted for 4 weeks, with a reminder email sent to participants after 2 weeks. Completion of the Round 1 questionnaire took participants on average 51 minutes (SD = 32.29). Qualitative responses gathered in Round 1 were analyzed using Sekayi and Kennedy’s³⁷ methodology, which involves coding and grouping statements. There were no predefined grouping criteria. Two members of the research team completed independent coding of the results, and a third member of the team coded a random sample of responses to explore inter-rater reliability. A consensus approach was used to iteratively refine the themes. The codes were then discussed with the wider research team. The themes were reviewed and refined, and 28 statements were produced.

In Round 2, the 28 statements were presented to participants. Participants were asked to rate how much they agreed with the statements on a four-point Likert scale. Participants were given the opportunity to respond, with their reasons, in a free text response area. The data collection period lasted for 3 weeks, with a reminder email sent to participants after 2 weeks. Completion of the Round 2 questionnaire took on average 8 minutes (SD = 4.68). The percentage of responses on each point on the scale was calculated for each item.

In Round 3, participants were re-presented with all items and agreement distributions from Round 2. Participants were asked to rate how much they agreed with the statements on a four-point Likert scale with an additional free text response area. The data collection period lasted for 3 weeks, with a reminder email sent to participants after 2 weeks. Completion of the Round 3 questionnaire took an average of 15 minutes (SD = 23.48).

To examine consensus in Rounds 2 and 3, percentage of agreement was calculated for each point on the scale. Based on the guidelines by Langlands and colleagues,⁴⁰ statements rated with “agree” or “strongly agree” by ≥80% of participants were considered an integral component of diagnostic assessments in females. Stability between rounds was examined through descriptive statistics; as the participants in each round were not matched, statistical analysis of stability was not possible.

Participants

Participants, referred to as experts, were clinicians with experience diagnosing autism in females. An expert can be defined as “a group of knowledgeable people: Those who can provide relevant input to the process, have the highest authority possible, and who are committed and interested” (p. 33).⁴¹ We explored previous guidance/literature to develop the expert panel inclusion criteria, which highlighted variability in definitions of experts in this context. For example, Cumin et al.³ defined an expert as someone who had provided 250 diagnoses, practiced for at least 5 years and assessed at least 100 women, whereas Jordan⁴² recruited participants who had 3 or more years of professional experience conducting autism assessments and had conducted at least 20 assessments in the past 3 years. Wigham, Ingham, Le Couteur et al.⁴³ took a broader approach, recruiting clinicians conducting adult autism diagnostic assessments in the United Kingdom.

We felt it was important that the inclusion criteria encapsulated the need to be an appropriately trained, qualified clinician, practicing in the United Kingdom, who had sufficient experience in assessing for autism. This may include psychiatrists, speech and language therapists, nurses, and psychologists.²⁶ Therefore, in this study, an expert is defined as someone who: •

Is a registered health care professional who has received appropriate training to conduct autism diagnostic assessments AND

•

Has 12 months minimum experience of diagnosing autism through formal assessment AND

•

Has assessed for and diagnosed autism in females AND

•

Is practicing in the United Kingdom.

A sample size of at least 30 participants was sought, with a minimum number required being 15.⁴⁴ The study used a volunteer and snowball sample. We collated a list of clinicians based in the United Kingdom, who were known to practice within diagnostic services or those who had contributed to research protocols/journal articles pertaining to diagnostic services or assessments for autism. This list was developed by searching systematically through relevant literature, conducting an online search for diagnostic services in the United Kingdom, and asking colleagues if they were aware of clinicians working in the field. Participants were also recruited via word of mouth and professional networks, adverts within health care organizations, and online advertisements on professional sites such as LinkedIn. During the recruitment phase of the Delphi, 34 participants volunteered and consented to participate in the study.

Of the 34 volunteers, 24 completed Round 1 (67.6% retention rate), 18 completed Round 2 (52.9% retention rate), and 16 completed Round 3 (47.1% retention rate). Demographic data from Round 1 are presented in Table 1. Years of experience in autism assessments ranged from 2 years to 24 years, with a mean of 11.1 years of experience. The panel described completing autism assessments in the following settings: NHS outpatient services, private diagnostic services, inpatient settings including forensic inpatient settings, university clinics, and local mental health teams. They assessed adults (N = 21), children and young people (N = 14), and individuals with intellectual disabilities (N = 8). Clinicians reported that their knowledge of autism in females had developed through specialist autism courses, clinical training, independent research/reading, personal experience, and clinical supervision.

Table 1.

Round 1 Participants Demographic Information

Demographic information ^a	Round 1 (n = 24)	Round 2 (n = 18)	Round 3 (n = 16)
Gender
Female	23	17	15
Male	1	1	1
Nonbinary	0	0	0
Other	0	0	0
Ethnicity
Arab	0	0	0
Asian or Asian British	2	1	1
Black, Black British, Caribbean, or African	0	0	0
Mixed or multiple ethnicities	2	1	1
White or White British, Irish, Scottish, Northern Irish, or Welsh	19	16	13
Other	1	0	1
Age
18–24	0	0	0
25–34	3	1	1
35–44	12	6	6
45–54	7	8	7
55–64	1	1	1
65+	1	2	1
Professional discipline
Psychologist (including clinical psychologist)	16	10	10
Nurse	2	2	2
Psychiatrist	2	3	2
Speech and language therapist	2	2	1
Occupational therapist	2	1	1

Discrepancies between rounds are felt to be due to responder error and/or the occurrence of birthdays between the response periods.

Ethical considerations

Ethical approval was obtained from St. Andrews Healthcare Research Ethics Committee. All participation was voluntary. Participants provided written consent after being provided with the study information and were made aware of their right to withdraw from the study at any point without explanation. Participants provided email addresses so that they could be sent the study links. These were stored in a password-protected file on a secure laptop. No personal identifying information was collected in the raw data; thus, all data were anonymous.

Results

Round 1

Thematic analysis identified five overarching themes, each with several subthemes (Table 2). There were additional codes that related more broadly to autism assessments (across sexes and genders). For example, personal attributes of clinicians were discussed, including being able to display kindness and compassion and clinicians having sufficient skills to understand scoring criteria, administer psychometric assessments, and draft reports. For succinctness and clarity, these were not included in the generation of statements for Round 2. Twenty-eight statements were presented in Round 2.

Table 2.

Themes and Codes Identified Following Round 1

Main theme	Subthemes	Associated quote	Statements for Round 2
Attributes of the assessor	Knowledge and expertise • Knowledge of female autism phenotype • Understanding of masking	“I tend to prompt the informant to go into more detail with female patients to ensure we get the full extent of the information. For example, girls can have special interests that look quite ‘normal’ from the outside (horses, celebrities, make up etc.) but have an unusual quality to them (e.g., collecting lists of horse breeds and colours or categorising make up products in specific ways).”	1. A strong knowledge of the research field connected to the autism presentation in females is necessary. 2. Clinical/psychological knowledge, particularly in regard to differential/comorbid diagnosis of psychological issues such as attachment, or other mental disorders, such as trauma and/or personality disorder is required. 3. A strong knowledge of other developmental conditions is necessary. 4. An awareness of masking is necessary. 5. An awareness of the potential for male bias in the assessment tools used is required.
Clinical/psychological knowledge • Understanding of mental health difficulties, trauma, personality, and attachment	“You need to be familiar with a wide range of mental health conditions and with other neurodevelopmental conditions and be able to differentiate between these and traits of autism.”
Postclinical qualification education and learning	Reading and research	“Ability to engage critically with the narratives around women and autism in the scientific literature and more broadly in societal narratives.”	6. Autism assessors must undertake their own reading of research/articles on autism within women and girls. 7. Autism assessors must seek out CPD on autism in women and girls. 8. Autism assessors must receive training from autistic women. 9. New assessors need to shadow assessments of females.
	Continued professional development into autism in women and girls	“Subsequent CPD^a in female presentation, supervision around this.”
	Training from autistic women	“Training from people with lived experience would be ideal”
Shadowing of assessments	“The most useful training is in shadowing experienced clinicians and multidisciplinary team discussions.”
Diagnostic tools	Observational measures (e.g., ADOS)	“Not a huge fan of the ADOS⁴⁵ but it’s there if needed.”	10. Screening measures (e.g., AQ-10,⁴⁶ RAADS-14,⁴⁷ AQ,⁴⁸ EQ,⁴⁹ CAST)⁵⁰can be useful for screening autism indicators in females. 11. Care should be taken when interpreting screening tools due to the male bias that may be present. 12. Screening measures are useful for providing narrative information, above and beyond the cutoff scores. 13. It is essential to use screening tools for differential diagnoses, such as ADHD, depression, and anxiety. 14. It is essential to use screening tools for trauma/adverse childhood experiences as part of the assessment. 15. It is important to use measures of empathy as part of the assessment. 16. It is important to use measures of camouflaging as part of the assessment. 17. The ADOS-2 is not always helpful when assessing females.
Screening tools • For autistic traits • Male bias may be present • Differential diagnoses • Camouflaging • Empathy	“I like to screen for autism (SRS-2⁵¹ for example), masking (CAT-Q),⁵² other conditions such as OCD, ADHD, mental health difficulties and sensory sensitivities/differences.”“Some tools are out of date, contain gender bias, or are designed for use with children, so they have to be used with caution and may need adapting.”
Questions asked	Prompts around social communication	“Quality of social relationships-being on the outskirts of friendships or intense relationships, relying on others to maintain, using others to enable to engage in social activities”	18. More prompts/questions should be asked when assessing females to explore more subtle differences in social communication skills (e.g., about friendships, nonverbal communication). 19. More prompts/questions should be asked when assessing females to explore more subtle differences in restrictive and repetitive patterns of behaviors and interests (e.g., stimming, daily routines). 20. Questions about repetitive and intense interests should focus on the quality/intensity of the interest, rather than the topic itself. 21. It is important to ask questions about masking, mimicking, and mirroring others’ behaviors. 22. It is important to ask questions about experiences in different relationships (e.g., family, friends, colleagues). 23. It is important to ask questions about the experience of menstrual cycles, periods, and menopause. 24. It is important to ask questions about experiences of sexual relationships (e.g., consent, intimacy, vulnerability).
	Prompts around restrictive and repetitive patterns of behaviors and interests	“I ask lots of questions about areas of interest to see what form these take. It can be hard at first glance to see if an interest is intense, so you need to explore the impact of this, what form it takes, the onset, duration etc. Most of the other standard questions work well for females and males in my opinion—this is probably the section that I modify least—although I do go in-depth about sensory issues as these can be subtle for females.”
	Prompts about masking	“Questions around mirroring behaviours and repeating socially appropriate phrases in order to fit in, and superficial social communication used to mask a lack of understanding.”
	Prompts about relationships • Including sexual relationships	“I ask for lots of examples to help me better understand their experiences. I ask about the impact of these difficulties on different relationships and in different life domains (e.g., family, intimate, social, occupational, health).”
Gender-specific prompts • Periods • Sex • Menopause	“There are some specific questions about menstrual cycle and ability to manage that as a younger person.”
Sources of information	Professional collaboration • Consensus meetings • Supervision • Panel approach • Assess for camouflaging	“To discuss complex cases, to check things out when unsure, to enhance knowledge and for support to guard against burnout.”	25. Multiple informants and sources from different areas of an individual’s life and from different time-points in their life should be used. 26. It is important to consider the possibility of masking when interpreting information from informants. 27. Multidisciplinary meetings, supervision, and consensus are essential to making the diagnosis of autism in females. 28. Multidisciplinary discussions should be used to discuss the possibility of masking/camouflaging during the assessment.
Informants • Multiple informants • Possibility of camouflaging	“Use of informants can differ among adult women. While parents are still spoke to, other informants appear to be considered as having utility within the process.”

Continued Professional Development.

ADHD, attention-deficit/hyperactivity disorder; OCD, Obsessive Compulsive Disorder; ADOS, Autism Diagnostic Observation Schedule; AQ-10, Autism Quotient 10; RAADS-14, Ritvo Autism Asperger Diagnostic Scale Revised; AQ, Autism Quotient; EQ, Empathy Quotient; CAST, Childhood Asperger Syndrome Test; SRS-2, Social Responsiveness Scale; CAT-Q, Camouflaging Autistic Traits Questionnaire.

Round 2

In Round 2, consensus was achieved for 21 of 28 items (Table 3). Experts generally agreed on the attributes of the assessor, expectations of postclinical qualification education and learning, and questions to be asked. There was less agreement overall about the sources of information and diagnostic tools to be used.

Table 3.

Scores for Round 2

Item	Statement	Strongly disagree (%)	Disagree (%)	Agree(%)	Strongly agree (%)	Consensus (%; n = 18)
Attributes of the assessor
1	A strong knowledge of the research field connected to the autism presentation in females is necessary	0	0	55.6	44.4	100
2	Clinical/psychological knowledge, particularly in regard to differential/comorbid diagnosis of psychological issues such as attachment, or other mental disorders, such as trauma and/or personality disorder is required	0	0	27.8	72.7	100
3	A strong knowledge of other developmental conditions is necessary	0	0	44.4	55.6	100
4	An awareness of masking is necessary	0	0	22.2	77.8	100
5	An awareness of the potential for male bias in the assessment tools used is required	0	0	16.7	83.3	100
Postclinical qualification education and learning
6	Autism assessors must undertake their own reading of research/articles on autism within women and girls	0	0	16.7	83.3	100
7	Autism assessors must seek out CPD on autism in women and girls	0	0	33.3	66.7	100
8	Autism assessors must receive training from autistic women	0	11.1	61.1	27.8	88.9
9	New assessors need to shadow assessments of females	0	5.6	33.3	61.1	94.4
Diagnostic tools
10	Screening measures (e.g., AQ-10, RAADS-14, AQ, EQ, CAST) can be useful for screening autism indicators in females	0	16.7	66.7	16.7	83.3
11	Care should be taken when interpreting screening tools due to the male bias that may be present	0	0	27.8	72.2	100
12	Screening measures are useful for providing narrative information, above and beyond the cutoff scores	0	11.1	55.6	33.3	88.9
13	It is essential to use screening tools for differential diagnoses, such as ADHD, depression, and anxiety	0	33.3	50	16.7	66.7
14	It is essential to use screening tools for trauma/adverse childhood experiences as part of the assessment	0	38.9	50	11.1	61.1
15	It is important to use measures of empathy as part of the assessment	0	44.4	44.4	11.1	55.6
16	It is important to use measures of camouflaging as part of the assessment	0	16.7	44.4	38.9	83.3
17	The ADOS-2 is not always helpful when assessing females	5.6	16.7	44.4	33.3	77.7
Questions asked
18	More prompts/questions should be asked when assessing females to explore more subtle differences in social communication skills (e.g., about friendships, nonverbal communication)	0	0	27.8	72.7	100
19	More prompts/questions should be asked when assessing females to explore more subtle differences in restrictive and repetitive patterns of behaviors and interests (e.g., stimming, daily routines)	0	0	38.9	61.1	100
20	Questions about repetitive and intense interests should focus on the quality/intensity of the interest rather than the topic itself	0	0	33.3	66.7	100
21	It is important to ask questions about masking, mimicking, and mirroring others’ behaviors	0	0	22.2	77.8	100
22	It is important to ask questions about experiences in different relationships (e.g., family, friends, colleagues)	0	0	27.8	72.2	100
23	It is important to ask questions about the experience of menstrual cycles, periods, and menopause	0	22.2	55.6	22.2	77.8
24	It is important to ask questions about experiences of sexual relationships (e.g., consent, intimacy, vulnerability)	0	33.3	55.6	11.1	66.7
Sources of information
25	Multiple informants and sources from different areas of an individual’s life and from different time-points in their life should be used	0	16.7	61.1	22.2	83.3
26	It is important to consider the possibility of masking when interpreting information from informants	0	0	27.8	72.2	100
27	Multidisciplinary meetings, supervision, and consensus are essential to making the diagnosis of autism in females	5.6	16.7	55.6	22.2	77.7
28	Multidisciplinary discussions should be used to discuss the possibility of masking/camouflaging during the assessment	0	16.7	44.4	38.9	83.3

Qualitative responses were also provided by some experts. In terms of the attributes of the assessor, one expert wanted clinicians to understand the impact of hormones on mental health and another wanted clinicians to have knowledge of how autism relates to other neurodevelopmental conditions. Two experts stated that the attributes described should be present for all assessments not just female assessments.

In relation to clinician training, three experts commented on the recommendation for clinicians to have training by autistic women. Two thought this would be helpful but not essential, and a third elaborated that expert-by-experience training should be facilitated by several autistic people to provide a range of experiences.

There were several comments about recommended screening tools. Experts reflected on whether screening tools were useful to the assessment more generally and noted that it was dependent on the validity and interpretation of the screening tool. Experts also emphasized the need to not use the Autism Diagnostic Observation Schedule (ADOS-2)⁴⁵ in isolation.

Three experts commented that the recommended questions to ask may also be helpful to ask males. Two experts were not sure what questions about periods and sexual relationships would add to the diagnostic assessment.

Round 3

In Round 3, 18 items remained at the consensus level of 80% (Table 4). There was a trend observed that items reaching 100% agreement in Round 2 maintained 100% agreement in Round 3. Four items that were rated above 80% agreement in Round 2 were below the agreement threshold in Round 3. These were

Table 4.

Scores for Round 3

Item	Statement	Strongly disagree (%)	Disagree (%)	Agree(%)	Strongly agree (%)	Consensus (%; n = 16)
Attributes of the assessor
1	A strong knowledge of the research field connected to the autism presentation in females is necessary	0	0	56.3	43.8	100
2	Clinical/psychological knowledge, particularly in regard to differential/comorbid diagnosis of psychological issues such as attachment, or other mental disorders, such as trauma and/or personality disorder is required	0	0	18.8	81.3	100
3	A strong knowledge of other developmental conditions is necessary	0	0	25	75	100
4	An awareness of masking is necessary	0	0	25	75	100
5	An awareness of the potential for male bias in the assessment tools used is required	0	0	25	75	100
Postclinical qualification education and learning
6	Autism assessors must undertake their own reading of research/articles on autism within women and girls	0	0	43.8	56.3	100
7	Autism assessors must seek out CPD on autism in women and girls	0	0	62.5	37.5	100
8	Autism assessors must receive training from autistic women	0	31.3	62.5	6.3	68.7
9	New assessors need to shadow assessments of females	0	6.3	37.5	56.3	93.7
Diagnostic tools
10	Screening measures (e.g., AQ-10, RAADS-14, AQ, EQ, CAST) can be useful for screening autism indicators in females	0	6.3	93.8	0	93.7
11	Care should be taken when interpreting screening tools due to the male bias that may be present	0	0	31.3	68.8	100
12	Screening measures are useful for providing narrative information, above and beyond the cutoff scores	0	12.5	68.8	18.8	87.5
13	It is essential to use screening tools for differential diagnoses, such as ADHD, depression, and anxiety	0	62.5	37.5	0	37.5
14	It is essential to use screening tools for trauma/adverse childhood experiences as part of the assessment	0	62.5	37.5	0	37.5
15	It is important to use measures of empathy as part of the assessment	12.5	68.8	18.8	0	18.7
16	It is important to use measures of camouflaging as part of the assessment	0	31.3	37.5	31.3	68.7
17	The ADOS-2 is not always helpful when assessing females	0	12.5	62.5	25	87.5
Questions asked
18	More prompts/questions should be asked when assessing females to explore more subtle differences in social communication skills (e.g., about friendships, nonverbal communication)	0	0	50	50	100
19	More prompts/questions should be asked when assessing females to explore more subtle differences in restrictive and repetitive patterns of behaviors and interests (e.g., stimming, daily routines)	0	0	50	50	100
20	Questions about repetitive and intense interests should focus on the quality/intensity of the interest rather than the topic itself	0	0	37.5	62.5	100
21	It is important to ask questions about masking, mimicking, and mirroring others’ behaviors	0	0	31.3	68.8	100
22	It is important to ask questions about experiences in different relationships (e.g., family, friends, colleagues)	0	0	43.8	56.3	100
23	It is important to ask questions about the experience of menstrual cycles, periods, and menopause	0	31.3	56.3	12.5	68.7
24	It is important to ask questions about experiences of sexual relationships (e.g., consent, intimacy, vulnerability)	0	31.3	68.8	0	68.7
Sources of information
25	Multiple informants and sources from different areas of an individual’s life and from different time-points in their life should be used	0	37.5	50	12.5	62.5
26	It is important to consider the possibility of masking when interpreting information from informants	0	0	31.3	68.8	100
27	Multidisciplinary meetings, supervision, and consensus are essential to making the diagnosis of autism in females	0	43.8	50	6.3	56.2
28	Multidisciplinary discussions should be used to discuss the possibility of masking/camouflaging during the assessment	0	37.5	50	12.5	62.5

•

Autism assessors must receive training from autistic women.

•

It is important to use measures of camouflaging as part of the assessment.

•

Multiple informants and sources from different areas of an individual’s life and from different time-points in their life should be used.

•

Multidisciplinary discussions should be used to discuss the possibility of masking/camouflaging during the assessment.

There was also a trend for items that were below consensus in Round 2: consensus tended to further decrease in Round 3. One item did not reach 80% agreement in Round 2 but increased in Round 3 to exceed 80%. This was the statement: the ADOS-2 is not always helpful when assessing females.

Experts provided comments to accompany their ratings. Relating to the attributes of the assessor, like in Round 2, one expert stated it would be helpful for clinicians to understand hormones. Another expert commented that continued professional development and learning among teams are essential. As in Round 2, experts continued to query the usefulness and practicality of screening measures. In addition, one expert commented that the recommended questions to be asked should also be asked of males. Finally, one expert stated that having informants was not essential as adults are able to describe their own experiences.

In summary, experts agreed that a strong knowledge of a) research pertaining to autism in women; b) possible differential/co-occurring diagnoses and other relevant psychological experiences (e.g., attachment, trauma); c) other developmental conditions; and d) masking is necessary for autism assessments with females. It was recommended that autism assessors undertake their own reading and relevant continued professional development exploring autism in women, and new assessors should shadow assessments with women. In terms of tools to assist in an assessment, it was agreed that the ADOS-2 was not always helpful when assessing females for autism, whilst screening measures (e.g., Autism Quotient, Empathy Quotient, and Ritvo Autism Aspergers Diagnostic Scale Revised) were felt to be useful, especially for providing narrative information, above and beyond the cut off scores. That being said, experts agreed that an awareness of the potential for male bias in assessment tools is required, and thus care should be taken when interpreting screening questionnaires. Experts also agreed that more prompts/questions should be asked when assessing females to explore more subtle differences in social communication skills (e.g., about friendships, non-verbal communication) and repetitive patterns of behaviours and interests (e.g., stimming, daily routines). Questions about repetitive and intense interests should focus on the quality/intensity of the interest, rather than the topic itself. Experts also agreed that it is important to asking about experiences in different relationships and about masking behaviours. Masking should also be considered when interpreting informant information.

Discussion

Clinicians have reported feeling less confident in recognizing, screening, and diagnosing autism in females, with many wanting additional training in this area.^24,53 This study used a Delphi technique to gather consensus on clinician recommendations for diagnostic assessments in cis-gender females. The expert panel consisted of registered health care professionals practicing in the United Kingdom who had assessed and diagnosed autism in females. Of the recruited experts, 47.1% completed all three rounds of the Delphi study. Statements reaching consensus fell within five categories: attributes of the assessor, postclinical qualification and learning, diagnostic tools, questions asked, and sources of information.

Attributes of the assessor

Experts in this study agreed that assessing clinicians need to have: a strong knowledge of the research into the female presentation of autism and of other developmental conditions; an awareness of masking and the potential for male bias in the assessment tools; and clinical or psychological knowledge of differential explanations or co-occurring conditions, such as attachment, trauma, and mental health difficulties. Such knowledge and awareness should assist in differentiating between autism and personality disorder, with which autistic females are often misdiagnosed.¹² This would also assist in generating a holistic understanding of the person.

Postclinical qualification education and learning

Experts described a myriad of ways in which they developed their knowledge and practice on autism and females. This included independent research and attending courses. Research has highlighted a lack of autism focus within professional clinical training courses, leading to a lack of confidence and consistency in clinical practice.⁵³ Some experts emphasized the importance of ongoing learning, including training from autistic women, whereas others felt this was not necessary.

Diagnostic tools

Experts indicated that screening tools for autistic traits can be useful during diagnostic assessments, and that care should be taken when interpreting these measures due to the potential for male bias. Screening measures that can provide narrative information would be useful within diagnostic decision-making, alongside the quantitative scores. NICE²⁷ recommended using the Autism Quotient⁴⁶ and the Ritvo Autism Asperger Diagnostic Scale Revised,⁴⁷ although this recommendation is not sex- or gender-specific.

The use of screening tools for cooccurring or differential conditions such as anxiety, depression, attention-deficit/hyperactivity disorder (ADHD), and traumatic experiences was not endorsed by most of the experts. Qualitative responses suggested this may relate to the validity of specific tools. NICE²⁷ recommended clinicians assess for mental health conditions, although they provide no specific guidance on how to do this. It may be that relevant information is collected as part of the developmental history and narratively synthesized.

Most experts indicated that measuring empathy was not an important part of the assessment. Difficulties with empathy are not specified with diagnostic criteria of autism, and contemporary understandings of empathy in autism suggest a more nuanced profile of skills⁵⁴ than would be captured on one-dimensional measures, such as the Empathy Quotient.⁵⁵ Thus, our findings may reflect changes to the wider understanding of empathy in autism.

Experts agreed that the ADOS-2⁴⁵ was not always helpful in assessing females. This echoes the experiences of autistic women who felt the ADOS-2 was patronizing and limited in reliability.³⁹ Research indicates that ADOS-2 scores and profiles can differ between sexes, but this may not affect whether they meet the overall diagnostic criteria for autism.⁵⁶ It may therefore be the subjective experience of receiving an ADOS-2 that is a limitation, as opposed to the tool’s ability to identify signs of autism.

Masking was highlighted as a crucial factor to consider,⁵⁷ which was also emphasized by autistic women.³⁹ Despite 80% of experts agreeing that assessing for camouflaging is helpful in Round 2, in Round 3 consensus dropped below the threshold, with a third of experts indicating that measuring camouflaging is not an important part of the assessment. This contrasts with the suggestion made by Lockwood Estrin, et al.⁸ who suggested that a tool examining camouflaging would be beneficial for diagnostic assessments. Our findings may relate specifically to the use of screening tools, as consideration of camouflaging is supported by clinicians within later themes. Screening measures of camouflaging may lack construct validity, in that the tools capture anxiety-based behaviors, as opposed to autism-related behaviors.⁵⁸ Thus, alternative means of exploring experiences of camouflaging may be more useful.

Questions asked

Experts agreed that it is useful to ask questions about masking, mimicking, and mirroring. Qualitative information could provide a greater understanding of the client’s experiences of autism than a screening measure could.

Experts also agreed that it would be useful to ask additional and/or more probing questions when conducting autism assessments in females in comparison with assessments with males. This included exploring subtle differences in social communication skills, restrictive and repetitive patterns of behaviors and interests, and experiences in relationships. This could include how a person played with other children,⁴⁷ types of repetitive behaviors,⁵⁹ and the intensity of relationships.⁶⁰

Research into the experiences of menopause in autistic women has described higher levels of menopausal complaints and increased frequency of autistic burnout and meltdowns.^61,62 In contrast to the perspectives of autistic females,³⁹ many experts in this study indicated that asking questions pertaining to assigned-sex specific experiences (e.g., menstrual cycles and menopause) and experiences in sexual relationships would not be important within the diagnostic assessment.

Sources of information

Experts indicated that it was important to understand that a person may have camouflaged their difficulties during the appointments, potentially influencing the observational data. Some experts felt that the use of multidisciplinary meetings is important, including discussing the possibility of camouflaging, whereas others did not.

In summary, expert clinicians in this study identified several recommendations for autism diagnostic assessments with females. The findings of this study are in line with the recommendations provided by Cook et al.³² following their narrative review. This includes that clinicians should be aware of the possibility of male bias in the assessment tools, how the autism presentation differs between males and females, the possibility of masking, and an understanding that there may be differential or cooccurring conditions. In addition, both this study and that of Cook et al.³² recommend that clinicians should explore and probe for different behavioral descriptions of features of autism, considering sex differences, as well as examining camouflaging behaviors, and that there should be a focus on qualitative experience beyond the scores on screening tools. This study expands on Cook and colleagues’³² findings by making recommendations for postqualification education and learning and for specific prompts/questions to be asked about sex differences in presentation. Application of our findings may also contribute to reducing health inequalities experienced by autistic females, such as by promoting increased training and practitioner knowledge and an awareness of helpful measurement tools.²⁵ It is felt that many of the recommendations made here may be helpful for people of all sexes and genders.

Evaluation and drawbacks

Our study has drawbacks, which should be considered when interpreting the findings. First, attrition may have influenced the final statements by introducing responder bias. While the sample size exceeded the recommended minimum of 15 at all stages,⁴⁴ it is possible that agreement on certain statements may have differed if the sample size was consistent between Delphi rounds. Those with minority opinions may have dropped out, leading to an overestimation of consensus.⁶³ Furthermore, it appears that the recommendations generated in this study would be most appropriate for adolescents and adults rather than children. For instance, some of the recommended screening tools may not have been validated for use with children (e.g., the RAADS-14⁴⁷). It is possible that although some experts in this study reported working with children and young people, the findings reflect the wider bias in autistic women being diagnosed later in life.¹⁰ Thus, our study continues to highlight the late-diagnosis bias experienced in the female autistic population and emphasizes the need to explore recommendations for assessing females across the lifespan.

In Round 1, participants were asked whether they make adaptations or considerations when assessing autism in females from non-White British backgrounds. We felt this was important to include as those in global majority groups may experience disadvantages in obtaining an autism diagnosis.^64,65 Previous research has indicated that cultural differences within the autistic phenotype do occur, which clinicians should be aware of. For instance, a child’s limited eye contact with an adult is a sign of respect in Chinese culture,⁶⁶ and imaginative play is less common in the East than in the West.⁶⁷ During the ADOS-2,⁴⁵ a child’s birthday party scene is often imagined; however, what is typical regarding a child’s birthday party may vary across cultures and this is not reflected in the scoring matrix. This highlights the need for cultural awareness and adaptations within autism assessments. There is currently a dearth of literature about the experience of females from global majority backgrounds attending diagnostic assessments or receiving their autism diagnosis. For example, Diemer, Gerstein, and Regester⁶⁸ suggest that “autistic Black girls are effectively invisible in the literature.” Tromans, Chester, Kapugama, et al.⁶⁹ conducted a literature review and identified factors influencing this disparity including biases in the referral process, attribution of social and communication difficulties to an individual’s ethnic background, differing norms across cultures, and potential bias in diagnostic tools.

Our thematic analysis did not identify any overarching themes for this area. However, some participants stated that they did not assess non-White females differently to White females, one stated that they considered differences in eye contact and other cultural values, and one referenced the use of an interpreter. As a result, the subsequent rounds included no statements or consensus ratings relating to considerations in assessments for non-White females.

In addition, our participants were predominately White or White British, Irish, Scottish, Northern Irish, or Welsh (79.2%), female (95.8%), and worked as psychologists (66.7%). Data regarding the characteristics of autism assessors nationally are limited, however, there are data available around the composition of psychologists, which is similar to our sample.⁷⁰ While the effects of participant demographics were not tested against the results, limited variation in the sample could lead to biases. Future studies should actively engage with and seek participation from professionals from global majority backgrounds as well as professionals with experience facilitating assessments with such communities. Where studies are planning to use participants who have engaged in assessments, participants and research teams should include those from diverse cultural backgrounds.

Implications and recommendations

Although there is a developing evidence base describing the issues faced by women and girls within autism assessments, the extent to which this research is being integrated into clinical practice within autism diagnostic assessments is unclear. This Delphi study highlighted inconsistencies in sex considerations within the diagnostic process between clinicians. This finding is of concern given the high levels of underdiagnosis as reported in a recent population-based cohort study of primary care data. O’Nions, Petersen, Buckman, et al.⁷¹ highlighted inequalities in access to autism diagnostic assessments in the United Kingdom, with particular difficulties in certain geographical localities and older age groups. They concluded that there is a pressing need for better identification of autism in adults and better access to adult autism diagnostic services. Furthermore, research has suggested people have an 85% chance of being diagnosed as autistic in some centers compared with a 35% chance in others.⁷² As such, not only are there difficulties in accessing autism services, but there are also inconsistencies between services in terms of the approach to diagnosis, and ultimately, the diagnostic outcome. Barnard-Brak, Richman, and Almekdash⁷³ estimated that 39% more girls should be diagnosed with autism than actually were, by comparing clinic and community-based data sets to national population estimates for sex distribution. Therefore, it is likely that autistic girls and women are one of the groups most affected by inequalities in access to and outcomes from autism diagnostic assessments.

This may be compounded by a lack of clarity or consensus concerning terminology within existing research. Some authors separate sex and gender,⁵ some use sex and gender interchangeably,⁸ and others do not define their use of gendered terms (e.g., “female” and “male”),² making it challenging to delineate findings and recommendations by sex and/or gender. In our reports of other studies, we have specifically used the terminology used by each author when describing their findings. Looking ahead, it will be essential for researchers to clearly define the groups of interest, to allow for effective generalization and translation of findings into practice. This would also shine a light on underserved populations that are being missed.

Given the inconsistencies in practice highlighted by the Delphi and national data, it is evident that clinicians working within autism diagnostic services require guidance in the assessment of women and girls. This project aimed to take a step forward toward the development of such guidance. We sought the perspectives of autistic women³⁹ and clinicians operating in this field in the United Kingdom. It is hoped that assessing clinicians find these recommendations helpful in informing their practice and improving autism diagnostic assessments with females. This project provides both practical and feasible recommendations as well as the foundations for service development and quality improvement initiatives within diagnostic services. The next steps for us will be the development of formal guidance, which will combine the evidence base, perspectives of autistic women and girls, and the Delphi results. This guidance will build on the five themes highlighted within this research: attributes of the assessor, postclinical qualification education and learning, diagnostic tools, questions asked, and sources of information. It will be the subject of wide consultation with a range of stakeholders, aiming to improve the experience and accuracy of diagnosis for girls and women undergoing autism assessment, increase consistency across services, and support assessing clinicians.

Footnotes

Acknowledgments

The authors thank Dr. Deborah Morris, Dr. Elanor Webb, and Dr. Lucy Cygan for advice on the study design.

Authorship Confirmation Statement

G.T., V.C., and P.M. contributed to the study conception and design. All authors contributed to the material preparation, data collection, and analysis. The first draft of the article was written by G.T., and all authors commented on previous versions of the article. All authors read and approved the final article.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

No funding was received for conducting this study. The authors have no relevant financial or nonfinancial interests to disclose.

Supplementary Material

References

1. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders: DSM-5. American Psychiatric Publishing: Arlington, VA; 2013.

2. Zeidan

, Fombonne

, Scorah

, et al. Global prevalence of autism: A systematic review update. Autism Research. 2022;15(5):778–790; doi: 10.1002/aur.2696

3. Cumin

, Pelaez

, Mottron

. Positive and differential diagnosis of autism in verbal women of typical intelligence: A Delphi study. Autism. 2022;26(5):1153–1164; doi: 10.1177/13623613211042719

4. Head

, McGillivray

, Stokes

. Gender differences in emotionality and sociability in children with autism spectrum disorders. Molecular Autism. 2014;5(1):19; doi: 10.1186/2040-2392-5-19

5. Wood‐Downie

, Wong

, Kovshoff

, Cortese

, Hadwin

. Research review: A systematic review and meta‐analysis of sex/gender differences in social interaction and communication in autistic and nonautistic children and adolescents. J Child Psychol Psychiatry. 2021;62(8):922–936; doi: 10.1111/jcpp.13337

6. Lai

, Lombardo

, Auyeung

, Chakrabarti

, Baron-Cohen

. Sex/gender differences and autism: Setting the scene for future research. J Am Acad Child Adolesc Psychiatry. 2015;54(1):11–24; doi: 10.1016/j.jaac.2014.10.003

7. Edwards

, Wright

, Sargeant

, Cortese

, Wood‐Downie

. Research review: A systematic review and meta‐analysis of sex differences in narrow constructs of restricted and repetitive behaviours and interests in autistic children, adolescents, and adults. J Child Psychol Psychiatry. 2024;65(1):4–17; doi: 10.1111/jcpp.13855

8. Lockwood Estrin

, Milner

, Spain

, Happé

, Colvert

. Barriers to autism spectrum disorder diagnosis for young women and girls: A systematic review. Rev J Autism Dev Disord. 2021;8(4):454–470; doi: 10.1007/s40489-020-00225-8

9. Van Wijngaarden-Cremers

PJM

, van Eeten

, Groen

, et al. Gender and age differences in the core triad of impairments in autism spectrum disorders: A systematic review and meta-analysis. J Autism Dev Disord. 2014;44(3):627–635; doi: 10.1007/s10803-013-1913-9

10.

10. Loomes

, Hull

, Mandy

WPL

. What is the male-to-female ratio in autism spectrum disorder? A systematic review and meta-analysis. J Am Acad Child Adolesc Psychiatry. 2017;56(6):466–474; doi: 10.1016/j.jaac.2017.03.013

11.

11. Dworzynski

, Ronald

, Bolton

, Happé

. How different are girls and boys above and below the diagnostic threshold for autism spectrum disorders? J Am Acad Child Adolesc Psychiatry. 2012;51(8):788–797; doi: 10.1016/j.jaac.2012.05.018

12.

12. Lai

, Baron-Cohen

. Identifying the lost generation of adults with autism spectrum conditions. Lancet Psychiatry. 2015;2(11):1013–1027; doi: 10.1016/S2215-0366(15)00277-1

13.

13. Hull

, Lai

, Baron-Cohen

, et al. Gender differences in self-reported camouflaging in autistic and non-autistic adults. Autism. 2020;24(2):352–363; doi: 10.1177/1362361319864804

14.

14. Lai

, Lin

, Ameis

. Towards equitable diagnoses for autism and attention-deficit/hyperactivity disorder across sexes and genders. Current Opinion in Psychiatry. 2022;35(2):90–100; doi: 10.1097/YCO.0000000000000770

15.

15. Milner

, McIntosh

, Colvert

, Happé

. A qualitative exploration of the female experience of autism spectrum disorder (ASD). J Autism Dev Disord. 2019;49(6):2389–2402; doi: 10.1007/s10803-019-03906-4

16.

16. Hull

, Petrides

, Allison

, et al. “Putting on my best normal”: Social camouflaging in adults with autism spectrum conditions. J Autism Dev Disord. 2017;47(8):2519–2534; doi: 10.1007/s10803-017-3166-5

17.

17. Gould

, Ashton-Smith

. Missed diagnosis or misdiagnosis? Girls and women on the autism spectrum. Good Autism Practice (GAP). 2011;12(1):34–41.

18.

18. Lai

, Lombardo

, Ruigrok

ANV

, et al.; MRC AIMS Consortium. Quantifying and exploring camouflaging in men and women with autism. Autism. 2017;21(6):690–702; doi: 10.1177/1362361316671012

19.

19. Cook

, Hull

, Crane

, Mandy

. Camouflaging in autism: A systematic review. Clinical Psychology Review. 2021;89:102080; doi: 10.1016/j.cpr.2021.102080

20.

20. Cage

, Troxell-Whitman

. Understanding the reasons, contexts and costs of camouflaging for autistic adults. J Autism Dev Disord. 2019;49(5):1899–1911; doi: 10.1007/s10803-018-03878-x

21.

21. Milner

, Colvert

, Hull

, et al. Does camouflaging predict age at autism diagnosis? A comparison of autistic men and women. Autism Research. 2024;17(3):626–636; doi: 10.1002/aur.3059

22.

22. Giarelli

, Wiggins

, Rice

, et al. Sex differences in the evaluation and diagnosis of autism spectrum disorders among children. Disabil Health J. 2010;3(2):107–116; doi: 10.1016/j.dhjo.2009.07.001

23.

23. Cridland

, Jones

, Caputi

, Magee

. Being a girl in a boys’ world: Investigating the experiences of girls with autism spectrum disorders during adolescence. J Autism Dev Disord. 2014;44(6):1261–1274; doi: 10.1007/s10803-013-1985-6

24.

24. Freeman

, Grigoriadis

. A survey of assessment practices among health professionals diagnosing females with autism. Research in Developmental Disabilities. 2023;135(104445):104445; doi: 10.1016/j.ridd.2023.104445

25.

25. Lai

M-C

, Amestoy

, Bishop

, et al. Improving autism identification and support for individuals assigned female at birth: Clinical suggestions and research priorities. Lancet Child Adolesc Health. 2023;7(12):897–908; doi: 10.1016/S2352-4642(23)00221-3

26.

26. National Institute for Health and Care Excellence. Autism Spectrum Disorder in Under 19s: Recognition, Referral and Diagnosis. Available from: https://www.nice.org.uk/Guidance/CG128 [Last accessed: October 16, 2024].

27.

27. National Institute for Health and Care Excellence. Autism Spectrum Disorder in Adults: Diagnosis and Management. Available from: https://www.nice.org.uk/Guidance/CG142 [Last accessed: October 16, 2024].

28.

28. Harper

, Smith

, Happé

, et al. Autistica Action Briefing: Diagnosing Autistic Women and Girls. Autistica. Available from: https://www.autistica.org.uk/downloads/files/Autistica-Action-Briefing-Diagnosing-Autistic-Women-And-Girls.pdf [Last accessed: October 16, 2024].

29.

29. England

NHS

. A National Framework to Deliver Improved Outcomes in All-Age Autism Assessment Pathways: Guidance for Integrated Care Boards. Available from: https://www.england.nhs.uk/long-read/a-national-framework-to-deliver-improved-outcomes-in-all-age-autism-assessment-pathways-guidance-for-integrated-care-boards/ [Last accessed: October 16, 2024].

30.

30. Lai

, Szatmari

. Sex and gender impacts on the behavioural presentation and recognition of autism. Current Opinion in Psychiatry. 2020;33(2):117–123; doi: 10.1097/YCO.0000000000000575

31.

31. Murphy

, Flower

, Jellett

. Women seeking an autism diagnosis in Australia: A qualitative exploration of factors that help and hinder. Autism. 2023;27(3):808–821; doi: 10.1177/13623613221117911

32.

32. Cook

, Hull

, Mandy

. Improving diagnostic procedures in autism for girls and women: A narrative review. Neuropsychiatr Dis Treat. 2024;20:505–514; doi: 10.2147/NDT.S372723

33.

33. Corbett

, Muscatello

, Klemencic

, West

, Kim

, Strang

. Greater gender diversity among autistic children by self-report and parent-report. Autism. 2023;27(1):158–172; doi: 10.1177/13623613221085337

34.

34. Walsh

, Krabbendam

, Dewinter

, Begeer

. Brief report: Gender identity differences in autistic adults: Associations with perceptual and socio-cognitive profiles. J Autism Dev Disord. 2018;48(12):4070–4078; doi: 10.1007/s10803-018-3702-y

35.

35. Hsu

, Sandford

. The Delphi technique: Making sense of consensus. Practical Assessment, Research, and Evaluation. 2007;12(1):10; doi: 10.7275/pdz9-th90

36.

36. Jünger

, Payne

, Brine

, Radbruch

, Brearley

. Guidance on conducting and REporting DElphi Studies (CREDES) in palliative care: Recommendations based on a methodological systematic review. Palliative Medicine. 2017;31(8):684–706; doi: 10.1177/0269216317690685

37.

37. Sekayi

, Kennedy

. Qualitative Delphi method: A four round process with a worked example. TQR. 2017;22(10):2755–2763.

38.

38. Braun

, Clarke

. Using thematic analysis in psychology. Qualitative research in Psychology. 2006;3(2):77–101; doi: 10.1191/1478088706qp063oa

39.

39. Trundle

, Alexander

, Chester

. Put autistic women and girls at the heart: Exploring the experiences of autistic women during diagnostic assessments in the UK. AIA. 2025;11(2):100–113; doi: 10.1108/AIA-08-2024-0055

40.

40. Langlands

, Jorm

, Kelly

, Kitchener

. First aid for depression: A Delphi consensus study with consumers, carers and clinicians. J Affect Disord. 2008;105(1–3):157–165; doi: 10.1016/j.jad.2007.05.004

41.

41. Gutierrez

. Experimental techniques for information requirements analysis. Information & Management. 1989;16(1):31–43; doi: 10.1016/0378-7206(89)90025-6

42.

42. Jordan

. The Use of Clinical Judgment in Differentiating Symptoms of Autism Spectrum Disorder from those of Other Childhood Conditions: A Delphi Study. (Doctoral dissertation, University of Denver). https://digitalcommons.du.edu/cgi/viewcontent.cgi?article=2587&context=etd Accessed 16th October 2024.

43.

43. Wigham

, Ingham

, Le Couteur

, et al. A survey of autistic adults, relatives and clinical teams in the United Kingdom: And Delphi process consensus statements on optimal autism diagnostic assessment for adults. Autism. 2022;26(8):1959–1972; doi: 10.1177/13623613211073020

44.

44. Ludwig

. Predicting the future: Have you considered using the Delphi methodology? Journal of Extension. 1997;35(5).

45.

45. Lord

, Rutter

, DiLavore

, et al. Autism Diagnostic Observation Schedule, Second Edn. Western Psychological Services: Torrence, CA; 2012.

46.

46. Allison

, Auyeung

, Baron-Cohen

. Toward brief “red flags” for autism screening: The short autism spectrum quotient and the short quantitative checklist in 1,000 cases and 3,000 controls. J Am Acad Child Adolesc Psychiatry. 2012;51(2):202–212.e7; doi: 10.1016/j.jaac.2011.11.003

47.

47. Ritvo

, Ritvo

, Guthrie

, et al. The Ritvo Autism Asperger Diagnostic Scale-Revised (RAADS-R): A scale to assist the diagnosis of autism spectrum disorder in adults: An international validation study. J Autism Dev Disord. 2011;41(8):1076–1089; doi: 10.1007/s10803-010-1133-5

48.

48. Baron-Cohen

, Wheelwright

, Skinner

, Martin

, Clubley

. The autism spectrum quotient (AQ): Evidence from asperger syndrome/high-functioning autism, males and females, scientists and mathematicians. J Autism Dev Disord. 2001;31(1):5–17; doi: 10.1023/A:1005653411471

49.

49. Baron-Cohen

, Wheelwright

. The empathy quotient: An investigation of adults with Asperger syndrome or high functioning autism, and normal sex differences. J Autism Dev Disord. 2004;34(2):163–175; doi: 10.1023/B:JADD.0000022607.19833.00

50.

50. Scott

, Baron-Cohen

, Bolton

, Brayne

. The CAST (Childhood Asperger Syndrome Test): Preliminary development of a UK screen for mainstream primary-school- age children. Autism. 2002;6(1):9–31; doi: 10.1177/1362361302006001003

51.

51. Constantino

, Gruber

. The Social Responsiveness Scale Manual, Second Edition (SRS-2). Los Angeles, CA: Western Psychological Services; 2012.

52.

52. Hull

, Mandy

, Lai

, et al. Development and validation of the camouflaging autistic traits questionnaire (CAT-Q). J Autism Dev Disord. 2019;49(3):819–833; doi: 10.1007/s10803-018-3792-6

53.

53. Tromans

, Chester

, Kapugama

, Elliott

, Robertson

, Barrett

. The PAAFID project: Exploring the perspectives of autism in adult females among intellectual disability healthcare professionals. AIA. 2019;5(3):157–170; doi: 10.1108/AIA-09-2018-0033

54.

54. Shalev

, Warrier

, Greenberg

, et al. Reexamining empathy in autism: Empathic disequilibrium as a novel predictor of autism diagnosis and autistic traits. Autism Res. 2022;15(10):1917–1928; doi: 10.1002/aur.2794

55.

55. Allison

, Baron-Cohen

, Wheelwright

, Stone

, Muncer

. Psychometric analysis of the Empathy Quotient (EQ). Personality and Individual Differences. 2011;51(7):829–835; doi: 10.1016/j.paid.2011.07.005

56.

56. Rea

, Øien

, Shic

, Webb

, Ratto

. Sex differences on the ADOS-2. J Autism Dev Disord. 2023;53(7):2878–2890; doi: 10.1007/s10803-022-05566-3

57.

57. Dean

, Harwood

, Kasari

. The art of camouflage: Gender differences in the social behaviors of girls and boys with autism spectrum disorder. Autism. 2017;21(6):678–689; doi: 10.1177/1362361316671845

58.

58. Fombonne

. Camouflage and autism. J Child Psychol Psychiatry. 2020;61(7):735–738; doi: 10.1111/jcpp.13296

59.

59. Sutherland

, Hodge

, Bruck

, Costley

, Klieve

. Parent-reported differences between school-aged girls and boys on the autism spectrum. Autism. 2017;21(6):785–794; doi: 10.1177/1362361316668653

60.

60. Sedgewick

, Hill

, Pellicano

. ‘It’s different for girls’: Gender differences in the friendships and conflict of autistic and neurotypical adolescents. Autism. 2019;23(5):1119–1132; doi: 10.1177/1362361318794930

61.

61. Brady

, Jenkins

, Gamble-Turner

, Moseley

, Janse van Rensburg

, Matthews

. “A perfect storm”: Autistic experiences of menopause and midlife. Autism. 2024;28(6):1405–1418; doi: 10.1177/13623613241244548

62.

62. Kock

, Strydom

, O’Brady

, Tantam

. Autistic women’s experience of intimate relationships: The impact of an adult diagnosis. AIA. 2019;5(1):38–49; doi: 10.1108/AIA-09-2018-0035

63.

63. Gargon

, Crew

, Burnside

, Williamson

. Higher number of items associated with significantly lower response rates in COS Delphi surveys. J Clin Epidemiol. 2019;108:110–120; doi: 10.1016/j.jclinepi.2018.12.010

64.

64. Becerra

, von Ehrenstein

, Heck

, et al. Autism spectrum disorders and race, ethnicity, and nativity: A population-based study. Pediatrics. 2014;134(1):e63–e71; doi: 10.1542/peds.2013-3928

65.

65. Mandell

, Wiggins

, Carpenter

, et al. Racial/ethnic disparities in the identification of children with autism spectrum disorders. American J Public Health. 2009;99(3):493–498; doi: 10.2105/AJPH.2007.131243

66.

66. Hus

, Segal

. Challenges surrounding the diagnosis of autism in children. Neuropsychiatr Dis Treat. 2021;17:3509–3529; doi: 10.2147/NDT.S282569

67.

67. de Leeuw

, Happé

, Hoekstra

. A conceptual framework for understanding the cultural and contextual factors on autism across the globe. Autism Research. 2020;13(7):1029–1050; doi: 10.1002/aur.2276

68.

68. Diemer

, Gerstein

, Regester

. Autism presentation in female and Black populations: Examining the roles of identity, theory, and systemic inequalities. Autism. 2022;26(8):1931–1946; doi: 10.1177/13623613221113501

69.

69. Tromans

, Chester

, Gemegah

, et al. Autism identification across ethnic groups: A narrative review. AIA. 2021;7(3):241–255; doi: 10.1108/AIA-03-2020-0017

70.

70. Jameel

, Gin

, Lee-Carbon

, et al. The “Our Stories Project”: Understanding the needs, experiences, and challenges of trainee, aspiring and qualified clinical psychologists from minoritised backgrounds. Association of Clinical Psychologists. Available from: https://acpuk.org.uk/wp-content/uploads/2022/10/Our-Stories-Project-2.0-Final.pdf [Last accessed: January 29, 2025].

71.

71. O’Nions

, Petersen

, Buckman

, et al. Autism in England: Assessing underdiagnosis in a population-based cohort study of prospectively collected primary care data. The Lancet Regional Health–Europe. 2023;29; doi: 10.1016/j.lanepe.2023.100626

72.

72. Hill

. Some NHS Centres Twice as Likely to Diagnose Adults as Autistic, Study Finds. The Guardian. Available from: https://www.theguardian.com/society/2024/mar/04/some-nhs-centres-twice-as-likely-to-diagnose-adults-as-autistic-study-finds [Last accessed: October 16, 2024].

73.

73. Barnard-Brak

, Richman

, Almekdash

. How many girls are we missing in ASD? An examination from a clinic- and community-based sample. AIA. 2019;5(3):214–224; doi: 10.1108/AIA-11-2018-0048

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.01 MB

0.00 MB