A systematic review of higher education-based interventions to support the mental health and wellbeing of neurodivergent students

Abstract

Increasing numbers of neurodivergent students are engaging in higher education; however, support approaches vary within different institutions. Sometimes there are long waiting lists for specialised support, and most focus on academic adjustments, such as providing extra time in an assessment, rather than mental health and wellbeing. A systematic review, pre-registered on Prospero (CRD42024597980), was conducted to provide an overview of interventions supporting mental health and wellbeing of neurodivergent students in higher education. Ovid, Web of Science, and ERIC databases were searched in May 2025. Studies were included where the intervention aimed to improve mental health and/or wellbeing or improve the student experience, and the focus was on whether any strength-based approaches were used. Thirty-seven studies are included, conducted in seven countries. The Mixed Methods Appraisal Tool (MMAT) was used to assess the quality of included papers. Interventions varied widely and included: coaching, cognitive behavioural therapy, self-help, peer support, psychotherapy, counselling, mentoring, mindfulness, and neuro/bio feedback. The narrative synthesis demonstrates little evidence of strength-based approaches and found that neurodivergent students were rarely involved in designing the interventions. Most commonly, studies focused on attention deficit hyperactivity disorder (ADHD) (17 studies) or Autism (14 studies), with few interventions considering co-occurrence or other neurotypes.

Introduction

Neurodiversity can be described as a naturally occurring variation in how people think, process information, and interact with the world around them¹. It recognizes that neurological differences—such as autism or dyspraxia—are part of human diversity, rather than inherently pathological or disordered. However, defining neurodiversity can be complex, as interpretations vary depending on cultural, medical, and social perspectives². Some view it primarily through a medical lens, emphasising diagnosis and treatment, while others advocate for a social or rights-based approach that focuses on acceptance, inclusion, and accommodation^3,4. These differing viewpoints can make it challenging to arrive at a single, universally accepted definition⁵. The terms neurodivergent and neurodivergence were introduced by Asasumasu in 2000 as part of the shift in language becoming more inclusive⁶. As understanding of neurodiversity continues to evolve, so too do the conversations surrounding it, shaped by ongoing research, lived experiences, and shifting societal attitudes.

The number of neurodivergent students attending university is increasing⁷. For example, the percentage of UK students with a diagnosis of autism has increased four-fold over the past decade⁸. We do not have exact numbers for all neurotypes as HESA does not collect data, but it is expected these will have increased too. This may be partially attributed to the UK’s widening participation agenda, which has aimed to improve access to higher education for underrepresented groups, including neurodivergent students⁹. However, it is also important to recognise that there has been increased identification of neurodivergence in the general population¹⁰ and thus the change in higher education reflects population wider increases.

Neurodivergent students often face distinct challenges for their mental health and wellbeing¹¹. Rigid curricular structures and assessment practices frequently fail to accommodate diverse learning needs¹². Furthermore, there is a pressure to fit in, and whilst this is a challenge for all students¹³ it is often experienced more acutely by those who are neurodivergent due to differences in social communication¹⁴. In some cases, neurodivergent students may engage in masking behaviours to conceal their differences, which has been linked to negative impact on wellbeing¹⁵. Furthermore, research indicates that neurodivergent students experience higher rates of anxiety, depression, and other mental health concerns when compared to their neurotypical peers¹⁶.

Neurodivergent students are not alone in struggling. It is estimated that over 40% of young adults in higher education may be struggling with their mental health¹⁷. Emerging adulthood (18–25 years) is a period of heightened vulnerability, with most mental health problems first arising before the age of 25^18,19. Across the UK and globally, prevalence rates for anxiety, depression, and self-harm among this age group are rising^20–22. The university environment, while a potential catalyst for social mobility and long-term mental health benefits²³, also presents acute stressors—academic pressure, financial insecurity, social dislocation, and disrupted care continuity—that can exacerbate distress^24–26.

In response to the rising prevalence and complexity of student mental health needs, there has been a growing policy shift toward whole-institution approaches within higher education, aiming to promote good mental health²⁷. This reflects a recognition that reactive, individual-level interventions—while important—are insufficient to address the systemic, structural, and cultural factors that shape student mental health. The university environment itself, such as its policies, pedagogy, social infrastructure, and institutional culture, plays a crucial role in influencing wellbeing²⁸. Preventative approaches must therefore extend beyond one-to-one or reactive services to foster inclusive, supportive, and connected campus climates^29,30. Settings-based interventions, which embed mental health promotion within the broader social and organisational context of universities, have demonstrated promise^31–33.

In the UK, the University Mental Health Charter^29,30 exemplifies this shift. It outlines a preventative, whole-institution framework centred on promoting positive culture change and embedding mental health into all aspects of university life, from leadership and curriculum to student services. The Charter’s implementation has been supported by government policy, asking all English universities to adopt whole-university approaches^34,35. While this has led to the introduction of new wellbeing services at some institutions³⁶, the effectiveness and inclusivity of this approach—particularly for neurodivergent students—remains unclear. Provision and access to specialist support vary significantly across institutions, raising concerns about equity and adequacy.

As Waddington and Bonaparte³⁷ argue, if the primary mission of a university is to support student learning, then compassion should be central to its identity. A compassionate institution actively designs policies and practices that foreground human experience, fostering environments grounded in mutual respect, psychological safety, and strengths-based approaches. For neurodivergent students, such approaches aligned with the neurodiversity movement offer an important corrective to neuro-normative systems and enable diverse ways of thinking to flourish³⁸. Legal frameworks also reinforce this imperative: under the Equality Act 2010, universities have a statutory duty to provide support for disabled students. Updated guidance from the Equality and Human Rights Commission further clarifies that institutions must make anticipatory adjustments, regardless of formal diagnosis³⁹, underscoring the need for proactive, inclusive systems of care.

In this context, it is vital to consider how support is designed and delivered for neurodivergent students. In general, support for neurodivergent students remains reactive and deficit-focused. For instance, a student with ADHD might be labelled as disruptive or lazy due to difficulty concentrating or staying seated in class. Instead of recognizing these behaviours as signs of a different way of processing information, support is often given only after the student has failed multiple assignments or received disciplinary action⁴⁰. This reactive and deficit-focused approach can leave students feeling misunderstood and unsupported, and it may hinder their educational and emotional development. However, here too we are seeing a change in approach. The positive psychology movement introduced a strength-based approach, which views individuals holistically and focuses on recognising and building upon inherent strengths⁴¹. Such an approach offers a promising framework for enhancing their mental health and overall wellbeing⁴² and is increasingly being adopted in the education sector within schools using approaches such as the character strengths framework or the PERMA model, which focuses on Positive Emotion, Engagement, Relationships, Meaning, and Accomplishment in order to lead to flourishing^43,44.

As the conversation around neurodiversity continues to evolve, it is essential for higher education to adapt and embrace the diverse strengths of all students⁴⁵. In doing so, the sector can create a more equitable and supportive environment that enhances mental health and wellbeing for everyone⁴⁶. The neurodiversity movement started in the late 1980s and grew from online communities of neurodivergent people—the term neurodiversity was collectively developed by several people in these communities⁴⁷. There is still a long way to go. Currently, within the identification and support options provided for neurodivergent people, the language used is often pathologizing and can lead to stigma⁴⁸. There has been a growing shift to update the language used to be more inclusive and reject medicalised terminology and approaches to support⁶. Adopting strengths-based approaches offers the opportunity to support neurodivergent students without the stigma arising from medicalised or deficit-based approaches.

Previous systematic reviews of students in higher education have focused on what factors influence mental health and wellbeing and prevalence⁴⁹; showing, for example, that students with autism have an increased risk of developing mental health challenges⁵⁰. While previous systematic reviews have examined general student mental health and wellbeing in higher education and highlighted a range of approaches, such as peer support, mindfulness programmes, or counselling services⁵¹, none have specifically focused on neurodivergent students as a distinct population group. This is a notable gap in the literature, given the increasing number of neurodivergent students entering higher education and the unique challenges they may face. By centring neurodivergent experiences, this review provides a more targeted understanding of how current support systems are designed and implemented, and whether they are inclusive and appropriate for students with diverse cognitive profiles.

The aims of the review were:

1. To identify interventions for neurodivergent students designed to enable good mental health and wellbeing in higher education.

2. To categorise the types of interventions being used and ascertain whether any strength-based support is being evaluated.

Methods

The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines⁵² were followed (see Supplementary Information Tables 2 and 3 for checklists), and the review was registered with the International Prospective Register of Systematic Reviews (Prospero ID number: CRD42024597980).

Eligibility criteria

Qualitative and quantitative studies were included in this review.

The population of interest was neurodivergent adult (≥18 years) Higher Education students, including undergraduate or postgraduate students, studying anywhere in the world. Within the review the definition of neurodivergent adopted was the one used by Clouder et al.¹¹ which includes Autism (also referred to as Autism Spectrum Condition, Autism Spectrum Disorder and ASD), Attention Deficit Hyperactivity Disorder (also referred to as ADHD, Hyperkinetic Disorder or ADD), Dyslexia, Dyscalculia, Dysgraphia, Dyspraxia or Developmental Co-ordination Disorder (DCD), Obsessive Compulsive Disorder (OCD), and Tourette’s Syndrome.

Studies were included where the intervention aimed to (1) directly or indirectly improve mental health and/or wellbeing; or (2) improve the student experience, and in doing so may have supported better mental health and/or wellbeing. Interventions were included that focus on the individual or consider public health level approaches to fostering better mental health and/or wellbeing across a group or community.

A comparator/control was not an inclusion criterion.

Studies that focus on higher education or the transition into higher education were included.

Outcome measures of relevance included measures of student mental health and/or wellbeing, including, for example: stress, anxiety, depressive symptoms, resilience, empowerment, loneliness, quality of life, burnout, belongingness, self-efficacy, and self-esteem.

No exclusion criteria were applied beyond the predefined inclusion criteria stated above.

Information sources

In May and June 2024, a systematic search was conducted for studies written in English in the following databases: Ovid (PsycINFO, Medline, EMBASE), Web of Science (Core Collection), and Education Resources Information Center (ERIC). This was done using advanced searching and Boolean operators. Searches were limited to peer-reviewed articles published in journals. Studies conducted worldwide, written in the English language, were included. An updated search was also done in May 2025.

Search strategy

Search terms were identified using scoping searches in Ovid and then adapted to use in the other databases (see supplementary information for full search terms). Keyword searching in the title and abstract fields only was conducted using keywords for neurodivergent (as outlined in population), keywords for higher education (including university, college, students, graduate or undergraduate), and keywords for mental health and wellbeing (as outlined in outcome measures).

Selection process

Rayyan and EndNote software were used to store references, remove duplicates, and conduct screening. In the first stage of screening, the titles and abstracts of papers identified by the electronic searches were independently screened for inclusion by the research team. All titles and abstracts were screened by author 1. Twenty percent of titles and abstracts were screened by all authors to ensure consistency. Disagreements were resolved through discussion, and discussion was used to adapt and refine further title and abstract screening. Papers where there was any uncertainty were carried forward to the second stage. In the second stage, a full text review was conducted by author 1, and any uncertainties were also discussed again with all the authors to decide on the final included papers.

Data extraction

Data extraction was managed in Microsoft Excel. A data extraction table was designed based on the TIDieR checklist⁵³. This was revised to include more specific information about the interventions used as per the aims of the review and available data, and was then used to extract data from all included papers. Interventions were classified by type based on how the authors described their intervention in the paper.

The following data were extracted from all included studies:

General information—authors, year of publication, country and university of recruitment, number of study centres, neurotype of focus.

Participant information—sample size, undergrad/postgrad and year of study, mean age, ethnicity and gender distribution, comparator (if present).

Intervention details—type of intervention, rationale for intervention, tailoring of intervention and any materials provided to participants, intervention facilitator job role, format in which intervention is delivered (how, where, when and how often).

Method, analysis and results—adherence to plan/dropout rate, study method (quant/qual/mixed), timepoints of measurement, quantitative measures of wellbeing and/or mental health of students at pre, during and post-intervention, key findings, themes from analysis and conclusions.

Study risk of bias assessment

Given that the included studies were a mix of designs and included both quantitative, qualitative, and mixed designs, a quality assessment was conducted following the Mixed Method Appraisal Tool (MMAT). This critical tool was used to guide a detailed summary of the quality of studies, rather than a rating given⁵⁴. The assessment is based on five main criteria: sampling strategy relevant to address the research question; sample representative of the target population; measurement appropriateness; the risk of non-response; and appropriateness of statistical analysis used.

Synthesis methods

There was considerable variation in the types of studies reported, and inconsistencies in the data reported across studies; therefore, an Adapted Synthesis Without Meta-analysis (SWiM) reporting guidelines were used to report the synthesis. It was not appropriate to conduct a meta-analysis given the heterogeneity of studies⁵⁵.

Results

Study selection

As summarised in Fig. 1, searches generated n = 7551 records from all databases after removing duplicates. In total, 37 studies are included in the final review.

Fig. 1.

PRISMA flow diagram.

Study characteristics

As summarised in Table 1, studies were conducted in seven countries, with a large majority, 73% (n = 27), taking place in the USA. There were 10% (n = 4) conducted in the United Kingdom, 5% (n = 2) in China, and 3% (n = 1) in Australia, Norway, Israel, and Spain, respectively.

Table 1.

Included studies and participant details

[Review number] Authors and year (Reference number)	Country	Neurotype	Sample size	Ethnicity and gender distribution (M = Male, F = Female)
[1] Ahmann et al., 201883	USA	ADHD	1	White, F
[2] Anastopoulos et al., 202084	USA	ADHD	88	12% Hispanic/Latino, 64% Caucasian, 18% African American, 6% Multiracial. 59% F, 41% M
[3] Anderberg et al., 201785	USA	Autism	21,713	Not reported
[4] Andreassen, Jensen and Bråten, 201786	Norway	Dyslexia	34	70% F, 30% M
[5] Bellman et al., 201587	USA	ADHD/LD	41	68% Caucasian, 20% Asian, 15% African American, 5% Hispanic, 2% American Indian, 7% Multi-race. 47% F, 53% M
[6] Birdsey and Walz, 202188	UK	Autism	1	White, M
[7] Capriola-Hall et al., 202189	USA	Autism	32	81.30% Caucasian, 25% F
[8] Coleman et al., 202490	USA	Autism	4	75% F, 25% M
[9] Eddy et al., 202191	USA	ADHD	250	6.8% Hispanic/Latino, 66.3% Caucasian, 14.2% African American, 5.3% Asian, 0.4% native American, 10.6% multi-racial, 3.3% other. 66% F, 34% M
[10] Eddy et al., 201592	USA	ADHD	4	Caucasian, 25% F, 75% M
[11] Field et al., 201393	USA	ADHD	160	Not reported
[12] Gabriely et al., 202094	Israel	ADHD/LD	73	74% M, 26% F
[13] Gillespie-Lynch et al., 201759	USA	Autism	58	78% White, 20% Hispanic, 3.4% Black/Native American, 1.7% Muslim, 1.7% Indian, 1.7% as mixed ethnicity. 69% M, 31% F
[14] Gu et al., 201895	China	ADHD	54	44.5% F, 55.5% M
[15] Gustin et al., 202096	USA	Autism	11	81.8% M, 18.2% F
[16] Hale and Sanders, 202397	USA	Autism	1	M
[17] Harris et al., 202198	USA	ADHD	11	81.8% White, 9% Hispanic, 9% biracial. 23.3% M, 72.7% F
[18] Hillier et al., 201899	USA	Autism	52	86.5% Caucasian, 7.8% Hispanic, 3.8% Asian, 1.9% African American. 98% M, 2% F
[19] Kuo et al., 201658	Taiwan	Autism	5	40% F, 60% M
[20] LaCount et al., 2015100	USA	ADHD	17	35.3% M and 64.7% F
[21] LaCount et al., 2022101	USA	ADHD	36	Not reported
[22] Lei et al., 2018102	UK	Autism	122	Overall: 89.38% Caucasian, 2.28% Asian, 0.64% Caribbean, 2.54% Mixed 3.76% Other, 1.4% not reported. 75% M, 25% F
[23] Lester et al., 2019103	USA	ADHD	7	Not reported
[24] Lopez-Pinar and Vicente-Gispert, 2023104	Spain	ADHD	2	100% F
[25] Lucas and James, 2018105	UK	Autism	16	75% M, 25% F
[26] McDowall, Rimfeld and Krishnan 202460	UK	Dyslexia	105	61% F, 37% M, 2% non-binary. 53% White British, 10% Asian British/Asian, 3% Black British, Caribbean or African, 3% multiple ethnicities, 31% not reported
[27] Meinzer et al., 2021106	USA	ADHD	113	74.35% White/Caucasian, 8% Black, 3.5% Asian, 2.65% more than one race, 11.5% Hispanic/Latino. 49.5% M and 50.5% F
[28] Prevatt and Yelland, 2015107	USA	ADHD	148	73% Caucasian, 8% African American, 2% Asian American, 8% Hispanic, 2% Multiracial and 7% not reported. 49% F
[29] Pugliese and White, 2014108	USA	Autism	5	100% Caucasian, 100% M
[30] Richman et al., 2014109	USA	ADHD/LD	24	12.5% Black, 75% White, 8.3% Asian, 4.2% other. 50% M and 50% F
[31] Shaikh, 2018110	USA	ADHD	54	48.1% Caucasian, 16.7% Hispanic/Latino, 14.8% Multi-racial, 9.3% Asian/Asian American, 9.3% other/prefer not to answer, 1.9% American Indian/Alaskan Native. 66.6% F, 33.4% M
[32] Siew et al., 2017111	Australia	Autism	10	30% F, 70% M
[33] Stark et al., 202356	USA	Dyslexia/LD	2	White, F
[34] Swartz et al., 2005112	USA	ADHD	1	White, F
[35] Vajda, 202357	USA	ADHD	2	Not reported
[36] White et al., 2017113	USA	Autism	26	Not reported
[37] Zwart, Lavonne and Kallemeyn, 2001114	USA	ADHD	50	94% Caucasian, 6% other. 52% M, 48% F

In total, 22 (59%) of the included studies used quantitative methods, 3 (8%) used qualitative methods, and 12 (32%) used mixed methods. Twenty-four percent of manuscripts reported case studies, and 27% of studies compared an active intervention to a control condition.

Population

Most interventions looked solely at ADHD (46%, n = 17) or Autism (38%, n = 14). Three studies focused on ADHD alongside other learning difficulties (LD), which were not further specified. Two studies focused on dyslexia, one on dyslexia alongside other learning difficulties, which were not further specified. No studies addressed dyscalculia, dysgraphia, dyspraxia, OCD, or Tourette’s. In terms of participant characteristics, in studies that reported ethnicity, a large majority of participants were White/Caucasian. While 7 studies used a large sample (n > 100), 13 studies had a sample of between 21–99, which means 17 (46%) studies had a very small sample (n < 20).

Intervention details

A variety of different types of intervention were used, including Cognitive Behavioural Therapy (CBT) (27%, n = 10), coaching (19%, n = 7), self-help (8%, n = 3) peer support (11%, n = 4), counselling or psychotherapy (11%, n = 4), mentoring (11%, n = 4), mindfulness (8%, n = 3) and neurofeedback or biofeedback (5%, n = 2) (Table 2). Interventions were delivered with varying rationale: to reduce ADHD symptoms/traits (22%, n = 8), to improve executive functioning deficits (18%, n = 7), to improve daily functioning (16%, n = 6), to reduce anxiety or depression (18%, n = 7), to improve self-efficacy (14%, n = 5), to address burnout (3%, n = 1), to reduce alcohol use (3%, n = 1), to improve empathy (3%, n = 1) and to improve social difficulties (3%, n = 1).

Table 2.

Intervention details

Review number	Intervention	Facilitator	Intervention delivery	Method	Key findings and conclusions
[1]	Coaching	Coach and psychiatrist	Weekly, 8 weeks. Initial session, 2 h, then 1 h.	Qual	Integration of health and wellness coaching with psychiatric care improved management of ADHD for return to study.
[2]	CBT	Psychologist	Active treatment: 6–10 weekly 90 min CBT group sessions, and 30 min individual sessions. Maintenance: 1–2 CBT group sessions and 4–6 individual sessions.	Quant	Reduction in anxiety and depressive symptoms.
[3]	Psychotherapy	Therapist	Group/individual therapy, varied by student.	Quant	No significant difference between the two groups; both showed improvement after treatment. Autistic students stayed in therapy longer and took longer for improvement.
[4]	Self-studying	Self	Group meeting for instructions. Online diaries and strategies were self-recorded.	Quant	No significant difference in self-efficacy found.
[5]	Coaching	Coach	One hour in person sessions each week.	Mixed	Academic coaching can improve self-confidence and motivation in STEM students with disabilities.
[6]	CBT	Trainee clinical psychology students	16 sessions of CBT, 75 min per session, weekly over 5-months.	Quant	In the ideographic measures, social anxiety decreased. Could be useful intervention for autistic students.
[7]	CBT	Counsellor	1-1 counselling in 1 h sessions over a 12–16-weeks. Counsellor contacts between sessions, providing calls, texts and emails.	Quant	Participants assigned to the intervention had a significant decrease in depressive symptoms compared to the control group.
[8]	Heart Rate Variability (HRV) Biofeedback Therapy	Researcher	Biofeedback software measured participants HRV. 20-min sessions, weekly over 10 weeks.	Quant	There was an improvement shown in HRV coherence, but it had a detrimental effect on reported levels of anxiety.
[9]	CBT	Psychologist	Active treatment: 6–10 weekly 90 min CBT group session and individual sessions of 30 min. Maintenance phase: 1–2 CBT group sessions and 4–6 mentoring sessions.	Quant	The immediate access group significantly improved wellbeing, delayed condition did not. There is a positive impact on wellbeing for college students with ADHD.
[10]	CBT	Therapist	8 sessions of brief CBT.	Mixed	High self-reported value of treatment and reduction in anxiety.
[11]	Coaching	Coach	Two-hour intake, then one 30 min session per week over the phone for 24 weeks. Also provided with email and phone check-ins.	Mixed	Improvement in executive functioning skills. Coached students’ wellbeing scores were significantly higher than comparison group.
[12]	Mindfulness/device-guided slow breathing (DGB)	Teacher for mindfulness	Mindfulness: 8 weekly meetings of 2.5 h, a retreat day between 6th–7th day. DGB: daily practice with the device for 15 min for 3 weeks.	Quant	Self-report scale of hyperactivity and inattention showed decrease in mindfulness group. A reduction in breathing rate in the DGB group.
[13]	Mentoring	Students	Weekly 1 h group sessions following a curriculum and/or weekly 1-h 1-1 sessions. 2–9 mentees in each group.	Mixed	No changes in self-reported autism, anxiety and overall perceived social support by participation. A curriculum of self-advocacy is likely to encourage acceptance.
[14]	Mindfulness-based cognitive therapy	Therapist	6 weekly sessions.	Quant	Overall trend of lowered ADHD symptoms, significant change in depressive symptoms.
[15]	Cooking	Dietitian nutritionists	6 cooking modules in weekly sessions 2–4 p.m. on a Friday.	Quant	Exposure to wider variety of food and learning new cooking skills are promising to improve wellbeing.
[16]	Medication, CBT and exercise	Behavioural neurologist	Not reported	Quant	After 18 months they returned to college.
[17]	Neurofeedback	Researcher	16 total sessions over 8–10 weeks.	Quant	Decrease in depression scores, in 9 cases anxiety was lower. Mean self-efficacy scores improved. Can lower depression and anxiety and improve self-efficacy in college students.
[18]	Support Group	Counsellor/disability staff	1 h a week for a 7-weeks, groups were 4–7 participants.	Mixed	Higher self-esteem, reduced loneliness and lower generalised anxiety. Support groups could be an effective support strategy in college.
[19]	Enrichment Programme	Professor/therapist	Weekend sessions, with different focuses	Qual	The programme encouraged participants to use creativity. Students had increased positive learning attitude, creativity, art performance and interpersonal skills.
[20]	CBT	Clinical psychology faculty/grad students	20 1-h long training sessions over 10 weeks, weekly individual sessions.	Quant	Self-reported levels of inattention reduced significantly. Completion was associated with improvement in functioning. Combined individual and group CBT training has significant effects for college students with ADHD.
[21]	High intensity interval training (HIIT)	Researchers	Two in lab morning sessions, 1 week apart. HIIT set of stretches, 3 min low intensity cycling, 16 min of 8 bouts of 20 s cycling followed by 120 s of rest. 15 min to recover before cognitive task.	Quant	Moderate-sized effect of HIIT on inattentive symptoms and hyperactive/impulsive severity. Self-reported improvements in depression. HIIT has promise for providing improvements in executive functioning ADHD symptomatology and depression.
[22]	Transition programme	Academic, support services and clinical staff or autistic students/ autistic graduates	Two overnight stays in student accommodation on campus, curriculum delivered over 3 days. Sessions delivered face to face in a group. Three strands of sessions “work”, “rest” and “play”.	Mixed	Majority provided positive feedback. Concerns of autistic students about attending university can be significantly reduced by the attendance of a transition programme.
[23]	Mindfulness	Psychologist/researcher	8 weeks of group intervention for 1 h.	Mixed	5 participants psychological distress decreased. Mindfulness interventions for college students with ADHD is an effective for reducing ADHD symptoms and psychological stress.
[24]	CBT	Clinical psychologist	11 1-h weekly sessions, grouped into 5 modules.	Mixed	For 1st student self-reported scores for ADHD related behaviours, executive functioning and emotional problems improved significantly. For the 2nd student, the quant results did not show clear improvements. Qualitative data showed improvements for both.
[25]	Specialist mentoring	Specialist mentors	Weekly sessions- on average this was 8 in Autumn time point, 18 in Spring and 33 in Summer.	Mixed	Satisfaction was high, it helped with university life, social skills and wellbeing. Mentoring is beneficial for students with autism and can take pressure off university services.
[26]	Cognitive Reappraisal	Researchers	Participants took part in online task with scenarios developed by dyslexic students	Quant	All students had reduction in anxiety, showing cognitive reappraisal is of benefit.
[27]	Brief motivational interviewing (BMI) and behavioural activation (BA)/ supportive counselling (SC)	Therapist	5 sessions over 7 weeks. Session 1–4 were in person for 60 min. Session 5 was a 20-min phone call 2 weeks after the last in person session.	Quant	No significant difference in alcohol related negative consequences or depressive symptoms. The inclusion of BA with BMI resulted in greater decreases in negative alcohol related consequences when compared to BMI and SC.
[28]	Coaching	Counselling psychology doctoral students	Individual sessions once per week for 8-weeks. Between sessions coaches used pre-arranged texts, emails and calls.	Quant	There were significant effects between pre and post intervention. Positive changes were associated with lower self-ratings of ADHD symptoms, depression and anxiety.
[29]	Problem-solving therapy (PST)	Clinical psychology doctoral students	9 sessions of 90 mins.	Quant	Significant improvement in problem solving was only observed for 2 participants. PST could be a promising way to address challenges faced by autistic college students.
[30]	Coaching	Coaches	6–12 weekly coaching sessions for two 12-week semesters. 30 min sessions took place either in person or campus or via phone.	Mixed	No statistically significant differences found between treatment and comparison group. Qualitative interviews showed students found the coaching useful. Coaching holds promise in supporting students with ADHD/LD
[31]	Group therapy	Clinical psychologist and therapist	Students met individually first. 12 weekly 90 min group meetings.	Quant	Participants in the group therapy improved significantly more than the control group on measures of self-esteem, psychosocial confidence and emotional maturity. Using interpersonal group therapy for students with ADHD can improve their self-esteem.
[32]	Specialist mentoring (peer)	Postgrad students in psychology/occupational therapy/ social workers	1-1 session weekly for 1 h. 90 min per week of group activities, which also included social activities such as bowling.	Mixed	Significant improvement in social support scores, post reduction in general communication apprehension. Qualitative results showed positive features of the programme. A structured specialist programme of peer support can be beneficial to the wellbeing of autistic students.
[33]	Mentoring	Clinical mental health graduates/counsellors	6 sessions over zoom, lasting between 30–50 min. One student met every 2 weeks, one student met weekly.	Mixed	One participant scored increases on all measures, other participant scores increased on 5 of the measures. Both participants in this study found that there were benefits from taking part in the solution focused mentoring.
[34]	Coaching	Doctoral students in counselling and psychology	Choice of weekly or biweekly face to face meetings. Phone calls and emails were also used between meetings.	Quant	Participant showed improvement in four of the seven areas of goals. The study presented a single case study of coaching for a participant with ADHD with promising results on achieving set goals.
[35]	Coaching	Researcher	Individual weekly online coaching sessions for 8 weeks, 30–40 min each. Between sessions offered regular communication and check ins via text message.	Qual	Students changed negative mindset about themselves, students became more confident and self-regulated learners, students learned new tools to implement on their own in the future. Coaching on a larger scale can provide strength-based tools and strategies for students to understand their neurodiversity.
[36]	CBT	Counsellor	Step 1: 6 counselling sessions, every fortnight, various people present- student only/ student-parent/ school personnel. Step 2 students had 1-1 counselling 12-3 sessions over a 12–16-week period. Counsellor also acts as a coach between sessions provided calls, texts and emails to check in.	Quant	Ratings so far show the programme is helpful for students.
[37]	Peer-Based Coaching	Students	Training group seminar 4 h and ongoing training in the academic year. Sessions 1 h a week for at least 5 weeks. Students fill out a student needs checklist to identify what they want help with.	Quant	The experimental group had significant improvements on the self-efficacy scales and motivation, time management and anxiety. Findings suggest a peer-based coaching programme can be helpful to enhance self-efficacy.

Notably, the rationale for interventions is predominantly to address deficits. Only two of the included studies discussed students’ strengths as part of the intervention^56,57. The rationale of these studies was to improve self-efficacy and to improve executive functioning, respectively. One study focused on creativity⁵⁸ and was not, therefore, explicitly deficit-focused.

One study⁵⁹ consulted autistic students prior to developing the intervention to assess the needs of students and develop the intervention on this basis. There was also one study⁶⁰ which codesigned parts of the intervention with students.

In terms of how the interventions were facilitated, there was a variety of approaches taken. Trainee psychologists under supervision facilitated some of the interventions (22%, n = 8), some were delivered by the researchers (14%, n = 5), and some by students themselves or peers (8%, n = 3). The remaining interventions were delivered by specialists with expertise or experience in the method of intervention (56%, n = 21).

The format of the intervention varied across the included studies. In terms of the length of the intervention, there were insufficient details in 6 of the included studies to report this. For the remaining studies, there was a range of less than 6 weeks to over 21 weeks for the intervention; less than 6 weeks (14%, n = 5), between 6–10 weeks (43%, n = 16), between 11–20 weeks (16%, n = 6) and over 21 weeks (14%, n = 5).

Outcomes

Studies used a range of outcome measures. For quantitative and mixed methods studies, these included the Learning and Study Strategies Inventory (LASSI⁶¹), UCLA Loneliness Scale⁶², the GAD-7⁶³, Beck’s Depression Inventory⁶⁴, Beck’s Anxiety Inventory⁶⁵, the Clinical Global Impression Scale⁶⁶, and the Weiss Functional Impairment Rating Scale Self-Report (WFIRS⁶⁷).

Risk of bias

The MMAT is presented in Table 1 in the Supplementary Information. All studies were eligible to be rated by the MMAT based on the screening questions. For all included qualitative studies (n = 3), the qualitative approach was appropriate to answer the research questions. Furthermore, the methods of data collection were broadly adequate to address the research question. However, it was unclear if findings were adequately derived from the data, and studies lacked clarity over whether the results were substantiated by the data and if there was coherence between qualitative data sources, collection, analysis, and interpretation.

For randomised control trials (n = 5), complete outcome data were included in all studies and in all but one study, the participants adhered to the assigned intervention and groups were comparable at baseline. The weakest area for RCTs was whether randomisation was adequately performed as this was no or unclear in 4 studies. For non-randomised studies (n = 13), there was considerable inconsistency. Most studies included appropriate measurements and over half had complete outcome data. Fewer than half clearly had a representative population. Similarly, mixed results were found for whether the intervention was delivered as planned and management of confounding variables.

For quantitative descriptive studies (n = 4), the quality varied considerably with one study receiving a positive appraisal across all five criteria and another having an unclear or negative appraisal for all criteria. Three of the studies were recorded as having a low nonresponse bias. There was a lack of clarity for most studies in terms of whether the statistics was appropriate and whether there was a representative sample. Finally, for mixed method studies (n = 12), the majority of the studies did meet most criteria with two studies meeting all criteria and a further four meeting 80% and three meeting 60%. Where studies could not be positively appraised, this was largely due to a lack of clarity.

The MMAT helped identify strengths and limitations within individual studies, which informed the overall interpretation of the evidence base. No studies were excluded based on the MMAT answers, as they did not impact the aims of this review.

Discussion

This systematic review aimed to search and critically examine the existing literature on interventions designed to support neurodivergent students in maintaining good mental health and wellbeing within higher education settings. The purpose was not only to identify and catalogue the types of interventions currently being implemented but also to analyse and group interventions according to their characteristics, such as design and target neurotype. Furthermore, the review sought to investigate the overarching theoretical or practical approaches adopted when supporting neurodivergent students, such as whether they take a strength-based or deficit-based stance. In particular, it identified whether any of the existing interventions were aligned with strength-based practices that focus on harnessing the unique abilities and perspectives of neurodivergent students, as opposed to attempting to correct or minimise their differences.

A wide array of interventions aiming to promote mental health and wellbeing for neurodivergent students were identified, ranging from social skills training to cognitive-behavioural therapies and peer mentoring. However, very few interventions adopted a strength-based perspective. In fact, only two studies made explicit mention of strength-based approaches, and both were small-scale case reports with limited generalisability. The overwhelming tendency was to adopt a deficit-based model, in which the intervention’s goal was to mitigate or reduce characteristics associated with neurodivergence. This was particularly evident in studies aiming to improve empathy or address social difficulties, objectives that implicitly pathologise neurodivergent communication styles. Such approaches assume that neurodivergent ways of interacting are inherently flawed, rather than acknowledging that communication breakdowns often arise from mutual misunderstandings between neurodivergent and non-neurodivergent individuals^68,69. An alternative theoretical model, the double empathy problem, proposes that social communication challenges are two-way and rooted in the differences in how autistic and allistic (non-autistic) people perceive and relate to each other⁶⁸. Despite growing support for this more relational model⁷⁰ a strengths-based approach remains underrepresented in the interventions reviewed.

In terms of the neurotypes represented in the literature, the bulk of the research focused on either autism or ADHD. There was relatively little research attention given to other forms of neurodivergence, such as dyslexia or learning differences, and no identified interventions specifically targeted students with dyscalculia, dysgraphia, dyspraxia, OCD or Tourette’s. This lack of diversity in focus points to a considerable gap in the research, suggesting that many neurodivergent students may be left unsupported or invisible within current intervention frameworks⁷¹.

Another notable gap concerns the involvement of neurodivergent students in the research and intervention design process. Since research and resulting interventions can have a big impact on the lives of neurodivergent people, it is important to include them in meaningful ways throughout the research process^72,73. There is growing evidence that participatory and co-productive approaches lead to more relevant, respectful, and effective outcomes⁷⁴; however, only two studies in this review had engaged neurodivergent students in any meaningful way to inform the design or evaluation of interventions. This lack of consultation highlights the need for more inclusive research practices that centre the voices and expertise of neurodivergent individuals themselves.

The 2013 update to the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) acknowledged the possibility of co-occurring conditions, such as ADHD and autism, which has gradually begun to influence how services are conceptualised and delivered⁷⁵. Moreover, emerging research suggests that co-occurrence is more the norm than the exception among neurodivergent individuals⁷⁶. However, this review found that most interventions continue to be designed for specific, isolated neurotypes, rather than addressing the needs of students with multiple overlapping neurodivergent identities. Most of the included studies excluded participants if they had co-occurring neurotypes. This siloed approach risks overlooking the complexity and individuality of students' lived experiences, further reinforcing a one-size-fits-all model that may not adequately support those with intersecting needs.

Most studies were conducted in the United States. In the USA, a strong emphasis is placed on individual responsibility and productivity within a capitalist, neoliberal framework⁷⁷. As a result, the interventions developed and studied within this context may reflect these societal values, which may not necessarily align with educational or cultural contexts in other parts of the world. Consequently, the generalisability and applicability of these findings to countries with differing educational philosophies or healthcare systems may be limited, and caution should be taken when attempting to apply USA based models elsewhere.

Regarding the demographic composition of participants, the majority identified as White or Caucasian, with minimal representation from other ethnic backgrounds. This lack of diversity is concerning, as it limits our understanding of how intersecting identities, such as race and gender may influence the effectiveness or accessibility of interventions⁷⁸. Interestingly, there was an even gender split in the participants overall. This even representation of men and women contrasts with commonly held assumptions that certain neurotypes (like autism) are predominantly male, an assumption that may reflect biases in diagnosis rather than actual prevalence⁷⁹. Disability disclose could also be a factor for some groups, impacting support offered⁸⁰.

In summary, while there is a growing body of work aimed at supporting neurodivergent students in higher education, this review highlights major gaps in representation, methodology, and approach. Most existing interventions remain rooted in deficit models, are predominantly USA based, and fail to include meaningful involvement from neurodivergent students themselves.

This review adopted the definition of neurodivergent used by Clouder et al.¹¹. There is some discussion around which neurotypes are included under the neurodivergent umbrella. Results would differ with a shift in what neurotypes fall into this categorisation. Our key search terms related to specific neurotypes (e.g. ADHD), rather than neurodiversity in general. This approach of naming conditions may have prevented the identification of some sources, as terms and language used have been continually updated⁴⁸.

Due to the variety of different methods included in this review, a meta-analysis was not possible; therefore, the narrative synthesis of presented information was selected subjectively. Due to the focus on interventions in this review, it may be that some strength-based approaches in general support for neurodivergent students were not included. This is because an intervention approach can automatically assume a medicalised and most often deficit viewpoint.

Searches were limited to English language literature; therefore, evidence from studies reported in other languages was not included. This could mean that there are interventions which exist which are not English language based and have been missed in this review.

Within educational research, as demonstrated in this review, there is a wide variety of types of interventions used, making standardised checklists, such as TIDieR, more challenging to follow. Recently, two checklists have been developed—CLOSER and CIDER⁸¹—which will enable future research to report on intervention details within a specific framework design for educational research.

The MMAT risk of bias demonstrated that some of the included studies have a high risk of bias, especially those in the quantitative descriptive studies. There was a lowest risk of bias in mixed method studies; however, there were still 30% of questions that were answered no or unclear. This demonstrates that some of the included studies had high risk of bias.

Strength-based models have been advocated for students in secondary education⁸², however, as demonstrated not used frequently for higher education students. There should be further research which aims to co-create interventions that would support neurodivergent students’ mental health and wellbeing in higher education from a strength-based, rather than a deficit model. There is significant scope for future research to develop more inclusive, culturally sensitive, and strength-based interventions that reflect the full diversity of the neurodivergent student population. Most of the interventions included are delivered by professionals and took place over a significant time frame in the context of education, which could have some implications for the cost and resourcing needed to deliver interventions. For these to also be evaluated effectively, it is important to consider these points when designing and developing appropriate interventions to support students in this population.

This review aimed to systematically review interventions to support mental health and wellbeing in neurodivergent students in higher education. Interventions were mostly focused on ADHD or autism. There were a variety of different interventions used; however, all but two of the studies were focused on improving deficits rather than strength-based supports. Additionally, only two of the studies included neurodivergent students in the intervention design. These findings demonstrate a gap in support options for neurodivergent students, especially those who are multiply neurodivergent or for students who may not wish to access support focused on perceived deficits in functioning.

Author contributions

Conceptualisation: F.R., E.J.D., and N.B.; methodology: F.R., E.J.D., and N.B.; formal analysis: F.R., E.J.D., and N.B.; writing—original draft preparation: FR.; writing—review and editing: F.R., E.J.D., and N.B.; supervision: E.J.D. and N.B. All authors have read and agreed to the version of the manuscript.

Data availability

All data generated during this study are reported and available in this paper and Supplementary Information.

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at 10.1038/s44184-026-00196-4.