Sage Journals: Discover world-class research

Abstract

Background:

We aimed to investigate self-medication prevalence, mental health problems, and their relationship among autistic adults. We also sought to identify key factors associated with anxiety and depression in this population.

Methods:

An online cross-sectional questionnaire survey was conducted with self-identifying or formally diagnosed autistic adults who reported prior or current substance use. Descriptive statistics, chi-square tests, and backward stepwise logistic regression were applied.

Results:

A total of 475 participants were recruited. The prevalence of anxiety, depression, and eating disorders among the participants was 87.37%, 75.16%, and 51.16%, respectively. Notably, 80.42% of participants also identified as Attention-Deficit/Hyperactivity Disorder (ADHD). The most commonly reported substances for self-medication were alcohol (92.84%), cannabis (82.32%), and stimulants (81.26%), with 77.69% of participants reporting daily or weekly use. Logistic regression revealed that anxiety, dissociatives, and sedatives were significantly associated with depression, while depression, ADHD, and eating disorders were significantly associated with anxiety. Identifying as female was associated with lower odds of depression.

Conclusion:

Autistic adults who self-medicate show high rates of mental health problems and substance use. The complex relationship between self-medication, mental health, and co-occurring conditions underscores the urgent need for tailored support services. To reduce harmful self-medication, we must address stigma, promote acceptance, and improve access to appropriate mental health support for autistic adults.

Lay Abstract

Many autistic adults use substances such as alcohol or cannabis to cope with mental health challenges such as anxiety and depression. However, this approach often leads to more difficulties, and we need to understand these connections better to offer the right support. In our study, we asked 475 autistic adults who use substances to complete an online survey. We wanted to learn how common mental health problems are among this group and what factors make anxiety and depression more likely. We found that most participants had significant mental health challenges, including anxiety (87%) and depression (75%). Many also identified as Attention Deficit Hyperactivity Disorder (ADHD), and had eating disorders or chronic pain. The most commonly used substances were alcohol, cannabis, and stimulants, with most participants using them regularly. Our findings revealed that certain factors, like being male or nonbinary, using specific substances, or living with chronic pain, increased the likelihood of depression. Similarly, depression, ADHD, and eating disorders were linked to a higher risk of anxiety. These results highlight the need for better mental health and substance use support for autistic adults. Services should be more accessible and understanding of the unique experiences of autistic individuals. Addressing stigma and ensuring fair treatment could also help reduce both mental health struggles and reliance on substances for coping. Although our study provides valuable insights, most participants were white and of a similar age, so the findings may not apply to everyone. Nonetheless, our research offers a step forward in understanding and supporting autistic adults who face these challenges.

Keywords

autism self-medication substance use mental health depression anxiety

Introduction

Autistic individuals face a significantly higher risk of developing mental health conditions compared to the general population. Researchers have consistently shown that autistic people experience elevated rates of depression, anxiety, and suicidality (Lai et al., 2019). Hollocks et al. (2019) conducted a meta-analysis and found that the current and lifetime prevalence for autistic adults ranged between 27% and 42% for anxiety, and 23% to 37% for depression. These rates considerably exceed those in the general population, where World Health Organisation (2024a, 2024b) estimates the global prevalence of depression and anxiety disorders to be 5% and 4%, respectively. The COVID-19 pandemic has further exacerbated mental health challenges among autistic individuals, with studies reporting heightened anxiety, depression, and stress due to disrupted routines, reduced access to support services, and increased social isolation (Oomen et al., 2021; Pellicano et al., 2022). Furthermore, autistic individuals are at an increased risk of suicidal ideation and attempts. Cassidy et al. (2014) conducted a large-scale study revealing that 66% of autistic adults reported experiencing suicidal thoughts, and 35% had planned or attempted suicide. These figures are alarmingly high when compared to the general population, where researchers estimate the lifetime prevalence of suicidal ideation ranges from approximately 9.2% (Nock et al., 2008) to 17% (Cassidy et al., 2014) and 2.7% for suicidal plans and attempts (Nock et al., 2008). Cassidy et al. (2018) also identified that 72% of a sample of 164 diagnosed UK-based autistic adults scored above the recommended psychiatric threshold for suicide risk. More recently, Cassidy et al. (2022) highlighted that among coroner records of people who have died by suicide in England, at least 10.8% were evidenced to be autistic. Blanchard et al. (2021) conducted a systematic review and meta-analysis and found that the odds of self-injurious behaviour among autistic children and adults were 3.18 times higher (95% CI, 2.45–4.12) compared to nonautistic individuals.

One potential consequence of poor mental health among autistic individuals is an increased risk of self-medication. Self-medication refers to individuals using substances, such as alcohol, prescription medications, or illicit drugs, to alleviate the experience of mental and physical health difficulties without professional guidance or supervision. While self-medication may provide some temporary relief, it can lead to a range of negative outcomes, including substance misuse, addiction, and worsening of mental health conditions (Feingold & Tzur Bitan, 2022). Indeed, the relationship between substance use and mental health is often complex and recognised as potentially bidirectional; while individuals may initiate substance use to cope with pre-existing difficulties, substance use itself can exacerbate these conditions or lead to new mental health challenges (Pacek et al., 2013; Swendsen et al., 2010). Autistic individuals who engage in self-medication may also be less likely to seek professional help, or they may self-medicate as a means of coping with delays in accessing mental health services (Camm-Crosbie et al., 2019). Furthermore, co-occurring identities such as Attention-Deficit/Hyperactivity Disorder (ADHD) which is itself linked to higher rates of substance use (Wilens et al., 2011; Zulauf et al., 2014) may add another layer of complexity to self-medication behaviours in autistic populations.

Research to date suggests that autistic individuals may be particularly vulnerable to initiating and maintaining self-medication practices. Butwicka et al. (2017) conducted a Swedish population-based cohort study and identified a higher risk of substance use problems among autistic individuals compared to the general population, with an odds ratio of 5.2 (95% CI, 4.9–5.6) for any substance use problem, 8.5 (95% CI, 7.7–9.3) for a drug use disorder, 6.4 (95% CI, 3.8–10.5) for tobacco and 4.0 (95% CI, 3.7–4.4) for an alcohol use disorder. Weir et al. (2021) found that autistic individuals were nearly nine times more likely to report using recreational substances to “manage behaviour” (OR = 8.89; 95% CI, 2.05–81.12) and more likely to report using recreational substances to manage mental health difficulties (OR = 3.08; 95% CI, 1.18–9.08) compared to nonautistic individuals.

Despite the high prevalence of mental health difficulties and self-medication practices among autistic individuals, researchers need to conduct further studies to better understand the complex relationship between these factors. We aimed to address this by assessing the relationship between self-medication and mental health among self-medicating autistic adults. Our primary objectives were to examine the association between self-medication practices and mental health outcomes and to identify key factors associated with anxiety and depression among self-medicating autistic adults. Our secondary objectives were to assess the prevalence of mental health problems among self-medicating autistic people and to describe the different types of substances this population uses.

Method

Study Design

We employed an online one-time cross-sectional questionnaire survey, co-constructing it with autistic and neurodivergent individuals who have lived experience of substance misuse, self-medication, and mental health difficulties. These individuals provided valuable insider perspectives and depth of insight throughout each element of the study including the research objectives, data collection processes, and interpretations of findings, enhancing the study's relevance, authenticity, and applicability.

Sampling and Recruitment Procedure

We employed purposive sampling to recruit participants who met our specific criteria: self-identifying or formally diagnosed autistic adults aged 18 or older with prior or current self-medication experience. To implement this sampling method, we developed a targeted recruitment strategy involving tailored recruitment messages that clearly outlined our inclusion criteria. We then strategically placed these messages on social media platforms known to have active autistic communities, including Twitter, Facebook groups, and LinkedIn pages.

To further reach our target population, we leveraged the study authors’ extensive networks. In particular, David Gray-Hammond, an international autistic expert and advocate in autistic self-medication and substance misuse, shared the survey through his following. This approach proved particularly effective in enabling us to find participants who met our inclusion criteria. Additionally, we partnered with the London Autism Group Charity and Academy, both autistic-led organisations. These organisations helped distribute our recruitment materials to their extensive networks of autistic adults, furthering our reach into the population we were targeting. Other autistic-led organisations with relevant audiences that distributed the survey included Autistic Not Weird, The Nurture Program, Tigger Training Consultancy, Bobbi Elman Consultancy, Autisticality, and Autistic Inclusive Meets. It should be noted that the online, text-based questionnaire format likely limited participation to autistic individuals comfortable with written communication, potentially excluding nonspeaking individuals or those relying primarily on augmentative and alternative communication methods. Furthermore, the recruitment criteria focused on autistic identity, and the study did not differentiate between participants with a formal diagnosis and those who self-identified.

We conducted our recruitment efforts over a two-month period, from July to August 2023. Before participants could access the full survey, we used a brief screening questionnaire. This questionnaire asked potential participants about their age, autism identification or diagnosis, and self-medication history. Only individuals who met all our criteria could proceed to the main survey. This step helped us ensure that our sample accurately represented our target population.

Measures

The questionnaire was designed to capture a comprehensive profile of the participants’ experiences with substance misuse, self-medication practices, and mental health issues. In the demographic information section, we prompted participants to disclose their age (with specified age ranges), gender (open-ended to accommodate all gender identities), and ethnicity (also open-ended to ensure inclusivity). Additionally, we required them to confirm their autistic identity, whether through self-identification or a formal diagnosis.

Questions about whether participants had ever used substances, either currently or in the past, were also added, including alcohol, nicotine, stimulants (including caffeine, amphetamines, and cocaine), sedatives, opioids, dissociatives, psychedelics, novel psychoactives, cannabis, and other (with participants asked to specify).

We included open-ended questions to explore the perceived positives and negatives of participants’ substance use experiences, as well as their experiences with specialist substance use treatment services, including the perceived effectiveness of such services, barriers to engagement, and suggestions for improvement. However, findings from these open-ended questions will be the focus of a separate forthcoming publication.

We also asked participants to disclose whether they currently self-identify or have a diagnosis of a range of mental health conditions including depression, anxiety, OCD, personality disorder, bipolar I and II, schizophrenia, schizoaffective disorder, dissociative identity disorder, and “other” (with participants asked to specify). We included questions about whether they currently have chronic pain, self-identify, or have a diagnosis of ADHD or an eating disorder. These variables, including anxiety and depression, were measured using binary (yes/no) self-report items, where participants were asked whether they currently self-identify or have received a formal diagnosis for each condition. We did not use standardised psychometric scales for these variables, as the aim was to identify prevalence and associations based on participant-identified conditions rather than severity ratings.

Ethics

The University of Bedfordshire's Institute for Health Research Ethical Committee provided ethical approval for the study. We provided participants with a detailed participant information form that described the study's aims, the voluntary nature of participation, confidentiality measures, and their right to withdraw without penalty. This form also included a trigger warning, given that the questionnaire would ask questions about personal experiences that could be potentially distressing or triggering. We established an arrangement with the London Autism Group Charity to provide support to participants who might encounter any emotional discomfort or distress during the survey. We made this support mechanism available to all participants, ensuring they had access to necessary assistance throughout their participation in the study. After reading the information form, we provided an informed consent form summarising the nature of involvement.

All collected data were treated with strict confidentiality and were stored on encrypted, password-protected Google Form servers. Our data management complied with GDPR and the Data Protection Act 2018.

Analysis

We used descriptive statistics to establish a demographic profile of the study participants, including prevalence and patterns of substance use and mental health status. Substance use was coded as a binary variable (yes/no) for each substance based on whether participants reported any use, regardless of recency or frequency.

For univariate analyses, we applied chi-square tests to assess the relationships between predictor variables and the outcomes of depression and anxiety. We also explored the relationship between depression and anxiety using chi-square tests. Given the exploratory nature of the study and the interrelatedness of the variables, we did not apply a formal correction for multiple comparisons (e.g., Bonferroni), as this may have increased the risk of type II errors and masked potentially meaningful associations. In addition, we used Phi and Cramer's V to quantify the strength of the relationship, and we calculated the Contingency Coefficient to confirm the association's significance. Furthermore, we estimated odds ratios (with 95% confidence intervals) to assess the relative risk of having one outcome given the presence of the other, which allowed us to assess the directional relationship between anxiety and depression.

Following these initial analyses, we conducted a backward stepwise logistic regression. This was appropriate for the exploratory nature of the study, where we had no predefined theoretical framework to guide the selection of predictor variables. Given the large number of potential predictors, and the absence of specific hypotheses, the stepwise method facilitated our identification of the most useful variables, thereby creating a more parsimonious model. This approach also helped us prevent overfitting by excluding variables that did not significantly contribute to the model's explanatory power. Additionally, the backward stepwise procedure offered further protection against multicollinearity. To assist interpretation of the regression findings, we later created a visual summary (Figure 1) of the significant associations observed. This is intended only as a descriptive representation of the two regression models and is not a formal structural or causal model.

Figure 1.

Data-informed conceptual model of statistically significant associations from two separate logistic regression analyses predicting depression and anxiety among self-medicating autistic adults.

In line with established statistical robustness criteria (Hosmer et al., 2013), we only included variables in the regression models if they had a minimum of 10 occurrences in at least one cross-tabulation cell. Consequently, we excluded variables that did not meet this criterion from the depression regression model, namely ethnicity, personality disorders, dissociative identity disorder, premenstrual dysphoric disorder, other specified disorders, trauma-related disorders, bipolar disorders, and schizophrenia-related disorders. Similarly, for the anxiety regression model, we were unable to include alcohol use or novel psychoactive substances due to insufficient variation in their distribution across anxiety status categories. While alcohol use was highly prevalent in the sample overall, it did not show enough variability between participants with and without anxiety to meet the threshold for inclusion. We binarised age and drug frequency variables (lower age vs. higher age and more frequent vs. less frequent use, respectively) and included them.

We considered anxiety and depression both as predictors and outcomes in each other's models, following clearance of concerns over multicollinearity through Variance Inflation Factor (VIF) tests. The VIF results confirmed that multicollinearity was not a concern for the substances we examined or for depression and anxiety as covariates.

To assess the model's goodness-of-fit, we assessed the −2 Log likelihood, Cox & Snell R Square, and Nagelkerke R Square values. We used the Hosmer and Lemeshow Test to evaluate the fit, ensuring the model's predictions were closely aligned with observed outcomes. We conducted all analyses using SPSS v26 (IBM Corp, 2020).

Results

Participant Characteristics

The study included 475 participants, the majority of whom (52%) fell within the 36–50 years age range. Younger participants comprised 6.74% (18–25 years) and 23.37% (26–35 years), while 17.89% were over the age of 50 years. Females constituted the majority (61.68%) of participants, followed by nonbinary and gender-diverse individuals (22.95%), and males (15.37%). A significant majority (90.53%) identified as White/Caucasian, with smaller representations from Mixed/Other (3.16%), Hispanic/Latinx (1.26%), Black (1.05%), Asian (0.84%), Indigenous/Native (0.63%), and Jewish (0.42%) ethnicities.

Mental Health, ADHD, and Chronic Pain Prevalence

We found high rates of anxiety (87.37%), depression (75.16%), and ADHD (80.42%) among participants. Chronic pain affected 63.79% of participants, while 51.16% reported eating disorders. Less prevalent conditions included Obsessive-Compulsive Disorder (17.47%), Personality Disorder (9.47%), and Trauma-Related Disorders (8.84%). We also noted instances of Dissociative Identity Disorder (4.42%), Premenstrual Dysphoric Disorder (0.84%), and other specified disorders (4.00%). Bipolar disorders affected 4.84% of participants, and Schizo-Related Disorders were reported by 1.89%.

Self-Medication Type and Prevalence

Alcohol emerged as the most commonly used substance for self-medication (92.84%), closely followed by cannabis (82.32%) and stimulants (81.26%). Nicotine use was reported by 68.63% of participants, while 47.79% used sedatives and 44.63% used psychedelics. Other substances included dissociatives (18.32%), opioids (36%), and novel psychoactives (12.00%).

Most participants (58.53%) reported daily substance use, with others using a few times a week (19.16%), monthly (3.58%), or infrequently/on special occasions (18.74%). We found that 13.89% of the sample had a history of engaging with specialist treatment services for substance use.

Univariate Analyses

Table 1 provides a full breakdown of the univariate analyses between the examined factors and outcome variables. We found a statistically significant association between gender and depression, with females showing a lower prevalence (70.31%, p = 0.007) compared to nonbinary and gender diverse (84.40%) and male (80.82%) participants. Anxiety prevalence did not significantly vary by gender. Depression showed significant associations with Personality Disorder (93.33%, p = 0.003), chronic pain (78.55%, p = 0.023), and the use of specific substances, namely dissociatives (85.06%, p = 0.018), psychedelics (79.72%, p = 0.039), cannabis (77.49%, p = 0.011), sedatives (81.50%, p = 0.002), and opioids (81.29%, p = 0.020).

Table 1.

Cross-Tabulations of Anxiety and Depression With Mental Health Problems, Chronic Pain, and ADHD.

	N (%) Depressed		χ² p value	N (%) Anxious		χ² p value
Demographic Characteristics
18–25 years (n = 32)	29 (90.63%)		0.214	27 (84.36%)		0.542
26–35 years (n = 111)	83 (74.77%)			101 (90.99%)
36–50 years (n = 247)	183 (74.09%)			215 (87.05%)
50+ years (n = 85)	62 (72.94%)			72 (84.71%)
Female (n = 293)	206 (70.31%)		0.007	253 (86.35%)		0.697
Nonbinary & Gender diverse (n = 109)	92 (84.40%)			97 (88.99%)
Male (n = 73)	59 (80.82%)			65 (89.04%)
White (n = 430)	321 (74.65%)		0.279	377 (87.67%)		0.512
Any other ethnicity (n = 35)	29 (82.86%)		0.279	32 (91.43%)		0.512
Mental Health/ADHD/Pain
	Yes	No		Yes	No
Depression (n = 357)	N/A	N/A	N/A	331/357 (92.72%)	84/118 (71.19%)	<0.001
Anxiety (n = 415)	331/415 (79.76%)	26/60 (43.33%)	<0.001	N/A	N/A	N/A
OCD (n = 83)	69/83 (83.13%)	288/392 (73.47%)	0.064	80/83 (96.39%)	335/392 (85.46%)	0.006
Personality disorder (n = 45)	42/45 (93.33%)	315/430 (73.26%)	0.003	41/45 (91.11%)	374/430 (86.98%)	0.427
Dissociative identity disorder (n = 21)	17/21 (80.95%)	340/454 (74.89%)	0.530	20/21 (95.24%)	395/454 (87.00%)	0.267
PMDD (n = 4)	3/4 (75.00%)	354/471 (75.16%)	0.994	2/4 (50.00%)	413/471 (87.69%)	0.024
Other (n = 19)	13/19 (68.42%)	344/456 (75.44%)	0.488	17/19 (89.47%)	398/456 (87.28%)	0.778
Chronic pain (n = 303)	238/303 (78.55%)	119/160 (69.19%)	0.023	269/303 (88.78%)	146/172 (84.88%)	0.219
ADHD (n = 382)	289/382 (75.65%)	68/93 (73.12%)	0.612	341/382 (89.27%)	74/93 (79.57%)	0.012
Eating disorder (n = 243)	188/243 (77.37%)	169/232 (72.84%)	0.254	220/243 (90.53%)	195/232 (84.05%)	0.033
Trauma-related disorders^a (n = 42)	34/42 (80.95%)	323/433 (74.60%)	0.363	39/42 (92.86%)	376/433 (86.84%)	0.262
Bipolar disorders^b (n = 23)	16/23 (69.57%)	341/452 (75.44%)	0.525	20/23 (87.39%)	393/452 (86.96%)	0.951
Schizo-related disorders^c (n = 9)	8/9 (88.89%)	349/466 (74.89%)	0.336	8/9 (88.89%)	311/466 (66.67%)	0.059
Self-Medication Type
	Yes	No		Yes	No
Alcohol (n = 441)	334/441 (75.74%)	23/34 (67.65%)	0.293	384/441 (87.07%)	31/34 (91.18%)	0.354
Nicotine (n = 326)	249/326 (76.38%)	108/149 (72.48%)	0.362	286/326 (87.73%)	129/149 (86.58%)	0.726
Stimulants (n = 386)	297/386 (76.94%)	60/89 (67.42%)	0.061	338/386 (87.56%)	77/89 (86.52%)	0.788
Dissociatives (n = 87)	74/87 (85.06%)	283/388 (72.94%)	0.018	76/87 (87.36%)	339/388 (87.37%)	0.997
Psychedelics (n = 212)	169/212 (79.72%)	188/263 (71.48%)	0.039	188/212 (88.68%)	227/263 (86.31%)	0.440
Cannabis (n = 391)	303/391 (77.49%)	54/84 (64.29%)	0.011	343/391 (87.72%)	72/84 (85.71%)	0.615
Sedatives (n = 227)	185/227 (81.50%)	172/248 (69.35%)	0.002	199/227 (87.67%)	216/248 (87.10%)	0.852
Opioids (n = 171)	139/171 (81.29%)	218/304 (71.71%)	0.020	151/171 (88.30%)	264/304 (86.84%)	0.645
Novel psychoactives (n = 57)	44/57 (77.19%)	313/418 (74.88%)	0.705	52/57 (91.23%)	363/418 (86.84%)	0.350
Drug Frequency
Daily (n = 278)	207 (74.46%)		0.673	239 (85.91%)		0.408
A few times a week (n = 91)	71 (78.02%)			84 (92.31%)
Monthly (n = 17)	11 (64.71%)			17 (82.35%)
Infrequently or on special occasions (n = 89)	68 (76.40%)			78 (87.64%)
Specialist Treatment Service History
Yes (n = 66)	56 (84.85%)		0.05	62 (93.94%)		0.083
No (n = 409)	301 (73.59%)		0.05	353 (86.31%)		0.083

PTSD & CPTSD combined.

Bipolar I & Bipolar II combined.

Schizophrenia & Schizoaffective disorder combined.

Note: For rows comparing the presence vs. absence of a condition or substance use (e.g., OCD Yes vs. OCD No; Alcohol Yes vs. Alcohol No), the first value [(N/N) (%)] indicates the rate of Depression/Anxiety within the group possessing the characteristic (Yes group), while the second value [(N/N) (%)] indicates the rate within the comparison group lacking the characteristic (No group). The associated p value is from the Pearson Chi-Square test comparing these two specific rates. For Specialist Treatment, these Yes and No group rates are presented on separate lines, with the p value comparing them shown on the “Yes” line. For multicategory variables (Age, Gender, Ethnicity, Drug Frequency), the percentage shown is the rate of Depression/Anxiety within that specific category. The p value for these variables typically reflects the overall Chi-Square test comparing rates across all categories within that variable (e.g., comparing across all four Drug Frequency groups simultaneously), not a specific pairwise comparison. P values <.001 are reported as .000 for brevity.

ADHD: Attention-Deficit/Hyperactivity Disorder. The bold values indicate <0.05 significance.

For anxiety, we observed a significant association with obsessive-compulsive disorder (96.39%, p = 0.006), and a lower prevalence associated with PMDD (50%, p = 0.024). The relationship between having a history of specialist treatment service and anxiety was not statistically significant (93.94%, p = 0.083).

Our analysis revealed a significant association between depression and anxiety (χ² = 37.253, df = 1, p < 0.001), with moderate strength (Φ = .280, V = .280, p < 0.001; Contingency Coefficient = .270, p < 0.001).

Risk estimates highlighted an increased likelihood of experiencing depression among individuals with anxiety, with an odds ratio of 5.153 (95% CI [2.932, 9.057]). For those without anxiety, we found significantly raised odds of having depression (OR = 3.956, 95% CI [2.482, 6.307]).

Multivariate Analyses

Table 2 presents the binary logistic regression results for examining factors associated with depression. We found that anxiety was strongly associated with depression (β = 1.715, p < 0.001, OR = 5.557), while being female was associated with lower odds of depression (β = −.730, p = 0.004, OR = .482). The use of dissociatives (β = .877, p = 0.031, OR = 2.403) and sedatives (β = .603, p = 0.016, OR = 1.828) was associated with significantly higher odds of depression, whereas novel psychoactives showed a negative although nonsignificant association (β = −.805, p = 0.059, OR = .447). Chronic pain is also associated with depression, though this did not reach statistical significance (β = .420, p = 0.070, OR = 1.521). Our model for depression explained between 11.8% and 17.6% of the variance in depression status (−2 Log likelihood: 472.675, Cox & Snell R Square: .118, Nagelkerke R Square: .176). The Hosmer and Lemeshow test confirmed the model's good fit to the data (χ² = 3.213, df = 7, p = 0.865).

Table 2.

Binary logistic regression results for depression and anxiety.

Depression	B	SE	Wald χ²	df	p	OR	95% CI for OR
Depression	B	SE	Wald χ²	df	p	OR	Lower	Upper
Disassociatives	.877	.407	4.636	1	.031	2.403	1.082	5.337
Sedatives	.603	.250	5.826	1	.016	1.828	1.120	2.984
Novel psychoactives	−.805	.427	3.562	1	.059	.447	.194	1.031
Chronic pain	.420	.232	3.272	1	.070	1.521	.966	2.397
Anxiety	1.715	.303	32.089	1	<.001	5.557	3.070	10.059
Female	−.730	.251	8.432	1	.004	.482	.294	.789

Anxiety	B	SE	Wald χ²	df	p	OR	95% CI for OR
Anxiety	B	SE	Wald χ²	df	p	OR	Lower	Upper
Depression	1.640	.292	31.533	1	<.001	5.157	2.909	9.143
ADHD	.765	.323	5.604	1	.018	2.149	1.141	4.048
Eating disorder	.538	.297	3.284	1	.070	1.712	.957	3.062

Note: ADHD: Attention-Deficit/Hyperactivity Disorder; B: Unstandardized regression coefficient (log-odds); SE: Standard Error; Wald χ²: Wald Chi-Square statistic; df: degrees of freedom; OR: Odds Ratio; CI: Confidence Interval. P values <.001 reported as .000 for brevity.

We also found that depression (β = 1.640, p < 0.001, OR = 5.157) was significantly associated with higher odds of anxiety. Attention-Deficit/Hyperactivity Disorder (β = .765, p = 0.018, OR = 2.149) was also significantly associated with anxiety. Eating disorders showed a weaker association that did not reach statistical significance (β = .538, p = 0.070, OR = 1.712). The model summary for examining factors associated with anxiety explained between 8.3% and 15.6% of the variance in anxiety status (−2 Log likelihood: 319.112, Cox & Snell R Square: .083, Nagelkerke R Square: .156). The Hosmer and Lemeshow test demonstrated a satisfactory fit (χ² = 2.887, df = 5, p = 0.717).

To aid interpretation of the findings, we developed a summary diagram (see Figure 1) which visually maps the variables found to be statistically associated with depression and/or anxiety in the two separate exploratory logistic regression models. While the figure resembles a conceptual model, it should not be interpreted as a formal path analysis or mediation model. The arrows reflect directional structure only as defined by the dependent variables in each regression model, and not causal or mediational relationships. Furthermore, the model does not account for interrelationships among predictor variables, many of which are likely correlated.

Discussion

Our study reveals significant associations between depression, anxiety, and the use of specific substances among self-medicating autistic adults. We found that the use of dissociatives and sedatives was associated with increased odds of depression, while the use of novel psychoactives reduced these odds. These findings suggest that autistic individuals may not use certain substances as coping mechanisms to alleviate depression but these substances may also contribute to the development or exacerbation of depressive symptoms. This aligns with previous research that has identified the potential bidirectional relationship between substance use and mental health conditions (Pacek et al., 2013; Swendsen et al., 2010).

The analysis also revealed that having an eating disorder or being ADHD was associated with increased odds of anxiety. This finding aligns with previous studies that have reported high comorbidity rates between anxiety disorders and eating disorders (Swinbourne & Touyz, 2007) as well as between anxiety disorders and ADHD (Kessler et al., 2006). It also demonstrates that these co-occurring phenomena are particularly influential among self-medicating autistic adults, as we identified the presence of eating disorders and ADHD as key factors independently associated with anxiety in our multivariate model. These results suggest that autistic individuals who engage in self-medication practices may experience a particularly pronounced complex interplay between anxiety, eating disorders, and ADHD.

We also found a strong interrelationship between anxiety and depression, with anxiety serving as a key factor independently associated with depression and in our model vice versa. This finding is consistent with previous research on autistic populations, which has consistently reported high rates of co-occurrence between these two mental health conditions (Hollocks et al., 2019; Hudson et al., 2019). The strength of this interrelationship in our sample may be particularly pronounced due to the added complexity of self-medication practices, which can both alleviate and exacerbate mental health problems.

Alexithymia, a neurological phenomenon that impacts how individuals identify and describe their own emotions, may also contribute to the interrelationship between anxiety and depression we observed in this study. Researchers have shown that alexithymia is more prevalent among autistic individuals compared to the general population (Kinnaird et al., 2019) and is associated with an increased risk of developing both anxiety and depression (Dikmen et al., 2020; Oakley et al., 2022). In the context of self-medication, individuals with alexithymia may struggle to recognise and express their emotions, leading them to rely on substances as a means of coping with emotional distress. This, in turn, could exacerbate their anxiety and depression, creating a vicious cycle that further reinforces the interrelationship between these two mental health conditions. Furthermore, individuals with alexithymia may find the use of substances to provide controlled interoceptive sensations particularly appealing, as it offers a tangible way to experience and regulate their emotions (Brewer et al., 2016). This highlights the importance of considering alexithymia when researchers and clinicians develop interventions and support strategies for autistic individuals who engage in self-medication practices.

While we observed a high prevalence of depression across all gender groups, autistic females overall had significantly lower rates of depression. This difference remained significant even after we controlled for other factors in the multivariate analysis, suggesting that being female is an independent protective factor against depression among autistic adults who engage in self-medication. This finding differs from previous research on general autistic populations where researchers have found higher rates of depression among autistic females compared to males (Lever & Geurts, 2016). However, on a broader level, previous research has highlighted that suicide is the leading cause of death among men under 50 in the UK (Office for National Statistics, 2020) and that transgender and nonbinary individuals face a significant risk of suicidal ideation and attempted suicide (Bailey et al., 2014; Stonewall, 2018). Researchers need to further explore the relationship between gender and depression among self-medicating autistic people to better understand and address the unique challenges faced by this population.

Our study also revealed that chronic pain affected 63.79% of the sample, far exceeding the 20–35% rates researchers have previously identified among the general autistic population (Huguet & Miro, 2008; King et al., 2011; Palermo et al., 2009) as well as the wider general population (Jones & Shivamurthy, 2022). We also found that chronic pain was strongly associated with depression, suggesting that clinicians and researchers should consider addressing chronic pain and its management as a key consideration when developing interventions and support strategies for autistic individuals who engage in self-medication.

In terms of the prevalence of mental health problems among self-medicating autistic adults, we found that 87.37% of participants experienced anxiety, 75.16% experienced depression, and 51.16% had an eating disorder. These rates are substantially higher than those researchers have reported in the general autistic population, where the prevalence of anxiety ranges from 20% to 56% (Howlin & Magiati, 2017; Lai et al., 2019; Mazzone et al., 2012; Spain et al., 2016), depression affects 11% to 70% (Howlin and Magiati, 2017, Lever and Geurts, 2016, Lugnegard et al., 2011), and eating disorders are present in approximately 7.9% (Karjalainen et al., 2016) to 23% (Huke et al., 2013). The disparities are even more pronounced when compared to the general population, with global prevalence estimates of 4% for anxiety, 5% for depression (World Health Organisation, 2024a, 2024b), and 0.2% for eating disorders (Castaldelli-Maia & Bhugra, 2022). Castaldelli-Maia & Bhugra (2022) also reported that the worldwide general prevalence of bipolar disorders and schizophrenia to be 0.5% and 0.3%, respectively, which compares to 4.84% and 1.89%, respectively, in our study. Lai et al. (2019) have previously reported bipolar and schizo-related disorders to be present among 5% and 4% of autistic people, respectively.

A particularly striking finding is that a substantial proportion of the participants (80.42%) identified as ADHD. This far exceeds that of the general autistic population, where ADHD co-exists in 28% of autistic individuals (Lai et al., 2019). The high proportion of autistic and ADHD (AuDHD) individuals in our study strongly indicates that this population may be particularly vulnerable to self-medication. This supports the findings of Sizoo et al. (2010) who identified higher rates of substance misuse among AuDHD individuals than compared to autistic people alone, as well as Wilens et al. (2011) who identified ADHD as significantly associated with the development of substance use problems among young ADHD adults. This is also particularly concerning given that professionals may easily overlook or misunderstand this population, since the dual diagnosis of autism and ADHD is a relatively recent development and listed as mutually exclusive prior to the DSM-V (American Psychiatric Association, 2013).

Regarding the types of substances used by self-medicating autistic adults, our findings revealed extensive use of various substances. Participants most commonly reported using alcohol (92.84%), cannabis (82.32%), and stimulants (81.26%). More concerningly, 77.69% of participants reported self-medication frequency as daily (58.53%) or as several times a week (19.16%). While our survey did not directly assess substance dependency, such high frequency of use may indicate elevated risk for problematic patterns of use and associated harms. Specifically, researchers estimate the total prevalence of alcohol use disorders to be 5.1% (95% CI: 4.9–5.4%) (Rehm & Shield, 2019), cannabis use disorders to be present globally among 22% (95% CI: 18–26%) (Leung et al., 2020) and stimulants use disorders—among the US general population—to be between 0.2% and 2.1% (Compton et al., 2018). The high prevalence of substance use in our study's sample suggests that autistic adults who engage in self-medication may be at a significantly increased risk of developing substance misuse compared to the general population. The high frequency of use also raises concerns about the potential long-term consequences on physical and mental health. Previous research has shown that chronic substance use can lead to adverse health outcomes, such as increased risk of cardiovascular disease, liver damage, and cognitive impairment (Rehm et al., 2017; Schulte & Hser, 2013), ultimately increasing the risk of death (Esser et al., 2022).

Our findings have significant implications for policy and practice. The high prevalence of mental health conditions and substance use among autistic adults who self-medicate underscores the urgent need for targeted interventions and support services. The data-informed conceptual model illustrating factors associated with depression and anxiety (Figure 1), while based on exploratory findings from a single study and not a validated predictive tool, represents a preliminary step towards understanding the complex interplay of factors identified. It may serve to highlight potential areas of focus and generate hypotheses for policymakers and practitioners to consider in the development and evaluation of interventions, service planning, and clinical practice, subject to further research and validation. Policymakers and practitioners must review their practices to ensure they provide adequate support and do not inadvertently contribute to these problems. Long wait times and suboptimal mental health support quality may increase risks (Camm-Crosbie et al., 2019; Weir et al., 2021). Additionally, autistic people may self-medicate due to a lack of support and the neuronormative approach of services, coupled with diagnostic overshadowing. Organisations should re-evaluate the most effective ways to support autistic individuals’ mental health, including promoting autistic-safe community spaces that build social capital and reduce loneliness, and prioritising autistic-led one-to-one interventions over generic neurotypical-led therapies that may be inappropriate, risk diagnostic overshadowing, and contribute to internalised ableism, thus increasing the risk of poor mental health and harmful self-medication.

Addressing the broader context of our findings, factors such as societal stigma and discrimination surrounding autism may contribute to the high rates of mental health difficulties observed in our sample. Evidence suggests stigma negatively impacts autistic mental health (Han et al., 2022; Papadopoulos et al., 2019; Papadopoulos & Prew, 2024) and can increase anxiety (Cage et al., 2018; Lei et al., 2024; Pearson & Rose, 2021)—a condition we found to be highly prevalent and strongly associated with depression among participants. Furthermore, experiences of stigma or services lacking neurodiversity-affirming practices may deter autistic individuals from seeking professional support (Thompson-Hodgetts et al., 2020), potentially increasing reliance on self-medication. Therefore, efforts to reduce stigma and improve the cultural competence of services could be crucial in mitigating mental health challenges and harmful self-medication, though further research is needed to confirm these pathways.

Current national policy documents often overlook the importance of addressing substance misuse among autistic people, and our study contributes preliminary insights into this underresearched issue. Based on our findings, we suggest that healthcare professionals—particularly those prescribing medication—consider adopting approaches that promote a sense of safety and open communication. Establishing therapeutic, trusting relationships may increase the likelihood of disclosure around neurodivergent identity, mental health difficulties, and substance use, potentially enabling more supportive responses and harm reduction. These findings also point to the possible value of developing specialist services with staff trained in autism-informed care. Involving autistic individuals with lived experience of self-medication in service design and delivery could be beneficial, though further research is needed to evaluate this. These considerations may also apply to substance misuse treatment settings, including inpatient environments, which previous studies suggest are often ill-equipped to support autistic individuals effectively (Sakai et al., 2014; Strauss et al., 2019).

Future researchers should seek to develop a more nuanced understanding of the reasons underpinning self-medication practices among autistic individuals. The data-informed conceptual model we present offers a foundation for generating hypotheses and exploring potential causal pathways in future work. The potential role of alexithymia as well as ADHD in the context of self-medication would also benefit from further research attention given the high prevalence of AuDHD individuals engaging in self-medication in our study, suggesting a potential increased vulnerability. Such research could inform the development of targeted interventions, particularly those delivered by service providers, to better support autistic individuals who are struggling with substance misuse. Furthermore, our study focused on autistic adults aged 18–50+, highlighting the need for future research to investigate self-medication practices among younger and older autistic populations. Exploring the unique challenges and experiences of these age groups could yield important findings that inform age-appropriate interventions and support strategies.

Our study has several limitations that readers should consider when interpreting the findings. Firstly, the cross-sectional design precludes the establishment of causal relationships between self-medication practices and mental health outcomes. Secondly, we relied on self-reported data, collected using a study-specific questionnaire, potentially affecting the reliability and validity of the results. Additionally, we did not use validated psychometric scales to assess the severity of mental health problems. This was a deliberate choice primarily to reduce participant burden and because our aim was to identify prevalence based on participant identified conditions rather than clinical severity ratings. While some psychometric scales tailored for autistic adults are available (e.g., the Anxiety Scale for Autism–Adults [ASA-A; Rodgers et al., 2020] and the Autistic Depression Assessment Tool – Adult [ADAT-A; Cassidy et al., 2021]), their use was beyond the scope of our prevalence-focused, community-led design. Future research, particularly studies aiming to measure severity or clinical outcomes, should consider incorporating such appropriately validated standardised questionnaires.

Thirdly, the exceptionally high prevalence of self-reported ADHD (80.42%) in our sample requires significant caution when interpreting all findings and limits generalisability. This rate suggests our sample predominantly represents autistic adults with co-occurring ADHD (AuDHD), a group known to have higher rates of substance use independent of being autistic (Wilens et al., 2011). Therefore, ADHD acts as a major potential confound in our analyses. It was not possible with our design to disentangle whether the observed self-medication patterns, prevalence rates of mental health difficulties, and factors identified as statistically significant (including ADHD being associated with anxiety in the model) are driven primarily by being autistic, ADHD, or the specific interaction of the two within this self-medicating group. Consequently, our findings, while potentially very relevant to AuDHD individuals who self-medicate, cannot be reliably generalised to the broader population of self-medicating autistic adults, particularly those who do not also identify as ADHD, and all interpretations of these statistical associations should be viewed through the lens of this significant potential confound.

Fourthly, the large number of univariate chi-square tests conducted raises the possibility of type I error inflation. Although we opted not to apply formal correction procedures such as Bonferroni due to the exploratory aims of the study, this decision may have increased the likelihood of false-positive findings.

Fifthly, our data collection did not differentiate between participants who held a formal autism diagnosis and those who self-identified, preventing analysis of potential differences between these groups. Sixthly, we recruited the sample through nonrandomised methods via online platforms and existing networks, which further limit generalisability to the broader autistic population and likely introduced sampling bias. The sample was predominantly female, White, and aged between 26 and 50 years, suggesting an underrepresentation of certain demographic groups, particularly in terms of ethnicity and potentially other intersecting identities. Finally, as previously noted, the online, text-based questionnaire format likely excluded nonspeaking autistic individuals or those requiring significant communication support, further impacting the representativeness of the sample.

Footnotes

ORCID iD

Chris Papadopoulos

Ethical Considerations

The study was approved by the University of Bedfordshire’s Institute for Health Research Ethics Committee. The procedures used in this study adhere to the tenets of the Declaration of Helsinki.

Consent to Participate

Informed consent was obtained from all individual participants included in the study.

Author Contributions

CP led the manuscript writing, performed the data analysis, and was primarily responsible for data interpretation. TA and KM contributed equally to the conceptualisation of the study and the interpretation of data, and provided critical revisions for important intellectual content. DGH conceived and designed the study, led the data collection, and significantly contributed to the interpretation of data and manuscript writing. All authors collectively ensured the accuracy and integrity of the work, and all authors approved the final manuscript for publication.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

The datasets generated and analysed during the current study are not publicly available due to ongoing research and future publications but are available from the corresponding author on reasonable request.

References

American Psychiatric Association. (2013). Diagnostic and Statistical Manual of Mental Disorders (5th ed). https://doi.org/10.1176/appi.books.9780890425596

Bailey

Ellis

S. J.

McNeil

(2014). Suicide risk in the UK trans population and the role of gender transition in decreasing suicidal ideation and suicide attempt. Mental Health Review Journal, 19(4), 209–220. https://doi.org/10.1108/MHRJ-05-2014-0015

Blanchard

Kalstrom

Ahlich

Wilens

Biederman

(2021). Risk of self-harm in children and adults with autism spectrum disorder: A systematic review and meta-analysis. JAMA Network Open, 4(12), e2139272. https://doi.org/10.1001/jamanetworkopen.2021.39272

Brewer

Cook

Bird

(2016). Alexithymia: A general deficit of interoception. Royal Society Open Science, 3(10), 150664. https://doi.org/10.1098/rsos.150664

Butwicka

Långström

Larsson

, et al. (2017). Increased risk for substance use-related problems in autism spectrum disorders: A population-based cohort study. Journal of Autism and Developmental Disorders, 47(1), 80–89. https://doi.org/10.1007/s10803-016-2914-2

Cage

Di Monaco

Newell

(2018). Experiences of autism acceptance and mental health in autistic adults. Journal of Autism and Developmental Disorders, 48(2), 473–484. https://doi.org/10.1007/s10803-017-3342-7

Camm-Crosbie

Bradley

Shaw

Baron-Cohen

Cassidy

(2019). ‘People like me don't get support’: Autistic adults’ experiences of support and treatment for mental health difficulties, self-injury and suicidality. Autism, 23(6), 1431–1441. https://doi.org/10.1177/1362361318816053

Cassidy

Au-Yeung

Robertson

Cogger-Ward

Richards

Allison

Bradley

Kenny

O'Connor

Mosse

Rodgers

Baron-Cohen

(2022). Autism and autistic traits in those who died by suicide in England. The British Journal of Psychiatry : The Journal of Mental Science, 1–9. Advance online publication. https://doi.org/10.1192/bjp.2022.21

Cassidy

Bradley

Cogger-Ward

Graham

Rodgers

(2021). Development and validation of the Autistic Depression Assessment Tool–Adult (ADAT-A) in autistic adults. Research Square. https://doi.org/10.21203/rs.3.rs-358997/v1

10.

Cassidy

Bradley

Robinson

, et al. (2014). Suicidal ideation and suicide plans or attempts in adults with asperger's syndrome attending a specialist diagnostic clinic: A clinical cohort study. The Lancet Psychiatry, 1(2), 142–147. https://doi.org/10.1016/S2215-0366(14)70248-2

11.

Cassidy

Bradley

Shaw

Baron-Cohen

(2018). Risk markers for suicidality in autistic adults. Molecular Autism, 9, Article 42. https://doi.org/10.1186/s13229-018-0226-4

12.

Castaldelli-Maia

J. M.

Bhugra

(2022). Analysis of global prevalence of mental and substance use disorders within countries: Focus on sociodemographic characteristics and income levels. International Review of Psychiatry, 34(1), 6–15. https://doi.org/10.1080/09540261.2022.2040450

13.

Compton

W. M.

Han

Blanco

Johnson

Jones

C. M.

(2018). Prevalence and correlates of prescription stimulant use, misuse, use disorders, and motivations for misuse among adults in the United States. The American Journal of Psychiatry, 175(8), 741–755. https://doi.org/10.1176/appi.ajp.2018.17091048

14.

Dikmen

P. Y.

Aysevener

E. O.

Kosak

Aydınlar

E. I.

Kocaman

A. S.

(2020). Relationship between MIDAS, depression, anxiety and alexithymia in migraine patients. Acta Neurologica Belgica, 120(4), 837–844. https://doi.org/10.1007/s13760-017-0856-x

15.

Esser

M. B.

Leung

Sherk

, et al. (2022). Estimated deaths attributable to excessive alcohol use among US adults aged 20 to 64 years, 2015 to 2019. JAMA Network Open, 5(12), e2239485. https://doi.org/10.1001/jamanetworkopen.2022.39485

16.

Feingold

Tzur Bitan

(2022). Addiction psychotherapy: Going beyond self-medication. Frontiers in Psychiatry, 13, 820660. https://doi.org/10.3389/fpsyt.2022.820660

17.

Han

Scior

Avramides

Crane

(2022). A systematic review on autistic people's experiences of stigma and coping strategies. Autism Research, 15(1), 12–26. https://doi.org/10.1002/aur.2652

18.

Hollocks

M. J.

Lerh

J. W.

Magiati

Meiser-Stedman

Brugha

T. S.

(2019). Anxiety and depression in adults with autism spectrum disorder: A systematic review and meta-analysis. Psychological Medicine, 49(4), 559–572. https://doi.org/10.1017/S0033291718002283

19.

Hosmer

D. W.

Lemeshow

Sturdivant

R. X.

(2013). Applied logistic regression (3rd ed). Wiley.

20.

Howlin

Magiati

(2017). Autism spectrum disorder: Outcomes in adulthood. Current Opinion in Psychiatry, 30(2), 69–76. https://doi.org/10.1097/YCO.0000000000000308

21.

Hudson

C. C.

Hall

Harkness

K. L.

(2019). Prevalence of depressive disorders in individuals with autism spectrum disorder: A meta-analysis. Journal of Abnormal Child Psychology, 47(1), 165–175. https://doi.org/10.1007/s10802-018-0402-1

22.

Huguet

Miró

(2008). The severity of chronic pediatric pain: An epidemiological study. The Journal of Pain, 9(3), 226–236. https://doi.org/10.1016/j.jpain.2007.10.015

23.

Huke

Turk

Saeidi

Kent

Morgan

J. F.

(2013). Autism spectrum disorders in eating disorder populations: A systematic review. European Eating Disorders Review, 21(5), 345–351. https://doi.org/10.1002/erv.2244

24.

IBM Corp. (2020). IBM SPSS Statistics for Windows (Version 26.0) [Computer software]. IBM Corp.

25.

Jones

Shivamurthy

(2022). OA02 Paediatric chronic pain and autism spectrum disorder: Is there a link? Rheumatology Advances in Practice, 6(Supplement_1), rkac066-002. https://doi.org/10.1093/rap/rkac066.002

26.

Karjalainen

Gillberg

Råstam

Wentz

(2016). Eating disorders and eating pathology in young adult and adult patients with ESSENCE. Comprehensive Psychiatry, 66, 79–86. https://doi.org/10.1016/j.comppsych.2015.12.009

27.

Kessler

R. C.

Adler

Barkley

, et al. (2006). The prevalence and correlates of adult ADHD in the United States: Results from the national comorbidity survey replication. The American Journal of Psychiatry, 163(4), 716–723. https://doi.org/10.1176/ajp.2006.163.4.716

28.

King

Chambers

C. T.

Huguet

, et al. (2011). The epidemiology of chronic pain in children and adolescents revisited: A systematic review. Pain, 152(12), 2729–2738. https://doi.org/10.1016/j.pain.2011.07.016

29.

Kinnaird

Stewart

Tchanturia

(2019). Investigating alexithymia in autism: A systematic review and meta-analysis. European Psychiatry, 55, 80–89. https://doi.org/10.1016/j.eurpsy.2018.09.004

30.

Lai

M. C.

Kassee

Besney

, et al. (2019). Prevalence of co-occurring mental health diagnoses in the autism population: A systematic review and meta-analysis. The Lancet Psychiatry, 6(10), 819–829. https://doi.org/10.1016/S2215-0366(19)30289-5

31.

Lei

Leigh

Charman

Russell

Hollocks

M. J.

(2024). Understanding the relationship between social camouflaging in autism and safety behaviours in social anxiety in autistic and non-autistic adolescents. Journal of Child Psychology and Psychiatry, 65(3), 285–297. https://doi.org/10.1111/jcpp.13884

32.

Leung

Chan

G. C. K.

Hides

Hall

W. D.

(2020). What is the prevalence and risk of cannabis use disorders among people who use cannabis? A systematic review and meta-analysis. Addictive Behaviors, 109, 106479. https://doi.org/10.1016/j.addbeh.2020.106479

33.

Lever

A. G.

Geurts

H. M.

(2016). Psychiatric co-occurring symptoms and disorders in young, middle-aged, and older adults with autism spectrum disorder. Journal of Autism and Developmental Disorders, 46(6), 1916–1930. https://doi.org/10.1007/s10803-016-2722-8

34.

Lugnegard

Hallerback

M. U.

Gillberg

(2011). Psychiatric comorbidity in young adults with a clinical diagnosis of asperger syndrome. Research in Developmental Disabilities, 32(5), 1910–1917. https://doi.org/10.1016/j.ridd.2011.03.025

35.

Mazzone

Ruta

Reale

(2012). Psychiatric comorbidities in asperger syndrome and high functioning autism: Diagnostic challenges. Annals of General Psychiatry, 11(1), 16. https://doi.org/10.1186/1744-859X-11-16

36.

Nock

M. K.

Borges

Bromet

E. J.

, et al. (2008). Cross-national prevalence and risk factors for suicidal ideation, plans and attempts. The British Journal of Psychiatry, 192(2), 98–105. https://doi.org/10.1192/bjp.bp.107.040113

37.

Oakley

B. F. M.

Jones

E. J. H.

Crawley

, et al. (2022). Alexithymia in autism: Cross-sectional and longitudinal associations with social-communication difficulties, anxiety and depression symptoms. Psychological Medicine, 52(8), 1458–1470. https://doi.org/10.1017/S0033291720003244

38.

Office for National Statistics. (2020). Suicides in England and Wales: 2019 registrations. https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/deaths/bulletins/suicidesintheunitedkingdom/2019registrations. Accessed June 28, 2024.

39.

Oomen

Nijhof

A. D.

Wiersema

J. R.

(2021). The psychological impact of the COVID-19 pandemic on adults with autism: A survey study across three countries. Molecular Autism, 12(1), 21. https://doi.org/10.1186/s13229-021-00424-y

40.

Pacek

L. R.

Martins

S. S.

Crum

R. M.

(2013). The bidirectional relationships between alcohol, cannabis, co-occurring alcohol and cannabis use disorders with major depressive disorder: Results from a national sample. Journal of Affective Disorders, 148(2-3), 188–195. https://doi.org/10.1016/j.jad.2012.11.059

41.

Palermo

T. M.

Eccleston

Lewandowski

A. S.

Williams

A. C. D. C.

Morley

(2009). Randomized controlled trials of psychological therapies for management of chronic pain in children and adolescents: An updated meta-analytic review. Pain, 148(3), 387–397. https://doi.org/10.1016/j.pain.2009.10.004

42.

Papadopoulos

Lodder

Constantinou

Randhawa

(2019). Systematic review of the relationship between autism stigma and informal caregiver mental health. Journal of Autism and Developmental Disorders, 49(4), 1665–1685. https://doi.org/10.1007/s10803-018-3835-z

43.

Papadopoulos

Prew

(2024). What should NHS postdiagnosis support services for parents/carers of young autistic people look like? recommendations based on a new theoretical model constructed following an evidence synthesis on the impact of autism stigma on parent/carer mental health. Commissioned report for NHS England & NHS Improvement. https://doi.org/10.17605/OSF.IO/RF45E

44.

Pearson

Rose

(2021). A conceptual analysis of autistic masking: Understanding the narrative of stigma and the illusion of choice. Autism in Adulthood, 3(1), 52–60. https://doi.org/10.1089/aut.2020.0043

45.

Pellicano

den Houting

Crane

Heyworth

Hill

Hudson

(2022). “I want to see my friends”: The everyday experiences of autistic people and their families during COVID-19. Autism, 26(1), 132–146. https://doi.org/10.1177/13623613211035936

46.

Rehm

Gmel

G. E.

Gmel

, et al. (2017). The relationship between different dimensions of alcohol use and the burden of disease—an update. Addiction, 112(6), 968–1001. https://doi.org/10.1111/add.13757

47.

Rehm

Shield

K. D.

(2019). Global burden of alcohol use disorders and alcohol liver disease. Biomedicines, 7(4), 99. https://doi.org/10.3390/biomedicines7040099

48.

Rodgers

Farquhar

Mason

Brice

Wigham

Ingham

Freeston

M. H.

Parr

J. R.

(2020). Development and initial evaluation of the Anxiety Scale for Autism–Adults (ASA-A). Autism in Adulthood, 2(1), 34–42. https://doi.org/10.1089/aut.2019.0044

49.

Sakai

Miller

Brussa

A. K.

MacPherson

Augustyn

(2014). Challenges of autism in the inpatient setting. Journal of Developmental & Behavioral Pediatrics, 35(1), 82–84. https://doi.org/10.1097/DBP.0000000000000014

50.

Schulte

M. T.

Hser

Y. I.

(2013). Substance use and associated health conditions throughout the lifespan. Public Health Reviews, 35, 3. https://doi.org/10.1007/BF03391702

51.

Sizoo

van den Brink

Koeter

, et al. (2010). Treatment seeking adults with autism or ADHD and co-morbid substance use disorder: Prevalence, risk factors and functional disability. Drug and Alcohol Dependence, 107(1), 44–50. https://doi.org/10.1016/j.drugalcdep.2009.09.003

52.

Spain

Happé

Johnston

, et al. (2016). Social anxiety in adult males with autism spectrum disorders. Research in Autism Spectrum Disorders, 32, 13–23. https://doi.org/10.1016/j.rasd.2016.08.002

53.

Stonewall. (2018). LGBT in Britain: health report. https://www.stonewall.org.uk/system/files/lgbt_in_britain_-_trans_report_final.pdf. Accessed June 28, 2024.

54.

Straus

Coburn

Maskell

Pappagianopoulos

Cantrell

(2019). Medical encounters for youth with autism spectrum disorder: A comprehensive review of environmental considerations and interventions. Clinical Medicine Insights: Pediatrics, 13, 1179556519842816. https://doi.org/10.1177/1179556519842816

55.

Swendsen

Conway

K. P.

Degenhardt

, et al. (2010). Mental disorders as risk factors for substance use, abuse and dependence: Results from the 10-year follow-up of the national comorbidity survey. Addiction, 105(6), 1117–1128. https://doi.org/10.1111/j.1360-0443.2010.02902.x

56.

Swinbourne

J. M.

Touyz

S. W.

(2007). The co-morbidity of eating disorders and anxiety disorders: A review. European Eating Disorders Review, 15(4), 253–274. https://doi.org/10.1002/erv.784

57.

Thompson-Hodgetts

Labonte

Mazumder

Phelan

(2020). Helpful or harmful? A scoping review of perceptions and outcomes of autism diagnostic disclosure to others. Research in Autism Spectrum Disorders, 77, 101598. https://doi.org/10.1016/j.rasd.2020.101598

58.

Weir

Allison

Baron-Cohen

(2021). Understanding the substance use of autistic adolescents and adults: A mixed-methods approach. The Lancet Psychiatry, 8(7), 546–554. https://doi.org/10.1016/S2215-0366(21)00160-7

59.

Wilens

T. E.

Martelon

Joshi

, et al. (2011). Does ADHD predict substance-use disorders? A 10-year follow-up study of young adults with ADHD. Journal of the American Academy of Child and Adolescent Psychiatry, 50(6), 543–553. https://doi.org/10.1016/j.jaac.2011.01.021

60.

World Health Organisation. (2024a). Depressive disorder (depression). https://www.who.int/en/news-room/fact-sheets/detail/depressive-disorder-(depression). Accessed June 28, 2024.

61.

World Health Organisation. (2024b). Anxiety disorders. https://www.who.int/en/news-room/fact-sheets/detail/anxiety-disorders. Accessed June 28, 2024.

62.

Zulauf

C. A.

Sprich

S. E.

Safren

S. A.

Wilens

T. E.

(2014). The complicated relationship between attention deficit/hyperactivity disorder and substance use disorders. Current Psychiatry Reports, 16(3), 436. https://doi.org/https://https://doi.org/10.1007/s11920-013-0436-6

Predictors of Depression and Anxiety Among Self-Medicating Autistic Adults

Abstract

Background:

Methods:

Results:

Conclusion:

Lay Abstract

Keywords

Introduction

Method

Study Design

Sampling and Recruitment Procedure

Measures

Ethics

Analysis

Results

Participant Characteristics

Mental Health, ADHD, and Chronic Pain Prevalence

Self-Medication Type and Prevalence

Univariate Analyses

Multivariate Analyses

Discussion

Footnotes

ORCID iD

Ethical Considerations

Consent to Participate

Author Contributions

Funding

Declaration of Conflicting Interests

Data Availability Statement

References