Manitoba cohort shows broad psychiatric excess in autistic people across childhood to adulthood
Autistic children, adolescents, and adults in Manitoba carried a far heavier burden of diagnosed mental illness than their non-autistic peers, and the excess extended to attempted suicide. In a province-wide analysis using linked administrative health records from 2018 to 2022, autistic people had higher prevalence of nearly every psychiatric outcome examined, with particularly striking elevations for schizophrenia, ADHD, conduct disorder, and mood or anxiety disorders.
The findings matter because they come from routine health data rather than a clinic-based sample. That gives the results unusual reach: they reflect what a universal health system actually records across the life course, from children in elementary school to adults in midlife. The pattern also suggests that autism-related mental health risk is not confined to one developmental window. It is persistent, broad, and clinically consequential.
Autism was associated with an age- and sex-adjusted relative risk of 5.24 for ADHD, 5.96 for conduct disorder, 2.00 for mood or anxiety disorders, 9.82 for schizophrenia, and 3.14 for attempted suicide. Substance use disorder was the exception among the major psychiatric categories assessed, without the same excess prevalence seen in the other outcomes. In a study built on population records, those are not small deviations at the edges of the curve; they are large separations in the middle of routine care.
Psychiatric risk climbed across developmental stages
Investigators used a validated case definition to identify autistic individuals and compared them with all other Manitobans without autism. The cohort was split into three age bands: 6 to 12 years, 13 to 17 years, and 18 to 45 years. Across those strata, autistic people had significantly elevated relative risks for the mental health outcomes under study.
The span of conditions is clinically important. ADHD and conduct disorder point to early neurodevelopmental and behavioral complexity, while mood and anxiety disorders reflect internalizing symptoms that may become more visible as social demands increase with age. Schizophrenia, with its very high relative risk, signals the need for careful psychiatric surveillance rather than assuming all distress in autistic patients stems from autism itself. Suicide attempts sit at the most urgent end of the spectrum and demand direct screening, not indirect inference.
A diagnosis of autism did not narrow the clinical field; it widened it.
Sex-specific patterns point to different diagnostic burdens
Sex-stratified results showed different psychiatric profiles for autistic males and females. Among autistic males, risks were higher for ADHD, conduct disorder, and schizophrenia. Among autistic females, risks were higher for mood or anxiety disorders and attempted suicide.
Those differences are not just statistical curiosities. They suggest that the mental health presentation of autism may be filtered through sex-linked patterns of symptom expression, help-seeking, clinician recognition, and diagnostic bias. Females with autism are often underrecognized earlier in life, and internalizing symptoms may dominate the clinical picture by the time they reach care. In contrast, externalizing and psychotic-spectrum diagnoses may more readily surface in males through school or emergency pathways. The administrative record cannot resolve every mechanism, but it does show that risk is not distributed evenly.
Autistic females in the cohort numbered 1,377, while autistic males numbered 4,523. Even with different group sizes, the direction of association remained consistent: females showed greater relative risk for mood or anxiety disorders and suicide attempts, while males showed greater relative risk for ADHD, conduct disorder, and schizophrenia. In daily practice, those patterns argue for autism-informed assessment that does not rely on a single template of psychiatric presentation.
Suicide prevention must be built into autism care, not added later
The suicide finding deserves direct attention. Autistic people had a 3.14-fold higher adjusted risk of attempted suicide than non-autistic individuals, and autistic females had a 2.22-fold higher adjusted risk compared with their non-autistic female peers. Suicide prevention in this setting cannot depend on passive referral pathways or on the assumption that someone else has already asked the right questions.
The biological and clinical logic is straightforward. Autism often coexists with communication differences, sensory overload, social exclusion, and cumulative stress from navigating environments not built for neurodivergent people. When those pressures intersect with depression, anxiety, or psychosis, risk can escalate quickly. Administrative data cannot measure every stressor, but the healthcare system can still respond by screening earlier, documenting risk more consistently, and tailoring follow-up to communication needs and sensory profile.
Routine suicide assessment in autistic patients should be treated as core care, not as an optional extra reserved for crisis visits. The dataset in Manitoba shows why.
Universal health records reveal where systems are falling short
The study drew on provincial health records covering 2018 to 2022 and used regression models to estimate sex- and age-adjusted relative risks. The approach strengthens confidence in the results because it captures care recorded in real-world practice across an entire population. It also reveals a familiar systems problem: large risk differences can persist for years before care pathways fully adapt.
The scale of the non-autistic comparison group was large, with 317,073 males and 314,261 females. Against that backdrop, the autistic cohort stood out clearly enough to show broad psychiatric excess despite the use of adjustment for age and sex. This is exactly the kind of signal a universal healthcare database should detect, and exactly the kind of signal that should change screening behavior.
In practical terms, the findings support several concrete shifts in care. Primary care clinicians, child and adolescent psychiatrists, emergency clinicians, and developmental specialists need autism-informed mental health training. Screening should be routine, repeated, and adapted to the patient’s communication style. Follow-up should be faster when mood symptoms, psychotic symptoms, or self-harm emerge. Family members and caregivers also need to be included when appropriate, because risk detection in autism often depends on collateral history rather than spontaneous disclosure.
Why these findings should change clinical habits now
The most important message is not simply that mental illness is more common in autism. It is that the excess spans multiple diagnoses, multiple ages, and both sexes, with suicide attempts appearing as a serious and measurable outcome. The data argue against fragmented care in which autism is managed in one silo and mental health in another.
A more integrated model is necessary. Autism-informed mental health care should account for diagnostic overshadowing, where new psychiatric symptoms are mistakenly folded into autism itself. It should also account for the opposite problem, where autism is missed because a patient presents only with depression, anxiety, aggression, or suicidality. Both errors delay treatment.
The cohort cannot answer every causal question, and administrative records do not capture symptom severity, psychosocial stressors, or undiagnosed cases. Even so, the message from Manitoba is plain: autistic people face disproportionate psychiatric harm, and the risk is visible early enough in the life course to justify systematic intervention. The next clinical task is not more speculation about whether these disparities exist. The task is building services that are ready for them.