Autism Tied to Higher Rates of Self-harm, Suicide


Even after accounting for sociodemographic factors, intellectual disabilities, and psychiatric diagnoses, autism is associated with an 83% increased risk of self-harm among females and a 47% increased risk among males.


Evidence shows those with autism have over threefold greater odds than their counterparts without the disorder of self-injurious behavior, suicidal ideation, suicide attempt, or suicide death, but reasons for these elevated risks are unclear.

Using various linked databases in the province of Ontario, Canada, researchers identified all individuals with an autism diagnosis from April 1, 1988, to March 31, 2018, and matched each on age and sex to four nonautistic individuals for the comparison group.

Investigators created two cohorts to separately evaluate outcomes of self-harm events leading to emergency healthcare and suicide death with the accrual period for both cohorts beginning at a person’s 10th birthday.

The self-harm cohort included 379,630 individuals while the suicide cohort included 334,690 individuals.


Over 15 years, autistic females showed the highest cumulative self-harm events, followed by autistic males, nonautistic females, and nonautistic males; over 25 years, autistic males had the highest cumulative incidence of suicide death, followed by autistic females, nonautistic males, and nonautistic females.

Autism had independent associations with self-harm events (females: relative rate [RR], 1.83 [95% CI, 1.61 – 2.08]; males: RR, 1.47 [95% CI, 1.28 – 1.69]) even after accounting for sociodemographic factors (varied directions of associations), intellectual disabilities (associated with increased risks), and psychiatric diagnoses including mood and anxiety, psychotic, addiction, and personality disorders (associated with increased risks).

For both females and males, final models showed autism per se was not significantly associated with suicide death, but certain correlates were linked to risk. Among both sexes, intellectual disabilities were associated with reduced risks and psychiatric diagnoses were associated with increased risks.

As a substantial proportion (28.4%) of the suicide cohort did not have data on self-harm, researchers were unable to examine the association of self-harm with suicide death.


That psychiatric diagnoses increased suicide risks among people with autism suggests supports to reduce such risks “should consider multifactorial mechanisms, with a particular focus on the prevention and timely treatment of psychiatric illnesses,” write the authors.


The study was conducted by Meng-Chuan Lai, MD, PhD, Centre for Addiction and Mental Health, Toronto, Ontario, Canada, and colleagues. It was published online August 8 in JAMA Network Open.


The autism cohort didn’t capture those diagnosed in private practices or with subtle presentations not yet diagnosed. Misclassification of autistic people in the nonautistic cohort may have resulted in underestimation of suicide-related outcomes. The administrative data don’t reliably identify diagnoses associated with suicide risks such as attention-deficit/hyperactivity disorder or subcategories of mood disorders, and don’t contain information about risk and protective mechanisms of suicide behaviors such as family history.


The study received support from ICES, an independent nonprofit research institute; the Innovation Fund of the Alternative Funding Plan for the Academic Health Sciences Centres of Ontario; the Academic Scholars Award from the Department of Psychiatry, University of Toronto; and the Canadian Institutes of Health Research Sex and Gender Science Chair. Lai reported receiving personal fees from SAGE Publications as an editorial honorarium outside the submitted work. Author Natasha R. Saunders, MD, Child Health Evaluative Sciences, SickKids Research Institute, Toronto, reported receiving honoraria from the BMJ Group, Archives of Diseases in Childhood. No other disclosures were reported.

For more Medscape Psychiatry news, join us on Twitter and Facebook.

Source: Read Full Article