Autism Diagnoses Rose Faster Among Girls and Women in One Health System

This retrospective cohort study looked back at electronic health records from a large US health care system. It included 171,134 people who had at least one clinical visit every year from 2016 through 2024. That requirement meant the researchers could follow diagnosis patterns over time among people consistently connected to the system.

Participants included children ages 0 to 9, adolescents ages 10 to 18, and adults ages 19 and older. The average participant age was 26.78 years, and 63.8% of participants were female.

The researchers counted newly recorded autism spectrum disorder diagnoses each year. Autism spectrum disorder, often called autism, is a broad diagnostic term used for differences in social communication and patterns of behavior or interests. The study also included an older diagnostic code for Asperger disorder, a label no longer used as a separate diagnosis in current US diagnostic manuals.

Overall, females received an autism diagnosis 3.4 years later on average than males. The average diagnosis age was 15.7 years for females and 12.3 years for males. This difference was statistically significant, meaning it was unlikely to be explained by chance alone in this group of records.

The clearest change was seen after 2020. Among females, the annual rate of newly recorded autism diagnoses rose by 19.0% after 2020. The researchers estimated that the true yearly change could plausibly range from 12.1% to 33.4%.

The pattern differed by age group. Among children, newly recorded diagnoses among males declined by 2.0% per year from 2016 to 2024. For female children, diagnoses rose after 2020 by 17.4% per year. Among adolescent males, diagnoses declined from 2016 to 2018 and then stabilized. Among adolescent females, diagnoses increased from 2020 to 2024 by 22.3% per year.

For adults, the picture was less certain. Newly recorded diagnoses showed a downward trend among males across the study period. For females, the estimated annual change from 2016 to 2024 was positive, at 8.0%, but it was not statistically significant. In other words, the available data did not provide strong enough evidence to rule out a chance finding for adult females.

The researchers used a statistical method designed to identify changes in trends over time. Their conclusion was cautious: the increase in newly recorded diagnoses among females, alongside stable or decreasing rates among males, suggested improved identification of autistic girls and women rather than a uniform rise across the population.

That interpretation is not proof. Health records show when a diagnosis was entered, not when a person’s autism-related traits began or exactly why a clinician made the diagnosis. The study also cannot separate possible changes in awareness, referral patterns, access to assessment, diagnostic practices, or health care use after the COVID-19 pandemic.

The findings come from one health care system and include only people who maintained yearly clinical contact. The results may not represent people with less regular access to care or communities with different health systems. The abstract also does not provide details about participants’ race, ethnicity, income, insurance, or other factors that could affect access to autism assessment.

Further research in other settings would be needed to see whether the same trend appears more broadly and to understand what is helping—or delaying—autism identification among girls and women. For now, the study adds evidence that patterns of diagnosis can differ by sex and age, and that newly recorded diagnoses are not the same thing as a change in the underlying number of autistic people.