Autism spectrum disorder (ASD) is a neurodevelopmental condition that affects social communication and behavior. The prevalence of ASD has increased markedly in recent decades, with estimates ranging from 1% to 3.3% in the western Europe and the U.S. ASD has historically shown a high male-to-female ratio (MFR), commonly attributed to diagnostic practices, biological theories, and delayed or missed diagnosis among females, who often exhibit comparatively stronger social and communication skills. Although the MFR has declined over time, few large population-based studies have simultaneously disentangled the effects of calendar period, age at diagnosis, and birth cohort, factors influenced by changing parental demographics and the broadening of diagnostic criteria. These temporal impacts on MFR in ASD patients were well explained in a recent study published in the BMJ using Swedish population data.
This population-based, prospective, birth cohort study included a total of 2,756,779 liveborn children identified from the Swedish medical birth register between 1985 and 2020. The primary endpoint was diagnosis of ASD identified from the Swedish national patient register till a maximum age of 37 years (median age = 20.1 years, interquartile range [IQR]: 8.8-29.4 years) between January 1, 1987, and December 31, 2022. The register covered all inpatient diagnoses data from 1987 onward and outpatient diagnoses from 2001-2005 onward. ASD incidence rates and MFRs were analyzed using Poisson age-period-cohort models with sex interactions. Incidence rates, cumulative ratios, 95% confidence interval (CI) were estimated for 2016, 2022, and 2024.
Among 2,756,779 individuals, 2.8% (n = 78,522) were diagnosed with ASD at the end of the follow-up period (2020). ASD diagnosis rates peaked during 2020-2022 across all age groups, ranging from 184.8 per 100,000 person-years in males and 45.8 among females aged 0-4 years to 654.5 among males aged 10-14 years and 602.6 in females aged 15-19 years. The median age at diagnosis was found to be 14.3 years (IQR: 9.9-17.6). The MFR was about 3.0 before age 10 years but declined with increasing age, reaching ≤1.0 after age 15 years in recent calendar periods and among post-2000 birth cohorts. ASD rates increased steadily from 1995 to 2022 across all age groups; rates rose from 19.8 to 117.2 at ages 0-4 years and from 56.5 to 564.1 at ages 10-14 years. MFRs remained 2-4 in younger children but declined with age. The highest incidence was observed among individuals aged 15-19 years in cohorts born after 1995.  Â
Models including calendar period, age at diagnosis, and birth cohort provided the best fit according to the Akaike information criterion, outperforming simpler age-period or age-cohort models. The best-fitting model showed a high MFR up to age 10 years, followed by a decline and plateau around age 15 years. Period effects increased steadily until 2010 and then declined. Cohort effects were stable until 1995, decreased until 2005, and increased again through 2022.
Supplementary analysis identified 43,146 individuals with autistic disorder. Incidence rates were similar to those observed for overall ASD during childhood and adolescence, aged from 0 to 19 years, but declined sharply after age 20 years. Overall MFR for autistic disorder was found to be 1.7 (95% CI: 1.7-1.7), indicating a smaller sex disparity than that observed for the broader ASD category.
This study’s limitations include the use of register-based data lacking detailed clinical characteristics and comorbidities, the exclusion of outpatient diagnoses before 2001, limited control for familial confounders, and partial reliance on projected estimates associated with greater uncertainty.
In conclusion, this long-term analysis of the Swedish population demonstrates a sustained increase in ASD diagnoses alongside a declining MFR over the past 35 years. By adulthood, the MFR may approach parity in Sweden. These findings highlight delayed diagnosis in women and girls and underscore the need for improved understanding of sex-specific ASD presentations, as well as enhanced screening and diagnostic practices.
Reference: Fyfe C, Winell H, Dougherty J, et al. Time trends in the male to female ratio for autism incidence: population-based, prospectively collected birth cohort study. BMJ. 2026;392:e084164. doi:10.1136/bmj-2025-084164
