Males are diagnosed with autism significantly more often than females. The CDC’s most recent surveillance data, from 2022, found a prevalence of about 49 per 1,000 among boys compared to 14 per 1,000 among girls, a ratio of roughly 3.4 to 1. But that ratio is shrinking fast, and growing evidence suggests the true gap between sexes may be far smaller than decades of data have indicated.
The Current Numbers
Among children aged 4 and 8 in the United States, boys are about 3.4 times more likely than girls to have an autism diagnosis. That figure comes from the CDC’s Autism and Developmental Disabilities Monitoring Network, which tracks prevalence across 16 sites nationwide. In raw terms, nearly 1 in 20 boys and about 1 in 70 girls in the surveillance population had autism in 2022.
Those numbers, however, reflect who gets diagnosed, not necessarily who is autistic. A large population-based study published in The BMJ tracked every child born in Sweden over multiple birth cohorts and found that the male-to-female ratio drops dramatically with age. Among children diagnosed before age 10, the ratio hovered between 2 and 4, consistent with the CDC figures. But by age 20, after accounting for later diagnoses, the ratio fell to just 1.2 to 1 in 2022. Projections from that same study suggested the cumulative ratio could reach near-parity by 2024, meaning roughly equal lifetime rates of diagnosis for men and women.
Why Girls Get Diagnosed Later
Females are consistently diagnosed about 18 months later than males on average, and they are significantly more likely to receive what researchers call a “late diagnosis,” defined as age 13 or older. In one analysis of state-wide diagnostic records, about 21% of autistic females were diagnosed at 13 or later, compared to 14% of males. That 18-month delay isn’t trivial. It means years of missed support during critical developmental windows.
Several forces drive that delay. The standard diagnostic tools used to evaluate autism were originally developed using samples of predominantly white boys. At least one item on the most widely used clinical assessment, which scores repetitive hand movements, has been shown to be harder for clinicians to identify in girls. This means a girl and a boy with the same underlying level of autism severity can receive different scores on the same test.
Before finally receiving an autism diagnosis, many women first collect a series of other labels. In a study of autistic adults reflecting on their diagnostic histories, personality disorders were the most common perceived misdiagnosis, followed by anxiety disorders, mood disorders, chronic fatigue or burnout-related conditions, and ADHD. Women were specifically more likely than men to report being misdiagnosed with personality disorders, anxiety, and mood disorders. These conditions can genuinely co-occur with autism, which makes untangling them even harder, but the pattern points to a systemic tendency to interpret autistic traits in women through a different clinical lens.
How Autism Looks Different in Females
One of the biggest reasons girls slip through the diagnostic net is a behavior pattern called camouflaging, or masking. Autistic females tend to actively suppress their autistic traits in social settings at higher rates than males. This can look like forcing eye contact, rehearsing appropriate facial expressions, mimicking the social behavior of peers, or consciously playing a “character” in conversations to appear more typical.
Masking has a cost. Research has found that in autistic women, camouflaging is linked to dampened emotional expressiveness, particularly positive emotions. If a woman is intensely excited about a specific interest, for example, she may be more aware than her male counterparts that displaying that fascination would seem unusual, so she suppresses it. The cumulative effort of constant self-monitoring can lead to emotional exhaustion and reduced well-being, which may partly explain the high rates of anxiety and burnout diagnoses that precede an autism diagnosis in women.
Social expectations also play a role. Girls who are disruptive or who show less empathy, traits that are relatively tolerated in boys, tend to face more social pushback. That pressure teaches autistic girls early on to hide the behaviors that would otherwise flag them for evaluation.
Biology and the “Female Protective Effect”
The gender gap in autism isn’t purely a product of diagnostic bias. There does appear to be a genuine biological component, though it’s smaller than the 3-to-1 diagnosis ratio suggests. One well-studied theory is the “female protective effect,” which proposes that female biology requires a higher genetic load to produce autistic traits.
In practical terms, this means a girl may need to carry more autism-associated genetic variants than a boy before those variants change her brain development enough to cross the diagnostic threshold. A neuroimaging study of children and adolescents tested this idea by looking at how cumulative genetic risk for autism affected brain connectivity. In boys, both with and without autism, higher genetic risk scores correlated with changes in a key brain network involved in filtering important information from background noise. In girls, the same genetic risk had no measurable effect on that network. The autism risk genes appeared to interact with sex-specific biological processes, buffering girls from their impact.
Prenatal hormones have also been investigated. The “prenatal sex steroid theory” originally proposed that higher levels of testosterone during fetal development contribute to autistic traits. Some studies found correlations between testosterone measured in amniotic fluid and parent-reported autistic traits in childhood, while others found no direct link. Girls with congenital adrenal hyperplasia, a condition that exposes them to elevated androgens before birth, do score higher on measures of autistic traits than their relatives. But the picture is more complicated than testosterone alone: research from a large Danish biobank found no significant difference in amniotic testosterone between autistic and non-autistic boys, suggesting other prenatal hormones are likely involved.
The Gap Is Closing
The most striking trend in recent autism research is how quickly the male-to-female ratio is narrowing, especially among older age groups. The Swedish birth cohort study documented what researchers describe as a “substantial catch-up effect” for females. In that dataset, the cumulative ratio at age 20 dropped from about 1.9 in 2016 to 1.2 in 2022. Among younger children, the ratio has remained more stable at around 3 to 1, likely because early-childhood diagnoses still rely on the same behavioral benchmarks that tend to miss girls.
The narrowing ratio reflects growing awareness among clinicians, parents, and autistic women themselves that autism doesn’t always look like the stereotype. As more women seek and receive evaluations in adolescence and adulthood, the diagnosed population is gradually catching up to the actual autistic population. State-level data in the U.S. confirms this pattern: diagnostic rates for females are rising independently of the overall increase in autism prevalence, and being female remains an independent predictor of later diagnosis regardless of the time period examined.
None of this means the ratio will reach exactly 1 to 1. The female protective effect suggests some genuine biological difference in susceptibility. But the current best evidence points to a true sex difference that is far more modest than the historical 4-to-1 ratio that has shaped public perception for decades.