Can Autism Be Caused by Fetal Alcohol Syndrome?

Prenatal alcohol exposure does not cause autism in the way a virus causes the flu, but the two conditions share enough biology and behavioral overlap that researchers have spent decades trying to untangle their relationship. Children diagnosed with fetal alcohol spectrum disorder (FASD) are diagnosed with autism spectrum disorder (ASD) at rates roughly two to seven times higher than children in the general population, depending on the study and how the conditions are measured.¹ Whether that elevated rate reflects a genuine causal link, a shared set of brain vulnerabilities, or diagnostic confusion between two conditions that look alike on the surface is still an open and genuinely complicated question.

How Often the Two Conditions Overlap

A systematic review and meta-analysis examining children with FASD found that about 2.6% met criteria for ASD, roughly twice the rate seen in the general U.S. population at the time of the analysis.¹ That may sound modest, but it is worth noting that ADHD was far more common in the same group, found in over half the children. ASD was the least frequent of the co-occurring conditions studied, yet it still appeared at a meaningfully elevated rate.

A separate longitudinal cohort study looking specifically at children diagnosed with fetal alcohol syndrome (the most severe end of the FASD spectrum) found an even steeper increase. After adjusting for sex, race, where the children lived, and birth date, the odds of an ASD diagnosis were nearly seven times higher in the FAS group compared to children without an FAS diagnosis.² That study also found elevated rates of intellectual disability, speech and developmental delays, tics, and ADHD, suggesting that prenatal alcohol doesn’t target one neurodevelopmental pathway so much as it disrupts brain development broadly.

The gap between those two figures, a twofold increase versus a nearly sevenfold one, reflects differences in how FASD severity is classified, how ASD is diagnosed, and how the studies were designed. But both point in the same direction: children with prenatal alcohol exposure are more likely to end up with an autism diagnosis than you would expect by chance.

Where the Behavioral Profiles Look Alike

The overlap between FASD and ASD is real enough to create genuine diagnostic headaches. Both groups of children can struggle with social skills, have difficulty reading social cues, and display behaviors that look atypical to parents and clinicians. A study comparing children with ASD and children with FASD found that socially inappropriate behaviors and difficulty getting along with peers were common in both groups.³ On standardized assessments, FASD children can score high enough on autism-screening tools to raise flags, even when they don’t meet full diagnostic criteria for ASD.

Sensory processing adds another layer of similarity. Children with FASD often show sensory sensitivities, including over- or under-responsiveness to sound, touch, and visual stimuli, that overlap with the sensory profiles commonly described in autistic children. Attention problems are rampant in both conditions as well, though for somewhat different reasons. These shared surface features are one reason that the relationship between FASD and ASD has been so hard to pin down: the tools clinicians use to screen for autism pick up on behaviors that prenatal alcohol exposure can also produce.

Where They Diverge

The similarities are striking, but the differences are clinically meaningful. The same study that found overlapping social difficulties also found that certain core features of autism were rare in children with FASD. Difficulty initiating social interaction, limited sharing of emotions, and impaired nonverbal communication were common in the ASD group but uncommon in the FASD group.³ In other words, children with FASD tend to want to connect socially but do it clumsily, while children with ASD are more likely to have reduced motivation or capacity for social engagement in the first place.

Repetitive and restrictive behaviors, another hallmark of autism, also help tell the two conditions apart. Research examining autism characteristics in children with FASD found that their greatest difficulties showed up in social and communicative functioning, not in repetitive or restrictive behaviors.⁴ A child with FASD might struggle to hold a conversation and misread social situations, but they’re less likely to develop the intensely focused interests, rigid routines, or stereotyped movements that characterize many autistic children. This distinction, social awkwardness versus social disinterest, and flexible-but-impulsive behavior versus rigid-and-repetitive behavior, is one of the most useful clinical clues when both diagnoses are on the table.

What Alcohol Does to the Developing Brain

One reason researchers keep asking whether prenatal alcohol could trigger autism is that alcohol’s effects on the fetal brain are so broad that they touch many of the same systems implicated in ASD. Alcohol disrupts brain development through multiple routes at once. It interferes with molecules that guide how neurons migrate and stick together, disrupts the early growth of serotonin-producing neurons that help organize other brain regions, and can trigger excessive cell death through oxidative stress.⁵ It also interferes with growth factors that help new brain cells survive and differentiate properly.

More recent work has shown that alcohol can impair the process by which neural stem cells differentiate into mature neurons. In laboratory and animal models, alcohol activated a cellular stress response that reduced the number of newly formed neurons in the developing brain, and blocking that specific stress pathway partially reversed the damage.⁶ The practical upshot is that alcohol doesn’t just kill brain cells; it can change how the brain wires itself, altering the types and numbers of neurons that form during critical developmental windows.

Some of these disrupted pathways, serotonin signaling, neural migration, cell adhesion, are the same pathways that genetic research has linked to autism. That convergence is what makes the FASD-ASD question so compelling: it’s biologically plausible that environmental damage from alcohol and genetic predisposition toward autism could affect the same brain circuits and produce overlapping outcomes.

The Epigenetic Connection

Beyond direct damage to developing neurons, alcohol also alters how genes are regulated. Research has explored the possibility that prenatal alcohol exposure causes epigenetic changes, modifications to gene expression that don’t alter the DNA sequence itself, that overlap with the kinds of gene-expression patterns seen in autism.⁷ These changes can affect how neurons communicate, how synapses form, and how brain circuits organize themselves, processes collectively called neural plasticity.

The idea is that even if a child doesn’t carry the genetic variants typically associated with ASD, alcohol-driven epigenetic disruption could push their brain development in a direction that mimics an autistic phenotype. This remains a hypothesis more than a settled conclusion. But it offers one of the more coherent explanations for why FASD and ASD overlap behaviorally without one simply “causing” the other in a linear way.

What Animal Studies Show

Some of the most direct evidence that prenatal alcohol can produce autism-like features comes from experiments in mice, where researchers can control the exposure precisely and rule out confounders that plague human studies. In one study, mouse pups exposed to alcohol in utero showed social interaction deficits, a behavior researchers use as an animal analog for a core autism feature. The effect was sex-specific: only male offspring of alcohol-exposed mothers showed the social deficits, while female offspring did not.⁸

That same study examined gene expression in the frontal cortex of the affected male pups and found changes in several genes already linked to autism in human research. Genes involved in brain signaling and synaptic function were either abnormally elevated or suppressed. When the researchers used genetically modified mice that lacked a specific immune-signaling receptor (TLR4), the social deficits and gene expression changes from alcohol exposure were blocked, suggesting that an immune-inflammatory pathway was mediating at least part of the damage.⁸

Animal studies can’t prove that the same thing happens in humans, but they do demonstrate a plausible biological pathway: alcohol activates inflammatory signaling in the fetal brain, which alters the expression of autism-related genes, which produces measurable social behavior changes. The sex difference is also intriguing, since autism in humans is diagnosed far more often in boys than in girls.

What Human Epidemiology Says About Dose and Timing

If prenatal alcohol straightforwardly caused autism, you would expect a clear dose-response relationship in human data: more alcohol, more autism. The picture is messier than that. A large U.S. study called SEED (Study to Explore Early Development) looked at alcohol use before and during pregnancy and found that light drinking of one to two drinks per week around the time of conception and very early pregnancy was not associated with increased ASD risk. In fact, for some time windows, light drinking showed an inverse association with ASD, though that likely reflects characteristics of the mothers who drink lightly rather than any protective effect of alcohol itself.⁹

At higher doses, the signal shifted. Drinking three or more drinks per week during the second month of pregnancy was associated with elevated odds of ASD, though the estimates were imprecise and the confidence intervals were wide enough that the finding could have been due to chance.⁹ Binge drinking in the first trimester was not associated with ASD or developmental delay in the same analysis. That doesn’t mean binge drinking is safe; it means the relationship between alcohol exposure and autism specifically, as opposed to FASD’s broader neurodevelopmental effects, doesn’t follow a neat line.

Timing matters as much as quantity because the fetal brain goes through rapid, sequential changes throughout pregnancy. The structures and circuits most relevant to social cognition develop during different windows than those governing motor control or facial features. This is part of why FASD itself is a spectrum: two pregnancies with similar total alcohol exposure can produce very different outcomes depending on when the heaviest drinking occurred.

Genetic Vulnerability as a Modifier

Not every child exposed to alcohol in utero develops FASD, and among those who do, the severity varies enormously. Part of the explanation is genetic. Research in both humans and animal models has identified genetic factors in both the mother and the developing child that influence how susceptible a pregnancy is to alcohol’s effects.¹⁰ Variations in how efficiently a mother metabolizes alcohol, for example, change how much alcohol reaches the fetus and for how long.

This is relevant to the autism question because some researchers have proposed that alcohol exposure may be most likely to produce autism-like outcomes in children who already carry some genetic susceptibility to ASD. Some authors have hypothesized that prenatal alcohol exposure could be linked to increased ASD risk specifically in children with underlying genetic vulnerability.¹¹ Under this model, alcohol wouldn’t cause autism on its own so much as it would push a genetically predisposed brain past a threshold. The evidence for this gene-environment interaction model is suggestive rather than conclusive, but it fits with the broader pattern in autism research, where most cases appear to involve a combination of inherited risk and environmental triggers rather than a single cause.

The Diagnostic Confusion Problem

One of the most consequential real-world effects of the FASD-ASD overlap is misdiagnosis or, more precisely, diagnosis by availability. FASD is widely underdiagnosed. Many clinicians are not trained to recognize it, and a diagnosis requires knowing about the mother’s alcohol use during pregnancy, information that is often unavailable or unreliable. Autism, by contrast, is better known, more routinely screened for, and carries less social stigma for caregivers.

A study examining children receiving developmental services in San Diego County found that among children who could be classified as having FASD, roughly two-thirds were receiving services based on an autism diagnosis.¹² The behavioral overlap between the two conditions was enough that these children qualified for services under the autism umbrella, even though FASD may have been the more accurate primary diagnosis. This doesn’t mean none of them were genuinely autistic, but it does suggest that some fraction of children diagnosed with ASD may actually have FASD, or may have both.

The distinction matters for treatment planning. Interventions for autism tend to focus on building social communication skills and managing sensory and behavioral rigidity. Children with FASD often benefit more from strategies targeting executive function, impulse control, and adaptive behavior in unstructured settings. A child misdiagnosed with ASD when their primary condition is FASD may receive therapies that address only part of their profile while missing the areas where they need the most support.

Why the Answer Stays Complicated

Reviews of the literature have consistently described the relationship between FASD and ASD as “controversial,” with studies highlighting both similarities and differences that resist a clean conclusion.¹¹ Part of the difficulty is that both FASD and ASD are spectrum conditions with enormous internal variability. A child with full fetal alcohol syndrome, including facial features, growth restriction, and severe cognitive impairment, looks very different from a child on the milder end of FASD who has normal IQ but struggles with attention and social skills. Similarly, an autistic child with minimal language and high support needs occupies a different clinical space from an autistic adult who lives independently but finds social situations exhausting. Comparing two moving targets makes it hard to draw clean lines.

Another obstacle is that the best way to test whether prenatal alcohol causes autism would be a randomized controlled trial, which is obviously unethical. Researchers are left with observational studies, animal models, and case series, all of which have limitations. Observational studies in humans struggle to separate the effect of alcohol from other risk factors that tend to travel with heavy drinking during pregnancy, including poverty, poor nutrition, exposure to other substances, and higher levels of stress. Animal models can isolate the alcohol variable but can’t fully replicate human brain development or human social behavior.

When Both Diagnoses Apply

For some children, the answer isn’t FASD or ASD but both. The two conditions are not mutually exclusive, and a child can genuinely have the brain damage caused by prenatal alcohol exposure and independently meet full diagnostic criteria for autism. In those cases, the question of whether alcohol “caused” the autism becomes somewhat academic. What matters clinically is that both sets of challenges are identified and addressed.

Families navigating a dual diagnosis face a particular set of difficulties. Services for developmental disabilities are often organized around single-diagnosis categories, and eligibility criteria may not account well for the compounding effects of two co-occurring conditions. A child with both FASD and ASD may have more severe executive function impairments, more complex sensory needs, and greater difficulty with adaptive skills than a child with either condition alone, yet their treatment plan may be modeled on one diagnosis or the other rather than the combination. Clinicians who are familiar with both conditions are relatively rare, which means caregivers often end up doing a lot of the coordination work themselves.

From a research perspective, the existence of genuinely co-occurring cases makes it harder to study each condition in isolation. If a study of FASD includes children who are also autistic without identifying them, the results will be muddied. Similarly, autism studies that don’t screen for prenatal alcohol exposure may unknowingly include children whose behavioral profiles are partly shaped by alcohol rather than by the genetic factors the study is trying to investigate. Better screening practices for both conditions would sharpen the science on each of them.