Could a grandfather’s health history hold clues about autism risk decades later?
A new California study has found that grandchildren were more likely to receive an autism-spectrum-disorder diagnosis when their grandfathers smoked or reported heavier liquor use while their partners were pregnant.
The finding is drawing attention because it looks beyond the usual focus on a child’s immediate surroundings and examines whether certain exposures may be linked across generations.
But researchers stress that the study does not prove that a grandfather’s smoking or alcohol use caused autism. Instead, it adds to an emerging area of research exploring how genetics, environmental exposures, and family health histories may interact over time.
Published August 25, 2026, in Autism Research, the peer-reviewed study used records from the long-running Child Health and Development Studies cohort in California. Researchers analyzed pregnancy exposure information collected between 1959 and 1967 and later connected those records with California birth and developmental records from 1989 through 2014.
The unusual design gave researchers something rarely available in this type of research: exposure information collected decades before autism diagnoses were recorded.
The Study Found A Pattern Across Three Generations
Researchers analyzed 14,059 third-generation children who had at least one available grandparent exposure measure. Among those children, 86 had autism-spectrum-disorder records.
The strongest associations involved grandfathers’ tobacco use and heavier liquor consumption.
Grandfathers who used tobacco had a higher relative risk of autism-spectrum-disorder diagnosis among grandchildren compared with those who did not use tobacco. Cigarette smoking specifically was also associated with increased relative risk.
Among grandfathers who smoked at least 20 cigarettes daily, the adjusted relative risk was about twice that of grandchildren whose grandfathers did not smoke.
That does not mean a child with this family history has a predictable outcome. Relative risk describes a difference between groups in a study population, not the chance that any individual child will develop autism.
The study also found an association between heavier liquor consumption and autism diagnosis risk. Grandfathers who reported five or more liquor drinks per week had a higher relative risk compared with those who reported no liquor use.
However, researchers noted that this estimate was less certain because the number of autism cases was relatively small.
Why Researchers Are Looking At Grandparents’ Health
Most autism research focuses on a child’s own genetics and early development. This study looked further back, asking whether certain exposures before a child’s parents were born might be connected with later generations.
Scientists are interested in whether environmental exposures could influence biological processes that affect how genes function. One possible area of interest is epigenetics, changes in gene activity that do not alter the DNA sequence itself.
But the researchers have not proven that an epigenetic process explains these findings.
The study offers a possible clue, not a final answer.
The Study Does Not Mean Grandparents Caused Autism
Autism is not explained by one behavior, one exposure, or one family decision.
Researchers generally understand autism as a complex developmental condition influenced by many factors, including genetics and environmental influences. No single study can explain why one child is autistic, and another is not.
Research like this can easily be misunderstood.
The study cannot tell a family whether a particular grandfather’s smoking or alcohol use contributed to a specific child’s autism diagnosis. It identifies a pattern across a population, not a cause for a person.
Save this article
Enter your email address and we'll send it straight to your inbox.
The authors also acknowledge that other factors (including inherited traits, family circumstances, other exposures, and factors not measured in the study) may help explain the association.
Related: 8 Grandparent Behaviors That Damage Family Relationships
The Numbers Are Important, But So Are The Limitations
The study’s findings are based on a relatively small number of autism diagnoses.
Only 86 children in the analysis had autism-spectrum-disorder records. That means some estimates, especially those involving smaller exposure groups, should be interpreted cautiously.
Questions also remain about how exposure information was collected.
Grandfather smoking and alcohol use were reported by their pregnant partners rather than measured directly from the men themselves. That means some information may not have been perfectly accurate.
The study also relied on California Department of Developmental Services records to identify autism diagnoses. Those records may capture children with higher support needs more consistently and may not include every autistic child.
Researchers say additional studies will be needed to determine whether these findings appear in other populations.
Autism Research Is Looking At A Bigger Picture
The study comes as autism identification has increased in the United States.
Current CDC surveillance estimates indicate that about 1 in 31 U.S. 8-year-olds were identified with autism in 2022. Researchers caution that changes in screening, diagnosis, and access to evaluations influence these numbers.
A rise in autism identification cannot be explained by one exposure or one environmental factor.
Scientists continue to study how inherited genetics, prenatal development, environmental influences, and early-life experiences interact. The goal is not to find one explanation, but to better understand the many pathways that may contribute to autism.
What Families Should Take From The Research
The most important takeaway is that this study focuses on understanding patterns; not assigning blame.
Researchers are increasingly interested in how health histories may extend across generations, but a population-level association cannot explain an individual child’s autism diagnosis.
For families, the focus remains on supporting children’s development, discussing concerns with healthcare providers, and seeking evaluation and services when needed.
This study may open new questions for scientists, but it does not change a central point about autism: it is a complex condition shaped by many factors, and no family should view one research finding as an explanation for why their child is autistic.
How can researchers share emerging findings about autism without creating unnecessary fear or placing blame on families?
