![[Adobe Stock]](https://sp-ao.shortpixel.ai/client/to_webp,q_glossy,ret_img,w_750,h_375/https://www.psypost.org/wp-content/uploads/2025/08/pregnant-woman-750x375.jpg)
A new study published in Autism Research reveals that the age of maternal grandparents when they have children is linked to the likelihood of their grandchildren developing autism. The research shows that this association varies substantially across different racial and ethnic groups. These variations suggest that environmental and social factors operate alongside biology to influence child development across multiple generations.
Autism spectrum disorder is a neurodevelopmental condition characterized by repetitive behaviors and challenges with social communication. In recent decades, the identified prevalence of autism has grown rapidly. In California, the proportion of children diagnosed with the condition increased dramatically between the early 2000s and recent years. During this period, the diagnostic rates among historically underrepresented minority groups surpassed the rates seen in white children.
As diagnostic rates have shifted, demographic patterns in family planning have also changed. The average age of parents at the time of childbirth has steadily increased in the United States. Past research has established a firm association between older parental age and a higher chance of having a child on the autism spectrum.
Recent evidence has expanded this timeline backward, pointing to the age of the grandparents. When grandparents conceive the parents, their age may carry biological and social implications for the grandchildren. To understand this multigenerational pattern, Ting Chow, a public health researcher at the University of California, Los Angeles, and her colleagues initiated a large scale investigation.
Previous investigations into grandparental age focused almost exclusively on populations of white, European ancestry. The researchers wanted to see if the same age patterns appeared within a highly diverse population. They hypothesized that studying various racial and ethnic groups could help determine whether these age links are universal biological traits or markers for differing environmental conditions.
The biological mechanisms behind multigenerational health outcomes are still under active investigation. One possibility involves changes to the germ cells, which are the sperm and eggs. Environmental exposures or the natural aging process can alter epigenetic marks. These marks act as a chemical control system that turns genes on or off without altering the underlying genetic code.
Aging can also lead to the accumulation of damage in the mitochondria, the energy producing structures inside cells. A child inherits mitochondria exclusively from the mother’s egg. A female fetus develops all the eggs she will ever have while still inside her mother’s womb. Because of this timeline, the grandmother’s age and environment during pregnancy can directly influence the cellular health of the eggs that will eventually become her grandchildren.
Older grandfathers might pass along new genetic mutations or epigenetic changes through their sperm. Unlike female eggs, male sperm reproduce continuously from adolescence onward. This constant replication provides more opportunities for cellular errors to accumulate as a man ages.
To explore these possibilities, the researchers analyzed birth records from California. They linked the health records of children born between 2001 and 2019 to the birth records of their mothers, who were born between 1983 and 2001. The team also incorporated diagnostic data from the California Department of Developmental Services to identify autism cases.
This complex data linkage created a study population of more than 1.7 million mother and child pairs for the grandmother analysis. Within this massive group, the researchers identified nearly 28,000 children diagnosed with autism. They then looked backward in time to determine the exact age of the maternal grandmother and maternal grandfather when the mother was born.
The researchers categorized the grandparent ages into four brackets. They defined younger grandparents as those between 18 and 24 years old. The reference group consisted of grandparents between 25 and 29 years old. They also created categories for those who were 30 to 34 years old and those who were 35 to 55 years old.
They used statistical models to compare the odds of an autism diagnosis among grandchildren based on these different age brackets. Their calculations accounted for the birth years of the child and the mother, the sex of the child, and the number of previous pregnancies the grandmother had experienced.
Overall, the team found that grandchildren had a slightly higher chance of being diagnosed with autism if their maternal grandparents were either unusually young or relatively old when the mother was born. But these broad averages masked distinct variations among different racial and ethnic communities. The researchers observed a pronounced U shaped statistical curve specifically among white grandparents.
For white grandmothers and grandfathers, both the youngest and oldest age brackets were associated with an elevated likelihood of autism in grandchildren. Among Hispanic grandparents, the mathematical relationship looked different. Elevated odds for autism in the grandchildren emerged only when Hispanic grandmothers and grandfathers were in the oldest age category.
The patterns changed again when looking at other demographic groups. For Asian Pacific Islander families, increased odds of autism were only present among older grandmothers. In Black families, the researchers noted a decrease in documented autism odds associated with younger grandmothers. At the same time, older Black grandfathers were linked to an elevated rate of autism in their grandchildren.
The study authors suggest these racial and ethnic variations point to different underlying mechanisms. Among white populations, exceptionally young childbearing often correlates with lower socioeconomic status. The challenges associated with limited resources may cascade across generations to influence the health of the grandchildren.
The inverse pattern seen with younger Black grandmothers might be explained by a phenomenon known as live birth bias. This statistical concept implies that the most vulnerable fetuses might not survive to birth. If socioeconomically disadvantaged younger grandmothers experience high rates of pregnancy loss, the children who are born might represent a particularly resilient subset, artificially lowering the observed autism rate in the data.
In contrast, older grandparents across most demographic groups displayed a positive link with autism in their grandchildren. This consistency aligns with the biological theories of accumulating cellular damage or genetic mutations. Environmental factors, which disproportionately affect minority neighborhoods, might also combine with advanced age to negatively influence reproductive health.
Historically, Black and Hispanic populations in California have been more likely to face occupational hazards, higher levels of financial stress, and systemic discrimination. Chronic stress and toxic environmental exposures can alter fetal development. If a grandmother experiences these hardships while pregnant, the developing eggs inside her female fetus could undergo changes that eventually affect the grandchild.
The researchers acknowledged several caveats in their analysis. Relying entirely on birth and diagnostic records means that some nuances of family life remain unmeasured. For instance, the state databases did not contain comprehensive information on lifestyle habits, such as smoking and alcohol use. The researchers also lacked data on family histories of psychiatric conditions.
The investigation was strictly limited to the maternal genetic line. The researchers focused on the maternal grandparents because paternal identifiers on older birth records were historically less complete. This data gap prevented an equivalent analysis of the paternal grandparents and their generational influence. Data availability also forced the researchers to exclude older mothers who were born before electronic records began in 1983.
Future research will need to incorporate detailed biological, social, and environmental data. Expanding this work to diverse populations outside of California could help clarify how different social contexts interact with genetic inheritance. Tracing both the maternal and paternal ancestral lines would provide a more complete picture of family health timelines.
The study, “Age, Race, and Ethnicity of Maternal Grandparents in Autism Spectrum Disorder, a California Multigenerational Study,” was authored by Ting Chow, Qi Meng, Jingyuan Xiao, Karl O’Sharkey, Zeyan Liew, and Beate Ritz.
