In one sentence
A large Danish registry study (preprint, not yet peer-reviewed) suggests that links between some maternal health conditions and autism partly reflect shared genes passed to the child, while other links may also involve genes that shape the prenatal environment.
What the researchers did
Using Danish national registers, researchers followed children born 1998–2015 and their relatives across three generations. They compared full maternal cousin pairs: “parallel” cousins (children of the mother’s sister) and “cross” cousins (children of the mother’s brother). Direct genetic effects that travel with DNA should look similar in both cousin types; effects that act through the mother’s pregnancy-related biology should be stronger in parallel cousins. They examined maternal diagnoses previously tied to offspring autism — for example postpartum haemorrhage, personality disorders, epilepsy, false labour, depression, anxiety, and connective-tissue conditions — and modelled autism risk in the cousins.
What they found
- Some maternal diagnoses (including postpartum haemorrhage, personality disorders, and epilepsy) were associated with autism in both parallel and cross cousins — a pattern consistent with shared direct genetic effects.
- Other conditions (including false labour, recurrent major depression, other anxiety disorders, and systemic connective-tissue involvement) showed stronger associations in parallel than in cross cousins — consistent with additional indirect genetic effects via the prenatal environment.
- Adjusting for the same diagnosis in the cousin’s own mother did not substantially change estimates, offering little support for purely non-genetic mechanisms tied to that diagnosis label alone.
What this means for families and therapists
- A maternal health diagnosis linked to autism in research does not automatically mean “the pregnancy caused autism.” Shared genetics often sit behind both.
- Clinicians can use this nuance when counselling families who feel blamed for prenatal events or maternal diagnoses.
- This does not change day-to-day supports for autistic children; it clarifies how to talk about risk without oversimplifying cause.
Limitations and what we don't know yet
- This is a medRxiv preprint and has not completed peer review.
- Registry diagnoses may miss milder or untreated conditions and lack molecular genetic data.
- The design is observational and cannot prove causation for any single pathway.
- Findings come from Denmark and may not generalise to all populations.
- Follow-up may miss later autism diagnoses, especially in younger cohorts.
This is a plain-language summary of Maternal health and autism risk: parsing direct and indirect genetic effects using 3-generation family linkage by Arildskov E.S., Khachadourian V., Grove J. et al., medRxiv, preprint (2026). Source license: CC-BY-NC-ND-4.0. It is not medical advice — talk to a qualified clinician before changing therapy.