Gestational thyroid hormones and autism-related traits in the EARLI and HOME studies.
Extreme mid-pregnancy free thyroxine predicts slightly higher autism-social scores, adding thyroid hormones to the prenatal risk checklist.
01Research in Context
What this study did
Zhong et al. (2024) checked frozen blood spots from two pregnancy groups. They looked at free thyroxine (FT4) and thyroid-stimulating hormone (TSH) taken around week 20 of pregnancy.
Later, when the children were 3-5 years old, parents filled out the Social Responsiveness Scale (SRS). The team asked: do higher or lower mid-pregnancy thyroid levels line up with more autism-like social traits?
What they found
Kids whose moms had very high FT4 (top a large share) scored about 4 points higher on the SRS. That means more social-communication quirks.
Higher TSH, on the other hand, was linked to slightly lower SRS scores. The pattern looked different in each cohort, so the link is not one-size-fits-all.
How this fits with other research
Ganz et al. (2009) first tied hormone genes to autistic traits. Caichen moves the story forward by showing actual hormone levels, not just genes, matter too.
Carter et al. (2011) blamed high testosterone for autism's male skew. Caichen flips the spotlight to thyroid hormones, hinting that more than one prenatal hormone pathway is in play.
Ptomey et al. (2021) linked oxytocin-receptor methylation to social problems. Both papers show measurable biology tracking with trait severity, giving us parallel biomarkers to watch.
Why it matters
You can't change mom's thyroid levels after the fact, but you can use the info to flag kids who might need closer developmental monitoring. If a parent mentions maternal Graves disease or high FT4 in pregnancy, think social skills on your assessment list. The SRS bump is small, yet it reminds us that biology and behavior start sharing stories before birth.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Add a quick question about mom's thyroid problems to your intake form; flag high FT4 history for extra social-skills screening.
02At a glance
03Original abstract
Thyroid hormones are essential for neurodevelopment. Few studies have considered associations with quantitatively measured autism spectrum disorder (ASD)-related traits, which may help elucidate associations for a broader population. Participants were drawn from two prospective pregnancy cohorts: the Early Autism Risk Longitudinal Investigation (EARLI), enrolling pregnant women who already had a child with ASD, and the Health Outcomes and Measures of the Environment (HOME) Study, following pregnant women from the greater Cincinnati, OH area. Gestational thyroid-stimulating hormone (TSH) and free thyroxine (FT4) were measured in mid-pregnancy 16 (±3) weeks gestation serum samples. ASD-related traits were measured using the Social Responsiveness Scale (SRS) at ages 3-8 years. The association was examined using quantile regression, adjusting for maternal and sociodemographic factors. 278 participants (132 from EARLI, 146 from HOME) were included. TSH distributions were similar across cohorts, while FT4 levels were higher in EARLI compared to HOME. In pooled analyses, particularly for those in the highest SRS quantile (95th percentile), higher FT4 levels were associated with increasing SRS scores (β = 5.21, 95% CI = 0.93, 9.48), and higher TSH levels were associated with decreasing SRS scores (β = -6.94, 95% CI = -11.04, -2.83). The association between TSH and SRS remained significant in HOME for the 95% percentile of SRS scores (β = -6.48, 95% CI = -12.16, -0.80), but not EARLI. Results for FT4 were attenuated when examined in the individual cohorts. Our results add to evidence that gestational thyroid hormones may be associated with ASD-related outcomes by suggesting that relationships may differ across the distribution of ASD-related traits and by familial likelihood of ASD.
Autism research : official journal of the International Society for Autism Research, 2024 · doi:10.1002/aur.3115