Prenatal Depression and Anxiety Disrupt Infant Sleep Through Separate Biological Pathways
A large prospective study finds that maternal depression and anxiety harm infant sleep through distinct mechanisms, one involving stress hormones and the other the gut microbiome, pointing toward targeted interventions for each condition.
Lead
For years, clinicians have known that prenatal psychological distress raises the risk of infant sleep problems. But they did not know whether depression and anxiety operated through the same biological channels, or whether treating them would require different approaches. A new study published in Sleep provides the clearest answer yet: they act through separate pathways. And the distinction matters for how doctors might intervene.
What they found
Researchers at Ningxia Medical University in China enrolled 2,288 mother-infant pairs in a prospective birth cohort. They assessed maternal depression and anxiety during pregnancy and tracked infant sleep outcomes through 12 months of age. A smaller sub-cohort of 112 pairs underwent deep molecular profiling, including meconium microbiota analysis by 16S rRNA sequencing and cord blood tryptophan metabolite measurements by LC-MS/MS.
The results showed that prenatal depression alone was a significant and independent risk factor for infant sleep disturbance. After adjusting for confounders, the odds ratio was 1.53 (95% CI 1.04 to 2.25). The effect was especially pronounced in female infants, where the odds ratio rose to 2.11 (p = 0.022).
Maternal anxiety, by contrast, was not directly linked to sleep disturbance in the same way. Instead, anxiety was associated with changes in the infant’s gut ecosystem. It predicted reduced meconium alpha-diversity, lower levels of beneficial Bifidobacterium, and decreased concentrations of 3-HAA and serotonin in cord blood. These microbiota and metabolite changes, in turn, were linked to poorer infant sleep through a serial mediation pathway.
The depression pathway ran through a different biological axis. Cord blood cortisol partially mediated the link between prenatal depression and infant sleep disruption, with a significant average causal mediation effect of -7.47 (95% CI -14.82 to -0.12, p = 0.048). This points to the hypothalamic-pituitary-adrenal (HPA) axis as the primary conduit.
To test the predictive value of these markers, the team built an XGBoost machine learning model. It achieved an area under the curve of 0.727, with microbial diversity, Streptococcus abundance, dopamine, and 3-HAA emerging as the top predictors.
Why it matters
These findings suggest that prenatal depression and anxiety are not interchangeable risk factors when it comes to infant sleep. They appear to disrupt the baby’s developing physiology in different ways. Depression exerts its effect through the mother’s stress hormone system, elevating cortisol that reaches the fetus and programs its developing sleep-wake circuitry. Anxiety, meanwhile, shifts the infant’s gut microbial community and the tryptophan-serotonin pathway, a key signaling system for sleep regulation.
The clinical implication is significant. If both conditions were treated identically during pregnancy, clinicians might address only half the problem. A depressed mother might benefit most from interventions that lower cortisol and buffer HPA axis activity. An anxious mother might respond better to approaches that support gut health, such as probiotics, prebiotics, or dietary modulation of the microbiome.
The study also raises the possibility of early risk identification. The XGBoost model, with its AUC of 0.727, suggests that a combination of microbial and metabolic markers could one day help identify infants at highest risk before sleep problems emerge.
Limits
The study has important limitations. The sub-cohort for multi-omics analysis was relatively small, which limits statistical power for the mediation findings. The cohort was drawn from a single region in China, and results may not generalize to other populations with different diets, lifestyles, and microbial exposures. Infant sleep was measured by parental report rather than objective methods such as actigraphy, which introduces potential reporting bias. And while the mediation analyses suggest causal pathways, observational data cannot prove causation.
Bottom line
Prenatal depression and anxiety harm infant sleep through distinct biological routes. Depression acts through the HPA axis and cortisol. Anxiety acts through the gut-brain axis, altering the infant microbiome and tryptophan metabolism. These separate pathways call for separate intervention strategies and open the door to personalized approaches for protecting infant sleep before birth.
Source
Liu C, Lin Y, Li Y, et al. “Differential Effects of Prenatal Depression and Anxiety on Infant Sleep: Dual-Pathway Mechanisms Involving the HPA Axis and the Gut-Brain Axis.” Sleep, 2026. DOI: 10.1093/sleep/zsag171. PMID: 42364158.