BACKGROUND:Maternal air pollution exposure has been associated with impaired fetal growth, yet most studies have overlooked microenvironmental and personal exposures. OBJECTIVES:To examine associations between maternal air pollution exposure in key microenvironments (home, workplace, and commuting route) and fetal growth (birth weight and small for gestational age) using three modelling approaches and personal, home-indoor, and home-outdoor monitoring. METHODS:We used data from 1024 pregnant women in the Barcelona Life Study Cohort (2018-2021). Exposure to nitrogen dioxide, black carbon, and fine particulate matter (PM2.5) and its metallic constituents (copper, iron, and zinc) in each microenvironment were estimated using land use regression models, dispersion models, and hybrid land use regression-dispersion models, and combined with time-activity data to estimate total microenvironment exposures. Personal, home-indoor, and home-outdoor nitrogen dioxide concentrations were measured using passive samplers. Associations with birth weight and small for gestational age were evaluated using linear and logistic mixed-effects models. RESULTS:Higher nitrogen dioxide and black carbon exposure at home and in total microenvironments, estimated by land use regression, dispersion, and hybrid models, were associated with lower birth weight. Increased black carbon exposure in the workplace (hybrid model) and PM2.5 exposure both at home (land use regression model) and in total microenvironments (land use regression and dispersion models) were also associated with reduced birth weight, as were higher home, workplace, and total microenvironmental exposure to metallic components of PM2.5 in land use regression models, although higher workplace PM2.5-zinc was associated with higher birth weight in hybrid models. Higher personal and home-outdoor nitrogen dioxide exposure were further associated with reduced birth weight. Similar patterns were observed for small for gestational age. CONCLUSION:Maternal air pollution exposure was associated with impaired fetal growth. Home-based exposure estimates and short-term nitrogen dioxide measurements may serve as practical exposure proxies during pregnancy.

Neuroimaging studies suggest that air pollution exposure interferes with brain functional development, potentially affecting long-term cognitive abilities. Early life may represent a critical window of vulnerability, yet most research has assessed exposure effects later in childhood, when postnatal influences accumulate. This study examined the early impact of prenatal exposure to airborne particulate matter (PM2.5) on the functional structure of local connections in the cerebral cortex of neonates. This population-based study was conducted in Barcelona between 2018 and 2021. PM₂.₅ exposure was estimated using land-use regression models incorporating time-weighted maternal mobility data across distinct environments. Resting-state functional MRI scans were obtained from 61 neonates at 29 days postnatally. Brain functional connectivity was assessed using local Iso-Distant Average Correlation (IDAC) measures across different spatial distances. Voxel-wise regression analyses were performed to examine associations between PM2.5 exposure and IDAC measures. Higher maternal PM2.5 exposure during gestation was associated with stronger local functional connectivity in the neonatal cortex, particularly in brain regions involved in sensorimotor function. In the premotor and supplementary motor cortices, these effects varied by spatial scale, showing stronger associations at longer local distances. The findings suggest that prenatal exposure to air pollution interferes with early brain functional organization. Notably, the observed connectivity changes could reflect both a delay in the segregation of primary sensorimotor areas and the acceleration of functional maturation in higher levels of the sensorimotor system. Longitudinal studies are needed to track how these exposure-related alterations evolve over time and their potential impact on developmental outcomes.

Artificial intelligence (AI) is rapidly transforming healthcare, with growing impact on maternal and child health (MCH) through advances in machine learning, deep learning, computer vision, generative models, and conversational systems. This article provides a comprehensive synthesis of current AI applications in MCH, structured across six key domains: predictive modeling, image analysis, deep learning and interpretability, generative and multi-omics approaches, conversational AI, and environmental and lifestyle analytics. Drawing on recent literature and the translational experience of the Spanish RICORS-SAMID network, we analyze how these technologies are being integrated into clinical, preventive, and assistive workflows. Across domains, AI demonstrates strong potential for early risk prediction (e.g., preeclampsia, fetal growth restriction, neonatal outcomes), automated image interpretation, biomarker discovery, and personalized decision support. However, despite promising performance metrics, most systems remain at the proof-of-concept stage, with limited external validation, scarce prospective evaluation, and incomplete integration into real-world clinical pathways. Key translational gaps include data heterogeneity, lack of interoperability, insufficient explainability, and challenges related to bias, fairness, and regulatory compliance. We argue that the next phase of AI in MCH must shift from static prediction toward longitudinal, mechanism-aware, and clinically actionable systems, supported by robust validation and multidisciplinary collaboration. Particular emphasis is placed on equity, as the benefits of AI must extend to low-resource settings where maternal and neonatal morbidity remains highest. By bridging technical innovation with clinical implementation, coordinated research networks such as RICORS can play a critical role in accelerating the safe, effective, and equitable deployment of AI in maternal and child healthcare.

Prenatal air pollution exposure has been associated with impaired fetal growth. However, the underlying biological mechanisms, potentially involving the placenta, remain poorly understood. This study aimed to investigate the effects of maternal air pollution exposure on placental biomarkers of oxidative stress and cellular ageing, and to determine whether these biomarkers contribute to the association between air pollution exposure and birth weight. The study included 499 participants from the Barcelona Life Study Cohort (BiSC), recruited between October 2018 and April 2021. Maternal exposure to nitrogen dioxide (NO2), black carbon, and particulate matter ≤2.5 μm (PM2.5) during pregnancy was estimated using hybrid Land Use Regression (LUR)-dispersion models. Placental relative mitochondrial DNA content (mtDNAcn) and telomere length (TL) were measured by qPCR, and associations between exposure and placental biomarkers were estimated using linear mixed-effects regression models adjusted for covariates. Increasing levels of NO2 and PM2.5 during the 1st and 3rd trimesters and across the entire pregnancy, as well as black carbon during the 3rd trimester, were significantly associated with an increase in placental mtDNAcn. No significant associations were observed between prenatal air pollution exposure and placental TL, nor between either placental biomarker and birth weight. In conclusion, prenatal air pollution exposure was associated with differences in placental mtDNAcn but not TL, and neither biomarker was significantly associated with birth weight.

Understanding the factors that shape daily mobility during pregnancy is essential for inclusive transportation planning that promotes active travel for all. Using smartphone-based Global Positioning System data from 860 pregnant women in Barcelona, Spain, we evaluated the correlates of active and passive travel in early and late pregnancy. We identified 33 correlates from 48 candidate variables including personal characteristics, the residential physical environment, the social environment, and temporal factors. The most important correlate across pregnancy was non-European ethnic origin, being associated with 10-15 min less daily active travel. In early pregnancy, commuting distance was the most important correlate, being positively associated with passive travel, while the COVID-19 pandemic was associated with less passive travel. In late pregnancy, residential walkability and having a university degree were positively associated with active travel. The neighbourhood education level was associated with more active travel, particularly during weekends. We discuss key priorities for supporting active travel during pregnancy.

Maternal exposure to air pollution has been associated with a wide range of adverse perinatal outcomes; however, the evidence on its influence on the placenta has remained limited. We studied the association of maternal exposure to air pollution with fetoplacental Doppler markers in a cohort of pregnant women recruited in three tertiary hospitals and their primary care centers in Barcelona, Spain (2018- 2021). Maternal exposure to nitrogen dioxide (NO2), black carbon (BC), and particulate matter <2.5 μm (PM2.5) was assessed at home, workplace, commuting route, and all these microenvironments combined using different modelling techniques: land use regression (LUR), dispersion (DM), and Hybrid LUR-DM (HM) models. Doppler examination of the Uterine artery (UtA) mean pulsatility index (PI) was assessed in the first, second, and third trimester of pregnancy, while the umbilical artery (UA) PI, middle cerebral artery (MCA) PI, and cerebroplacental ratio (CPR) were examined in the third trimester. Mixed effects linear regression models with recruiting hospital as the random effect were used to estimate associations, controlled for relevant covariates. Higher maternal exposure to NO2 and BC was consistently associated with increased UtA mean PI, particularly for home and total exposure estimates by all three exposure models, while PM2.5 showed similar associations in some exposure models (DM). We also observed a significant association between home NO2 exposure and lower CPR, suggesting subtle fetal hemodynamic adaptation to impaired placental function. Overall, our findings provide evidence that maternal exposure to ambient air pollution may adversely affect uteroplacental vascular function. Further prospective studies are needed to confirm these associations and assess their clinical relevance for maternal and fetal outcomes.

BACKGROUND:Preeclampsia is a pregnancy-related disorder characterized by systemic endothelial dysfunction and angiogenic imbalance, most notably elevated levels of soluble fms-like tyrosine kinase-1 and decreased placental growth factor. While preeclampsia has been associated with long-term cardiovascular and cognitive risks, the specific role of angiogenic imbalance in predicting postpartum memory impairment remains unclear. Identifying biomarkers that can anticipate future neurocognitive outcomes may offer opportunities for early intervention and monitoring. OBJECTIVE:To evaluate whether preeclampsia and angiogenic imbalance during pregnancy, defined by a soluble fms-like tyrosine kinase-1/placental growth factor ratio ≥38, are associated with subjective memory impairment 3 to 6 years postpartum. STUDY DESIGN:Cross-sectional study at a tertiary hospital in Barcelona, Spain. Individuals were prospectively recruited during pregnancy and reevaluated 3 to 6 years postpartum. Preeclampsia was defined per American College of Obstetricians and Gynecologists criteria. Angiogenic imbalance during pregnancy was defined as soluble fms-like tyrosine kinase-1/placental growth factor ≥38 (determined between 28 and 40 weeks of gestation). Subjective memory performance was assessed using the validated Memory Failures of Everyday Life questionnaire. Memory impairment was defined as a total Memory Failures of Everyday Life score ≥36. Logistic and linear regression models were used to examine associations, adjusting for relevant confounders. RESULTS:A total of 266 individuals were reevaluated between August 2023 and February 2025. 81 of them (30.45%) had a documented history of preeclampsia. Participants with an elevated soluble fms-like tyrosine kinase-1/placental growth factor ratio during pregnancy showed a higher prevalence of memory impairment (30.0% vs 16.2%, P=.03). In multivariable analysis, angiogenic imbalance remained significantly associated with increased odds of memory impairment (odds ratio=2.18, 95% confidence interval [1.02-4.65], P=.04). In contrast, preeclampsia diagnosis alone was not significantly associated with memory outcomes (odds ratio=1.35, 95% confidence interval [0.70-2.60], P=.37). CONCLUSION:An elevated soluble fms-like tyrosine kinase-1/placental growth factor ratio during pregnancy is associated with an increased risk of subjective memory impairment 3 to 6 years postpartum. These findings highlight the potential utility of angiogenic biomarkers as early indicators of long-term cognitive vulnerability, supporting the need for longitudinal follow-up and targeted preventive strategies in women exposed to angiogenic imbalance during pregnancy.

Preeclampsia and fetal growth restriction, which are clinical presentations of placental dysfunction, are frequently associated with angiogenic imbalance during pregnancy and increased long-term cardiovascular risk. Whether this risk is driven by the pre-pregnancy risk factors, clinical disease, or by an elevated sFlt-1/PlGF ratio remains unclear. This study aimed to assess the association between vascular assessments (ophthalmic artery Doppler and carotid intima-media thickness) 3–6 years postpartum and a history of preeclampsia and fetal growth restriction, and to evaluate if associations were impacted by an angiogenic imbalance during pregnancy. This was a cross-sectional study, which included individuals prospectively recruited during their index pregnancy between 2018 and 2021 and re-evaluated 3–6 years postpartum. Preeclampsia, fetal growth restriction, and sFlt-1/PlGF values were defined from pregnancy data. Postpartum assessment included ophthalmic artery Doppler and carotid intima-media thickness performed by a single operator. Multivariable linear regression models assessed associations between placental dysfunction, angiogenic imbalance, and vascular parameters, adjusting for maternal covariates. 354 participants were included, 148 with and 206 without a history of preeclampsia or fetal growth restriction. Both placental dysfunction and angiogenic imbalance during pregnancy were independently associated with a significantly higher ophthalmic artery peak systolic velocity ratio 3–6 years postpartum. Participants with a history of placental dysfunction showed higher values compared to those without [0.75 (0.67–0.81) vs. 0.69 (0.63–0.78), p  = 0.03], as did those with an elevated sFlt-1/PlGF ratio during pregnancy [0.76 (0.66–0.82) vs. 0.70 (0.64–0.78), p  = 0.03]. The highest values were observed in women who had experienced both conditions. When preeclampsia and fetal growth restriction were analyzed separately, the association remained significant for preeclampsia, whereas in the fetal growth restriction group, a significant difference was observed only in the right eye. No statistically significant differences were observed in carotid intima-media thickness. Both a history of angiogenic imbalance and a clinical placental dysfunction presentation (particularly preeclampsia) during pregnancy were associated with increased ophthalmic artery peak systolic velocity ratio 3–6 years postpartum, whereas carotid intima-media thickness did not differ between the study groups.

AIM:The goal of a high specificity led to a loss of sensitivity in the new 2023 ACR/EULAR APS classification criteria, and mainly affecting obstetric clinical phenotype (OAPS). To reclassify a cohort of 1,200 women diagnosed of OAPS based on the Sydney classification criteria by applying the 2023 ACR/EULAR APS criteria. METHODS:The European Registry on Obstetric Antiphospholipid Syndrome (EUROAPS) with 1,200 women that fulfil the Sydney classification criteria were reanalysed according to the obstetric (domain 4) and laboratory domain 8 of the new classification criteria. RESULTS:When our cases were re-classified by applying the clinical criteria of domain 4, 816 cases (68%) would be excluded, leaving 384 cases (32%). The criteria that moderate to high IgM positivity is insufficient to classify patients, it led to a drop of additional 46 women, with a final number of excluded cases of 862 (71.83%), leaving only 338 cases (28.17%). Furthermore, the implementation of 40 units threshold discarded 86 women with an antiphospholipid antibody positivity above the 99th percentile cutoff. Eventually, from the 1,200 cases initially classified by the Sydney criteria, only 256 (21,33%) will remain. CONCLUSION:The 2023 ACR/EULAR classification criteria substantially reduce the proportion of women with prior obstetric APS who remain classifiable in our cohort. Although these findings highlight a marked decrease in case identification compared with the Sydney criteria, the potential clinical and research implications of this reduction warrant further evaluation in prospective studies.

Exposure to endocrine-disrupting chemicals (EDCs) during pregnancy may influence the placenta and fetal growth; however, evidence is scarce regarding EDC mixtures, newer chemicals, and the role of angiogenic biomarkers and fetoplacental hemodynamics. We aimed to examine the associations between nonpersistent EDC mixtures and fetal growth, fetoplacental hemodynamics, and angiogenic biomarkers. We included 734 pregnant participants from the Barcelona Life Study Cohort (BiSC), Spain (2018-2021). Metabolites of phthalates, DINCH, insecticides, polycyclic aromatic hydrocarbons, pesticides, flame retardants, and parent compounds of phenols and parabens were measured in pools of week-long maternal urine samples at 18 and 34 weeks' gestation. Penalized LASSO-type multigroup Bayesian Weighted Quantile Sum Regression estimated associations with fetal growth, fetoplacental hemodynamics, and angiogenic biomarkers. Birthweight z-score decreased with low-molecular-weight phthalate (LMWP) (β = -0.119; CrI -0.224, -0.008) mixtures and increased with organophosphate mixtures (β = 0.143; CrI 0.042, 0.245). LMWP exposure was also associated with altered hemodynamics and angiogenic biomarkers; angiogenic biomarkers mediated the relationship with birthweight z-score (ACME = -0.032; 95% CI -0.062, -0.009; p = 0.002). This comprehensive study suggests that mixtures of low-molecular-weight phthalates and organophosphate compounds may alter fetal growth and that angiogenic biomarkers may play a role as mediator.

BACKGROUND:Few studies have considered the complex interplay of the urban exposome in association with multiple respiratory outcomes during infancy. AIM:Utilizing an outcome-wide exposome approach, we aimed to assess associations of pregnancy urban exposome with offspring wheezing and chest infections at different time points within the first 18 months of life. METHODS:The analysis included data from 1032 mother-child pairs from the Barcelona Life Study Cohort (BiSC) (2018-2021). In total, 44 urban exposome factors were assessed during pregnancy, including air pollution, noise, temperature, humidity, green and blue spaces, and socioeconomic and lifestyle factors. Wheezing and chest infection were evaluated simultaneously at 2, 6, 12, and 18 months. We applied mixed-response sparse reduced-rank regression (MsRRR), with resampling procedures, adjusting for potential confounders. This many-to-many modelling approach identifies exposures concurrently associated with multiple interrelated outcomes. RESULTS:We found 13 exposures consistently associated with wheezing and chest infection across four time-points. Maternal education was the most consistent protective factor with odds ratios (OR) ranging from 0.65 to 0.81 for university and 0.88 to 0.96 for secondary education (vs. primary education) (all p < 0.05). Other important protective factors were size of nearest green space and maternal light-intensity physical activity. NO2 (OR 1.02-1.08), outdoor temperature (OR 1.02-1.04), and noise annoyance (OR 1.01-1.03) were consistently associated with increased risk. Area-level socioeconomic status indicators showed inverse associations. CONCLUSION:Pregnancy urban exposome may influence both wheezing and chest infections in infancy. Identifying key determinants through an outcome-wide exposome approach can inform targeted public health interventions towards more holistic urban planning strategies.

Objective: To describe first-trimester maternal, biochemical, biophysical, and ultrasound characteristics in women with recurrent pregnancy loss (RPL) compared to women without RPL. Methods: This was a retrospective cohort study analyzing data from 4440 pregnant women, including 142 women with previous RPL. Maternal and pregnancy characteristics, first-trimester biochemical markers, biophysical assessments, early-onset preeclampsia (EOPE) risk, and perinatal outcomes were compared. Results: Women with RPL were older (37.8 vs. 34.0 years, p < 0.001) and had higher rates of antiphospholipid syndrome (4.9% vs. 0.9%, p < 0.001), other thrombophilias (5.6% vs. 0.8%, p < 0.001), and thyroid disorders (14% vs. 7.5%, p = 0.010) than women without RPL. First-trimester uterine artery pulsatility index (UtA-PI) values, pregnancy-associated plasma protein-A (PAPP-A) levels, mean arterial pressure, and final risk for EOPE were comparable between groups. However, the RPL group had higher rates of very high risk for PE (10.6 vs. 5.1, p = 0.011). Likewise, second-trimester UtA-PI was higher in this group (1.10 vs. 1.01, p = 0.045). Aspirin and low molecular weight heparin prophylaxis were more frequent in women with RPL (23.8% vs. 9.6%, p < 0.001; 14.7% vs. 0.1%, p < 0.001). Regarding perinatal outcomes, we found a higher incidence of second-trimester intrauterine demise in the RPL group (6.4% vs. 1.4%, p = 0.011), with no other differences observed in the remaining outcomes. Conclusions: Women with RPL exhibit distinct maternal characteristics and worse pregnancy outcomes, although first-trimester markers do not seem to significantly differ from findings in women without RPL. These findings underscore the importance of tailored screening and intervention protocols to improve perinatal outcomes in this high-risk population.

Maternal nutrition during pregnancy is key for offspring neurodevelopment. Given the role of glucose in brain function, assessing carbohydrate quantity and quality, including glycemic index (GI), glycemic load (GL) and carbohydrate quality index (CQI), may provide insights into early brain development. This study examined the associations between maternal dietary carbohydrate intake and neurodevelopmental outcomes in early childhood. The prospective cohort study included 1080 mother–child pairs from the Barcelona Life Study Cohort. Maternal dietary carbohydrate intake, GI, GL and CQI were assessed during mid-pregnancy using a food frequency questionnaire. Child neurodevelopment was evaluated at 8 and 28 months using the Developmental Profile 3 (DP-3) and at 18 months using the Bayley Scales of Infant and Toddler Development (BSID-III). Associations were analyzed using multivariable linear regression models adjusted for relevant maternal and child covariates. Increased maternal carbohydrate intake, GI, and GL were inversely associated with language development (β (95

BACKGROUND:Exposure to non-persistent endocrine disrupting chemicals (EDCs) is widespread, causing various adverse health effects, especially during pregnancy. Diet is a major source of EDCs and the identification of dietary patterns that give rise to higher exposure levels is crucial. We aimed to explore the relationship between dietary patterns during pregnancy and exposure to non-persistent EDCs. METHODS:We included 640 pregnant women from the Barcelona Life Study Cohort (BiSC), Spain. Using food frequency questionnaires (2nd trimester), we estimated ultraprocessed food (UPF) consumption and Mediterranean diet adherence, and derived data-driven dietary patterns. Metabolites of phthalates, 1,2-cyclohexane dicarboxylic acid di(isononyl) ester (DINCH), chlorpyrifos and pyrethroid, polycyclic aromatic hydrocarbons (PAHs, including phenanthrenes), and organophosphates, and parent compounds of phenols and parabens were measured in weekly urine pools (3rd trimester). Associations between dietary patterns and EDCs were evaluated through multivariable linear regressions. RESULTS:Higher compliance with a "Vegetarian" pattern or the Mediterranean diet was associated with lower levels of some of the phthalate metabolites (e.g., di-iso-nonyl phthalates [Geometric mean ratio (GMR) per unit increases in the "Vegetarian" score = 0.97, 95 %CI 0.95, 1.00]). Conversely, "Western" and "Animal protein" patterns, and UPF intake were associated with higher PAHs (e.g., "Western" and 2-Hydroxynaphthalene [GMR(95 %CI) 1.07(1.01, 1.13)]), pesticides (e.g., "Animal protein" and 3-Phenoxybenzoic acid [1.15(1.01, 1.32]), and phthalates metabolites levels (e.g., UPF and Monoethyl phthalate [1.05(1.01, 1.09)]). No association remained significant after correcting for multiple testing. CONCLUSIONS:Following a plant-based diet may help reduce the exposure to certain phthalate and organophosphate metabolites, whereas higher consumption of foods from animal origin and high in fat/sugar/salt may increase exposure to PAHs, pesticides, and phthalates metabolites in pregnancy.

Early brain development is highly sensitive to environmental influences. While prenatal exposure to airborne particulate matter (PM2.5) has been broadly associated with harmful effects, PM2.5 also contains trace elements such as iron, copper and zinc, which are essential for brain growth. This study examined both the overall impact of prenatal PM2.5 exposure and the specific role of these trace elements on neonatal myelinated white matter-a key marker of brain maturation. This population-based study included 93 neonates recruited from three major hospitals in Barcelona (2018-2021). PM2.5 exposure was estimated for the embryonic and late fetal periods using land-use regression models incorporating time-weighted maternal mobility data. MRI was performed at 29 days postnatally. Global myelinated white matter was manually segmented, and automated cortical myelination measures were obtained in 85 cases. Associations were examined using linear regression models with and without adjustment for potential confounders. Higher prenatal PM2.5 exposure was associated with lower myelinated white matter content. Trace elements showed a similar pattern, but their associations became nonsignificant after adjusting for overall PM2.5 exposure. The findings suggest that prenatal air pollution exposure may delay early myelination. Moreover, no specific associations were identified for iron, copper, or zinc. However, given the dynamic nature of white matter maturation, such delays may not necessarily be detrimental. This study underscores the impact of environmental factors on neonatal brain development and the importance of stringent air quality policies, while emphasizing the need for longitudinal research to assess long-term cognitive and behavioral outcomes.

BACKGROUND:Synthetic phenols are widely used chemicals with potential neurodevelopmental toxicity. Human studies are often limited by small sample sizes and exposure misclassification. Identifying the biological pathways affected by these substances is crucial for understanding key drivers of toxicity. We aimed to study associations between prenatal exposure to synthetic phenols and child behaviour, exploring the potential mediating role of maternal steroid hormones. METHODS:We pooled data from two European cohorts: the Barcelona Life Study Cohort (BiSC; Barcelona, Spain, 2018-21, N=1080) and Suivi de l'Exposition à la Pollution Atmosphérique durant la Grossesse et Effets sur la Santé (SEPAGES; Grenoble, France, 2014-17, N=484). Mothers older than 18 years having a singleton pregnancy of less than 19 weeks gestational age were eligible for inclusion in the cohorts; those having multiple pregnancies were excluded. Repeated urine samples (up to 24 in BiSC; up to 42 in SEPAGES) collected in the second and third trimesters were pooled and analysed for 12 synthetic phenols. Child behavioural outcomes were assessed at 18 months in BiSC and 24 months in SEPAGES using the Child Behavior Checklist for Ages 1·5-5 (CBCL). Concentrations of total cortisol, total cortisone, and 11-dehydrocorticosterone-steroid hormones involved in the hypothalamic-pituitary-adrenal axis-were measured from maternal hair samples. Associations between phenol exposure and behavioural outcomes were estimated by adjusted linear regression analysis, and mediation by steroid hormones was assessed with regression-based causal mediation analysis within the counterfactual framework. FINDINGS:1024 mother-child pairs were included in the study: 607 from BiSC and 417 from SEPAGES. Maternal exposure to methylparaben in the third trimester of pregnancy was associated with higher internalising scores (change in score of 0·44 [95% CI 0·10-0·79] points) and externalising scores (0·67 [0·12-1·24]) in the CBCL. In boys, maternal exposure to bisphenol S in the third trimester was linked to increased internalising scores (0·92 [0·15 to 1·75]; p=0·019) and could be linked to increased externalising scores (1·14 [-0·09 to 2·44]; p=0·070). In girls, second-trimester butylparaben and propylparaben exposure were associated with lower internalising (-1·03 [-1·84 to -0·09], p=0·033) and externalising (-0·68 [-1·23 to -0·12]; p=0·019) scores. No mediation by steroid hormones was observed. INTERPRETATION:Prenatal exposure to phenols might influence early behavioural development, with sex-specific patterns. There was no strong evidence of mediation by maternal steroid hormones, suggesting the involvement of alternative pathways in the biological effects of phenols. Together with previous findings, these results highlight the need for stricter regulation of these compounds to reduce prenatal exposure. FUNDING:French Agency for Food, Environmental and Occupational Health & Safety, the EU's Horizon 2020 research and innovation programme, and the French Fund - Fondation de France.