Association of preconception to early childhood residential greenness with child neurodevelopmental outcomes and the mediating role of air pollution: a prospective cohort study.

Prospective evidence on the associations of residential greenness from preconception to early childhood with child neurodevelopmental outcomes remains limited, and air pollution's mediating role in these associations is unclear. We conducted a cohort study of 5,284 singleton mother-child pairs from the Jiangsu Birth Cohort. Residential greenness was estimated using the Normalized Difference Vegetation Index (NDVI) derived from satellite imagery. Child neurodevelopmental outcomes were assessed at age three using the Bayley-III Screening Test (five domains: cognition, receptive communication, expressive communication, fine motor, and gross motor), with children classified as "at-risk", "emerging", or "competent"; the "at-risk" and "emerging" groups were pooled into the "emerging/at-risk" group per domain. Robust Poisson regression assessed the associations between NDVI and the risk of the emerging/at-risk group for each neurodevelopmental domain. Separate single-pollutant mediation analyses evaluated potential indirect associations through PM2.5, PM10, NO2, and O3. Higher NDVI during preconception, pregnancy, early childhood, and each trimester was associated with lower risks of the emerging/at-risk group across the five neurodevelopmental domains. Specifically, each 0.1-unit increase in preconception and third-trimester NDVI was associated with lower risks of the emerging/at-risk group for cognition [relative risk (RR), 95% confidence interval (CI): 0.85 (0.74, 0.98), 0.82 (0.71, 0.95)] and gross motor [RR (95% CI): 0.87 (0.77, 0.98), 0.77 (0.68, 0.88)]. Following false discovery rate (FDR) correction (threshold 0.10), favorable associations of NDVI with multiple neurodevelopmental domains persisted for preconception and third-trimester exposure. Mediation analyses showed mediated proportions of 68.0% and 63.5% through PM10 and NO2, respectively, for the association of preconception exposure with cognition, and 59.1%, 46.5%, 34.0%, and 39.3% through PM2.5, PM10, NO2, and O3, respectively, for the association of third-trimester exposure with gross motor. These findings support a favorable association between early-life residential greenness across multiple exposure windows and child neurodevelopmental outcomes, and lower air-pollution exposure may partly mediate these associations.