Neuroprogramming of prenatal phthalate exposures on fluid cognition: A latent variable modeling approach to quantify exposure burden and integrate neurobehavioral data

Background: Phthalates are endocrine-disrupting chemicals with neuroactive properties linked to maladaptive neurodevelopment in children. However, few studies have utilized latent variable methodologies to estimate their cumulative impact and assess the complex integration of cognitive processes that characterize fluid cognition—the ability to efficiently process, manipulate, and integrate information to solve reasoning problems. Objective: We investigated the prenatal trimester-specific neuroprogramming effects of the phthalate burden scores on fluid cognition in Mexican children. Methods: Children (n = 626) aged 6–7 years from a prospective pregnancy cohort in Mexico City were administered subtests from the CANTAB, completing the between error, strategy, and mean latency measures intended to evaluate a broad spectrum of cognitive domains representative of fluid cognition. Phthalate metabolites were measured in maternal urine collected at 2nd and 3rd pregnancy trimesters. A CFA validated and quantified two correlated latent phthalate burden scores representing prenatal exposure to low molecular weight (LMW) and high molecular weight (HMW) phthalates. Trimester-specific models using a covariate-adjusted SEM estimated the associations of latent phthalate burden scores with a latent construct of fluid cognition, an integration of working memory, executive function, and attention tasks. Results: In the 3rd trimester, higher LMW phthalate burden was associated with poorer fluid cognition (b = −1.860; [95 % CI = −3.505, −0.215]; p = 0.027), while HMW phthalate burden showed a positive association (b = 1.815; [95 % CI = 0.176, 3.453]; p = 0.030). Conversely, in the 2nd trimester, neither burden levels of LMW (b = −0.508; [95 % CI = −1.639, 0.623]; p = 0.378) nor HMW (b = 0.451; [95 % CI = −0.671, 1.573]; p = 0.431]; p = 0.44) phthalate demonstrated significant associations with fluid cognitive performance. Conclusion: The temporal sensitivity of prenatal phthalate exposures on fluid cognition showed effects in later stages, with higher LMW burden linked to poorer performance and HMW burden showing a positive association. Our findings emphasize latent variable approaches and the need for more research on exposure-driven integrated cognitive programming.