Young children in the United States spend a significant portion of their waking hours in out-of-home care, making chemical exposures in early care and education (ECE) environments a major concern. Children's vulnerability to adverse health effects from pesticide exposure during critical developmental windows is well-documented, with studies linking early-life exposure to respiratory problems, decreased lung function, neurological and behavioral developmental impacts, and increased leukemia and lymphoma risk. The extensive use of pesticides in agriculture and structural pest control (over one billion pounds annually in the US) contributes to environmental contamination. While prior research has detected pesticides in ECE facilities, quantitative assessments of the predictors of pesticide levels in these environments are lacking. Indoor dust, particularly carpet dust, serves as a crucial indicator of long-term indoor exposure, accumulating contaminants over time and providing a robust measure of exposure risk. In California, policies like the Healthy Schools Act aim to reduce children's pesticide exposures by limiting agricultural applications near schoolsites and promoting integrated pest management (IPM). This study aims to identify behavioral and environmental determinants of pesticide concentrations and loadings in carpet dust from ECE centers in Northern California, hypothesizing that factors such as proximity to agricultural pesticide applications, on-site pesticide storage, IPM practices, building age, professional pesticide applications, carpet placement, and observed pest infestations would influence pesticide levels.

Several studies have previously highlighted the presence of pesticides in early care and education facilities. Research has shown detectable levels of pesticides in indoor floor wipe samples and dust samples from various locations across the US. For example, studies in North Carolina and Ohio found detectable pesticide levels in all dust samples analyzed from childcare centers. However, a critical gap exists in the literature regarding a quantitative assessment of the predictors for pesticide levels in these facilities. Existing research has largely focused on residential environments, highlighting the influence of factors such as nearby agricultural pesticide use and individual behaviors on pesticide exposures. A systematic review highlighted key determinants like housekeeping practices and proximity to agricultural fields. Studies have shown associations between living near agricultural fields, home carpeting, professional pest control treatments, and higher pesticide detection rates in residential dust. Conversely, daily cleaning and doormats have been associated with lower pesticide concentrations.

This study utilized baseline data from the University of California, San Francisco, Healthy Children & Environments Study (HCES), a randomized-control trial examining the impact of an IPM intervention on pesticide exposure and health risks. Carpet dust samples were collected from 51 licensed ECE centers across four Northern California counties (two in the San Francisco Bay Area and two in the San Joaquin Valley) during 2017-2019. The study included a mix of program types (private, non-profit, Head Start, California State Preschool Programs, and blended funding) and sizes (10-200 children). Data collection involved director interviews and observational checklists (IPM Checklist and Health and Safety Checklist), assessing facility characteristics, pest problems, pesticide use practices, and IPM policies. Dust samples were collected using a high-volume surface sampler from areas where children spend time (circle time/nap time). Samples were analyzed at Southwest Research Institute (SwRI) for 14 pesticides using gas chromatography/mass spectrometry (GC/MS). Pesticide concentrations (ng/g) and loadings (ng/m²) were calculated. California Department of Pesticide Regulation (DPR) data on agricultural and structural pesticide applications within a 3-km radius of each center were incorporated into a geospatial analysis. Statistical analyses, including Spearman's correlation and Tobit multivariable regression models, were used to assess associations between pesticide levels and predictor variables (agricultural pesticide use, PMP applications, observed pesticide products, IPM score, pest observations, geographic region, and carpet placement). Values below the limit of detection (LOD) were imputed as LOD/√2 before log transformation.

All 51 ECE centers had at least one detectable pesticide in their carpet dust samples. The five most frequently detected pesticides were cis-permethrin (98%), trans-permethrin (98%), bifenthrin (94%), fipronil (94%), and chlorpyrifos (88%). Bifenthrin concentrations and loadings showed significant positive correlations with agricultural bifenthrin applications within 3 km. Fipronil concentrations and loadings were positively correlated with professional fipronil applications reported to DPR. Higher IPM Checklist scores correlated with lower chlorpyrifos concentrations. Multivariable models revealed that location in the San Joaquin Valley was the strongest predictor of higher loadings for the most frequently detected pesticides. Specifically, being located in the San Joaquin Valley was associated with substantially higher bifenthrin, chlorpyrifos, fipronil, and permethrin loadings. Placement of the sampled carpet on carpeted flooring or base carpeting was associated with lower chlorpyrifos loading. Higher IPM Checklist scores were associated with lower permethrin and chlorpyrifos loadings, although these effects were modest. The median concentrations of pesticides found were similar to that of a similar study in 2012, but higher for some pesticides such as bifenthrin and cypermethrin, and lower for others such as chlorpyrifos, suggesting changing patterns of pesticide use.

The study's findings highlight the ubiquitous presence of pesticides in the environments of young children. The strong association between pesticide loadings and geographic location in the San Joaquin Valley emphasizes the influence of large-scale agricultural pesticide use on children's exposure. The positive correlation between bifenthrin levels and nearby agricultural applications supports findings from residential settings. The lack of correlation between chlorpyrifos levels and agricultural use may relate to the long half-life of chlorpyrifos, necessitating the investigation of pesticide applications over extended periods. The association between IPM practices and lower pesticide loadings suggests that implementing robust IPM programs in ECE settings could effectively reduce pesticide exposures, even to legacy pesticides. The study's limitations include the inability to fully examine other predictors like daily cleaning practices and the cross-sectional nature of the data, preventing causal inference. Despite various cleaning practices, measurable pesticide concentrations were present in all centers, highlighting the persistent nature of these contaminants. This study contributes critical information toward understanding the factors influencing pesticide exposures in ECE settings.

This study provides valuable insights into the environmental and behavioral determinants of pesticide contamination in California childcare centers. The strong association between geographic location, specifically the San Joaquin Valley, and high pesticide levels underscores the critical importance of addressing agricultural pesticide use near these facilities. The positive influence of IPM practices on reducing pesticide exposures highlights the need for widespread adoption of these strategies in ECE settings. Future research should explore the long-term health effects of these exposures, examine additional environmental factors, and evaluate the effectiveness of various IPM interventions in reducing pesticide exposure among young children.

The cross-sectional nature of the study design limits the ability to establish causal relationships between the predictors and pesticide levels. The study did not fully assess all potentially relevant factors influencing pesticide contamination, such as the frequency of specific cleaning practices or the type of pest control methods employed. The generalizability of the findings may be limited to similar ECE settings in Northern California. The imputation of values below the LOD might introduce uncertainty into the analysis. Finally, the study relies on self-reported data for some variables, which may be subject to recall bias.