Spatial heterogeneity in area-level environmental context of hospital-based congenital structural anomaly burden in Southwest China: a retrospective study from a provincial pediatric referral center.

BACKGROUND: Congenital structural anomalies constitute a major source of pediatric morbidity and surgical burden. However, evidence regarding the spatial epidemiology of congenital structural anomalies and their area-level environmental context remains limited, particularly in geographically and socioeconomically heterogeneous regions. This study aimed to investigate the spatial heterogeneity of area-level environmental contextual associations with hospital-based congenital anomaly burden. METHODS: We conducted a retrospective hospital-based study including pediatric inpatients with congenital structural anomalies admitted to the largest provincial pediatric referral center in Yunnan Province between 2014 and 2024. Descriptive epidemiological analyses were performed to summarize demographic characteristics, temporal trends, and system-level distributions. High-frequency congenital anomalies were selected to construct stable county-level datasets for spatial analysis. Ordinary least squares (OLS) and geographically weighted regression (GWR) models were applied to examine spatially varying, area-level contextual associations between environmental factors and anomaly burden across five congenital anomaly systems (circulatory, digestive, urogenital, musculoskeletal, and other structural anomalies). RESULTS: A total of 56,434 pediatric inpatients with congenital structural anomalies were identified. Among pediatric inpatients, boys accounted for the majority (67.68%), and admissions were predominantly concentrated in early childhood. Digestive (29.63%) and urogenital (23.30%) anomalies accounted for the largest proportions of pediatric inpatients. Among 41,531 pediatric inpatients included in spatial modeling, GWR consistently demonstrated better performance than global OLS models across five congenital anomaly systems. Area-level environmental associations with referral-weighted institutional burden showed substantial spatial heterogeneity. Carbon monoxide demonstrated predominantly positive associations across anomaly systems, whereas sulfur dioxide showed pronounced spatial heterogeneity. Vegetation coverage showed consistent negative associations, while population density displayed geographically varying associations. CONCLUSIONS: Hospital-based congenital structural anomaly burden showed a substantial spatial heterogeneity in its area-level environmental context within a provincial pediatric referral center dataset. Spatially explicit approaches such as GWR may help characterize referral-weighted, area-level patterns of institutional burden beyond global models. These findings are descriptive and hypothesis-generating and require validation using population-based or multi-center studies.