距离与剥夺：地理和情境因素对儿童癌症结局的影响

Introduction: Cancer is the number one cause of death by disease in children in the US and while survival rates have improved significantly, differences persist in pediatric cancer outcomes by a number of factors. The objective of this study was to demonstrate how cancer registry data can be used to assess social and geospatial predictors of pediatric cancer mortality and their combined effects. Methods: This population-based longitudinal study used data from the Louisiana Tumor Registry and Iowa Cancer Registry and included all individuals aged 0-19 at primary cancer diagnosis from 2000 to 2020. Travel distance was estimated using the great-circle distance method in ArcGIS Pro, neighborhood deprivation was characterized with the Area Deprivation Index, and rurality was measured at the census tract level with Rural Urban Commuting Area codes. Cox regression models were used to assess the association between social and geospatial predictors and pediatric cancer mortality, adjusting for confounding variables identified through a directed acyclic graph. Results: Of the 6,982 children with cancer identified from the two states, 4,871 had complete information on home addresses at diagnosis and treatment facilities and were included in the final analysis. Of the cases from Iowa, 40.31% resided in rural census tracts, compared to 15.00% of cases from Louisiana. Five-year survival was higher for children in Iowa (86.63%, 95% CI: 85.19-88.14%) than Louisiana (83.41%, 95% CI: 82.00-84.88%). Compared to children with low travel distances and low neighborhood deprivation, those experiencing both high travel distance (distance in the fourth quartile) and high neighborhood deprivation (ADI in top 50%) have an 82% high hazard of cancer mortality (95% CI: 1.43-2.32). When looking at the combined effects of rurality and travel distance, mortality differences appeared to be driven more by distance than rurality. Interestingly, compared to children in urban areas with a low travel distance to treatment, the hazard of death was highest for children with high travel distances residing in urban areas (aHR=1.41, 95% CI: 1.17-1.71), followed by children with high travel distances in rural areas (aHR=1.28, 95% CI: 1.01-1.62). Conclusions: By integrating cancer registry data with geospatial and neighborhood social and economic indicators, we identified that children facing both long travel distances and high neighborhood deprivation experience significantly elevated hazards of mortality. Importantly, geography and context are not just a backdrop, but a determinant of survival. These findings highlight the need for multilevel interventions that target individual, hospital, neighborhood, and policy drivers of differences in cancer outcomes for children.