Regional disparities and socioeconomic determinants of pediatric lymphoid leukemia mortality in Brazil from 2000-2021

Abstract Introduction: Leukemias are the most frequent cancers in childhood and adolescence, and socioeconomic factors are known to be related to their outcomes. In Brazil, regional disparities in pediatric mortality from lymphoid leukemias (LL) are observed across states. The factors involved in these disparities are not yet fully understood. This study aims to describe the spatiotemporal distribution of LL mortality in the Brazilian pediatric population and understand the correlation between socioeconomic factors and this mortality across the country's 27 states. Methods: This is an ecological study which included the Brazilian population of individuals aged zero to 19 years in a state level from 2000 to 2021, divided into four age groups: 0-4y (zero to four), 5-9y (five to nine), 10-14y (10 to 14), and 15-19y (15 to 19 years). Number of deaths by LL (ICD-10 code C91) in states was obtained from the Brazilian Ministry of Health's open-access database. Eleven socioeconomic-related characteristics were selected and grouped into factors using principal component analysis (PCA). Dynamic generalized additive mixed models (GAMM) were used to explore the association between the PCA components and the age-specific mortality rate (ASMR) by LL in states over time. Results: PCA resulted in two main factors which were interpretated as: F1–economics and health market growth and F2–primary health care assistance coverage. Results showed that mortality from LL in Brazil has increased over time, particularly in northern (N) and northeastern (NE) states. In contrast, southern (S) and southeastern (SE) states showed stability or decreasing rates. The economy and health market growth was associated with an ASMR increase in contexts of worse socioeconomic conditions and with a decrease in contexts of better socioeconomic conditions. This decrease was observed earlier in S and SE states, which have better socioeconomic indicators. The primary health care assistance coverage contributed to an overall decrease in the ASMR (Coef.=-0.03, p=0.4), and, although not statistically significant, was kept in the analysis for interpretation. Conclusion: These findings showed that Brazil experienced an increase in pediatric LL mortality from 2000-2021, despite the reduction effects of economy and health market growth, and primary health care assistance. The mortality disparities observed between Brazilian states are linked to socioeconomic factors. States under poorer socioeconomic conditions experienced a higher increase in LL mortality, and the benefits of economic and health market growth took longer to manifest there. Conversely, strengthened primary health care services were associated with an earlier and more pronounced reduction in LL mortality in the North and Northeast regions, despite these regions' socioeconomic disadvantages. This protective effect likely reflects expanded primary health care coverage driven by government programs targeting vulnerable populations during the study period. Further research is needed to explore individual factors contributing to the observed increase in pediatric lymphoid leukemia mortality in Brazil.