Geospatial analysis of incidence and mortality rates of gastrointestinal cancer and socioeconomic status across Puerto Rico during 2013-2022

Colorectal (CRC), gastric, and pancreatic cancers are classified as gastrointestinal (GI) malignancies. Collectively, they rank among the top ten leading causes of cancer-related death in Puerto Rico (PR) for both men and women with CRC being the second most diagnosed and second leading cause of cancer death in PR. This study examines spatial and temporal variations of GI cancer incidence and mortality across municipalities in PR, while evaluating the role of sociodemographic factors and major disruptions -such as Hurricanes Irma and Maria (2017), subsequent earthquakes, and the COVID-19 pandemic- on these patterns across two periods. Age-adjusted incidence rates (AAIRs) and age-adjusted mortality rates (AAMRs) at municipality level were obtained from the PR Central Cancer Registry for 2013-2017 and 2018-2022. Sociodemographic data came from the American Community Survey. Global spatial autocorrelation was assessed using Global Moran's Index (I) and visualized through choropleth maps (Jenks Natural Breaks). Getis-Ord Gi* analysis identified clusters of municipalities with high-incidence or mortality (hot spots) and low-incidence or mortality (cold spots), using Bonferroni correction at a 90% confidence level. Wilcoxon rank-sum tests compared AAIRs, AAMRs, and socioeconomic indicators between periods and spot classifications. Statistical significance was set at p<0.05. Analyses were conducted by period using R (version 4.4.1) and R-Studio. Spatial clustering of AAIRs was significant across PR for the full period (Moran's I 0.17, p<0.05). Clustering in 2013-2017 was marginal (Moran's I 0.08, p=0.1) but became pronounced in 2018-2022 (Moran's I 0.19, p<0.05). Furthermore, AAMRs showed positive spatial autocorrelation for 2013-2017 period (Moran's I 0.22, p<0.001). In 2013-2017, 11 municipalities were hot spots and 12 were cold spots. In 2018-2022, 10 hot spots emerged-Juana Díaz and Guánica persisted across both periods while nine new cold spots appeared with no overlap. In 2013-2017 period, hot spots had 13% higher AAMRs than cold spots (p<0.001), while hot spots had significantly higher AAIRs in the latter period. Socioeconomic status did not correlate with hot or cold spots from 2013-2017; however, cold spots showed significantly higher socioeconomic status than hot spots in the 2018-2022 period: lower household income, and higher rates of poverty and SNAP participation (p<0.05). GI cancer in Puerto Rico exhibits persistent spatial clustering: AAIRs intensified in 2018-2022 and AAMRs shifted from higher AAMRs in hot spots during 2013-2017 to an income-related gradient in 2018-2022, where higher socioeconomic levels were linked to lower mortality. These findings suggest that systemic disruptions such as natural disasters and the COVID-19 pandemic may have deepened existing inequalities in vulnerable regions of Puerto Rico.