A Quantitative and Spatial Analysis of Pancreatic Cancer in South Central Appalachia

Pancreatic cancer (PC) has a grim prognosis and is the third leading cause of cancer death in the United States. Patients with the best prognosis are those who are found with the disease at a stage early enough to be surgically resectable. No current strategies to identify PC early, outside of specific high-risk individuals, exist. This study described population demographics for PC patients in South Central Appalachia, identified healthcare and demographic variables that will aid in early PC identification, and assessed patient location as a significant independent variable in predicting PC outcomes.

A retrospective analysis of tumor registry and electronic health record data on PC patients at an academic medical center in South Central Appalachia with a high-volume pancreas service was completed. Data from 2010 through 2020 were analyzed. Statistical analysis for descriptive and inferential statistics were completed using SPSS v. 29 for Macintosh. Spatial analysis was conducted using ArcGIS Pro 3.0.

The sample revealed a total of 931 individuals with pancreatic cancer. The mean age of diagnosis was 67.0; 52% were male, 89.8% of individuals identified as white. 69.8% presented with Stage IIb, III, or IV disease. The mean distance to the treatment facility was 56.6 miles, the mean driving time was 49.8 minutes. Significant variables associated with disease prognosis include distance to the facility; having a primary care provider, having insurance, and mortality could be predicted both by stage of disease at the time of diagnosis and insurance status. ix

Distance to the high-volume PC treatment center was found to serve as facilitator to an earlier time to treatment and early stage of diagnosis when patients live closer to the high-volume treatment center.

Secondary analysis of regional PC patient characteristics can be informative in identifying non-biological variables associated with disparate outcomes. This regional perspective can be valuable in identifying areas for healthcare resource investment to improve outcomes. Knowledge from this study will serve future nurse scientists interested in PC, and the methodology used by this study remains a powerful tool for future nurse scientists to use across the healthcare spectrum to understand how spatial location influences health outcomes.