Assessing socioeconomic disparities and the likelihood of multimodal treatment utilization in patients with pancreatic cancer using SEER Medicare data.

Abstract

677 Background: Optimal treatment for pancreatic cancer necessitates multimodal therapies combining surgery, chemotherapy, and radiation. Socioeconomic determinants of health may influence access to these treatments, leading to disparities in outcomes. This study examines the association between socioeconomic status (SES) and the likelihood of receiving multimodal treatment among pancreatic cancer patients using SEER Medicare data. Methods: Our study population consists of elderly persons diagnosed with pancreatic cancer from the 2008–2017 SEER-Medicare dataset. A multinomial logistic regression controlling for sociodemographic characteristics, cancer stage, comorbidities, geographic region, and ecological measures of urbanicity and SES was used to assess the likelihood of receiving one of three mutually exclusive treatment categories: no treatment, one modality (surgery, chemotherapy, or radiotherapy), or multiple modalities (different combinations of surgery, chemotherapy, and radiotherapy), with multiple modalities as the reference group. Survival analysis was conducted using Kaplan-Meier curves (KMC) and Cox proportional hazards (CPH) models. Results: Adjusted analyses showed that among stages 0–3 patients, living in areas with a high percentage of individuals without a high school education (non-HS) was associated with increased odds of receiving no treatment (OR = 1.26, p < 0.01) and one modality (OR = 1.20, p < 0.01) versus multiple modalities. Low median household income was linked to higher odds of no treatment (OR = 1.30, p < 0.01) and one modality (OR = 1.16, p < 0.05). This pattern persisted in stage 4 patients, where high non-HS percentage increased odds of no treatment (OR = 1.38, p < 0.01), and low median income was associated with higher odds of no treatment (OR = 1.25, p < 0.05). Non-Hispanic Black patients had higher odds of receiving no treatment in both stages 0–3 (OR = 1.26, p < 0.05) and stage 4 (OR = 1.91, p < 0.01) compared to Non-Hispanic Whites. Being married was associated with reduced odds of receiving no treatment for stages 0–3 (OR = 0.53, p < 0.01) and stage 4 (OR = 0.55, p < 0.01). Lower SES and being Non-Hispanic Black are associated with decreased likelihood of receiving multimodal treatment across all stages. Patients receiving multimodal therapies demonstrated improved survival on our KMC analysis, and CPH models indicated lower mortality for those receiving multiple modalities. Conclusions: Socioeconomic disparities significantly influence multimodal treatment utilization in pancreatic cancer. Lower SES indicators—high non-HS percentages, low median income—and being Non-Hispanic Black are associated with decreased use of multimodal therapies and poorer survival. Addressing these disparities is essential to ensure equitable access to optimal treatments and improve survival rates.