Socioeconomic Disparities in Supportive Care Medication Use Among US Medicare Beneficiaries with Pancreatic Cancer.

IntroductionPancreatic cancer is associated with substantial symptom burden, but access to supportive care medications (SCMs) may vary by socioeconomic context. We examined associations between neighborhood socioeconomic status (SES) and the receipt and timing of SCMs among older adults with pancreatic cancer in the United States.MethodsWe conducted a retrospective cohort study using SEER-Medicare data (from 2005-2020). The cohort included beneficiaries aged 65 years or older with incident pancreatic cancer and Part D coverage. SCMs included psychotropics, pain medications, anti-nausea agents, appetite stimulants, and sleep aids. Neighborhood SES was measured using Yost Index quintiles (with Q5 representing the highest SES). Multivariable logistic regression estimated associations with medication receipt, and Fine-Gray competing risk models estimated associations with the cumulative incidence of medication initiation, treating death as a competing event.ResultsAmong 81,535 beneficiaries, 69% received at least one SCM after diagnosis. Compared with beneficiaries in Q5, those in Q1 had lower adjusted odds of receiving antipsychotics (aOR 0.81; 95% CI 0.68-0.96), antidepressants (aOR 0.80; 95% CI 0.73-0.88), anxiolytics (aOR 0.87; 95% CI 0.78-0.97), anti-nausea agents (aOR 0.77; 95% CI 0.71-0.83), appetite stimulants (aOR 0.78; 95% CI 0.72-0.85), and sleep aids (aOR 0.67; 95% CI 0.60-0.76). Lower neighborhood SES was also associated with lower cumulative incidence of psychotropic, anti-nausea, appetite stimulant, and sleep aid initiation. Associations with pain medication receipt and initiation were smaller and less consistent.ConclusionLower neighborhood SES was associated with lower receipt and cumulative incidence of several SCM classes among older adults with pancreatic cancer. Because SEER-Medicare does not capture symptom burden or clinical indication, these findings describe medications dispensed through Part D rather than the adequacy of supportive care received and identify targets for research on the mechanisms underlying these differences and on interventions to improve equitable symptom management.