Association of individual low-income status and area deprivation with mortality in multiple myeloma.

INTRODUCTION: Lower individual-level socioeconomic status (SES) and area-level SES have each been associated with poor survival outcomes among patients with multiple myeloma (MM). A body of literature suggests that individual-level SES may be differentially associated with mortality depending on area-level SES, and vice versa. This study assessed the effect of the cross-level interaction between individual low-income status and area deprivation on mortality among patients with MM. MATERIALS AND METHODS: This retrospective cohort study used the Surveillance, Epidemiology, and End Results (SEER)-Medicare data (2006-2016). Individuals were defined as having low income if they were dually eligible for Medicare and Medicaid and/or if they received the Low-Income Subsidy. The county-level Social Deprivation Index (SDI) was linked to individual-level SEER-Medicare data and categorized into quintiles, from the least deprived (Quintile 1) to the most deprived (Quintile 5). Adjusted hazard ratios (HRs) for the associations between low-income status, area deprivation, and all-cause mortality were estimated from a mixed-effects Cox proportional-hazards (PH) model. RESULTS: The mortality hazard was higher for individuals with low income than individuals without low income in all quintiles of area deprivation, with the exception of Quintile 5 (Quintile 1: HR 1.53 [95% confidence interval [CI]: 1.32-1.77]; Quintile 2: HR 1.17 [95%CI: 1.01-1.36]; Quintile 3: HR 1.34 [95%CI: 1.18-1.53]; Quintile 4: HR 1.33 [95%CI: 1.17-1.52]; Quintile 5: HR 1.09 [95%CI: 0.96-1.23]). Among individuals without low income, individuals residing in the most deprived area had a higher mortality hazard than individuals residing in the least deprived area (HR: 1.22 [95%CI: 1.03-1.45]). In contrast, among individuals with low income, residing in a more deprived area, Quintile 2, was associated with a lower hazard of death than residing in the least deprived area, Quintile 1 (HR: 0.82 [95%CI: 0.67-0.99]), and there was no statistically significant difference between Quintile 1 and Quintiles 3, 4, and 5. DISCUSSION: In this analysis, there was a statistically significant cross-level interaction between individual low-income status and area deprivation on mortality. More research is needed to fully understand the mechanism behind these associations, but the findings show that patients and their health should be considered in the context of where they live.