Effects of neighborhood deprivation index on survival in gastric cancers.
Abstract
336 Background: Previous studies have shown that living in disadvantaged neighborhoods resulted in poor outcomes in several malignancies. Neighborhood Deprivation Index (NDI) identifies key variables from 13 measures, including the following dimensions of socioeconomic (SES) status: wealth and income, education, occupation, and housing conditions. In this study, we aimed to investigate the effect of NDI among patients with gastric cancer (GC). Methods: We conducted a retrospective analysis using the Surveillance, Epidemiology, and End Results (SEER) database from 1996-2015. NDI was divided into two groups: less disadvantaged areas (NDI <60) and highly disadvantaged areas (NDI >60), respectively. Demographic and clinical characteristics were summarized by NDI groups. All associations were compared using Kruskal-Wallis and Chi-Square tests, respectively. Kaplan Meier curves were summarized by the NDI group for overall survival (OS) and disease-specific survival (DSS). All associations were compared using a log-rank test. Cox multivariate regression was performed to measure the association between NDI and OS/DSS. All analyses were conducted in RStudio v4.2.3 at a significance level of ≤ 0.05. Results: A total of 19,205 GC (11,137 patients with NDI<60 and 8,068 NDI>60) patients were analyzed. Disadvantaged areas (NDI>60) had a higher proportion of Black and Hispanic, single, uninsured, rural patients with later disease stage as compared to NDI <60. Median OS was 10.0 months for patients with NDI <60 versus 9.0 months for those with NDI >60 (p=0.0007). Median DSS was 12.0 months for NDI <60 and 10.0 months for NDI >60 (p=0.010). Multivariate analysis revealed that factors associated with worse OS included NDI >60 [HR 1.044, p=0.009], age (>60) [HR 1.28, p<0.001], male sex [HR 1.082, p<0.001], single marital status [HR 1.21, p<0.001], uninsured status [HR 1.13], p <0.001], advanced disease stage (III/IV) [HR 2.05, p<0.001], and higher grade [HR 1.32, p<0.001]. On the other hand, non-Hispanic Black [HR 0.98] and Hispanic [HR 0.94] race and urban location [HR 0.91, p<0.001] were associated with improved OS. Factors associated with inferior DSS included NDI >60 [HR 1.037, p=0.038], age (p<0.001), male sex [HR 1.05, p=0.005], single marital status [HR 1.15, p<0.001], advanced disease stage (III/IV) [HR 2.375, p<0.001], and advanced grade (III/IV) [HR 1.40, p<0.001). In contrast, Hispanic [HR 0.94], non-Hispanic Black [0.96], and urban location [HR 0.90, p<0.001] were associated with improved DSS. Conclusions: Patients from areas with higher NDI had worse OS and DSS for GC. Tailored healthcare policies that aim to improve healthcare delivery to areas with high deprivation may help improve GC outcomes.
Article Details
Journal Info
Journal of Clinical Oncology
Lippincott Williams & Wilkins
Authors (5)
Sawyer Bawek
1Cleveland Clinic, Cleveland, United States
Mrinalini Ramesh
University at Buffalo, Buffalo, NY
Kayla Catalfamo
1Roswell Park Comprehensive Cancer Center, Buffalo, United States
Kristopher Attwood
NRG Oncology, Philadelphia, PA
Sarbajit Mukherjee
Department of Medicine, Roswell Park Comprehensive Cancer Center , Buffalo, NY,