Geographic Location and Its Contribution to Disparities in Ovarian Cancer Survival

Abstract

Background: Over 14,000 American women die of ovarian cancer (OC) every year. Disparities in survival are observed by race and socioeconomic status (SES), even after adjusting for treatment received. Geographic location has been identified as an independent predictor of survival. This study aims to determine the impact of geographic location on advanced-stage OC survival in California (CA, USA), relative to race and SES.Methods: Women diagnosed with advanced-stage epithelial OC between 1996 and 2014 were identified through the CA Cancer Registry, with follow up obtained through 2016. Cox proportional hazard models with a smooth for residential location were used to identify geographic patterns in survival. We assessed the impact of distance traveled for care, distance to closest high-quality-of-care (QOC) hospital, race, and SES on survival, adjusting for treatment received, age, and cancer characteristics.Results: Among the 20,095 women diagnosed at late stages, survival time ranged from 0 to 20.2 years, with a median of 2.0 years. Median survival time was shortest for women of Black race (1.3 years) and the lowest SES (1.5 years). Patients who traveled the greatest distance to receive care (>32km) had better survival (hazard ratio [HR],0.85; 95% confidence interval [CI],0.80-0.90), whereas women living furthest (> 48km) from a high-QOC hospital had poorer survival (HR,1.06; 95% CI,1.00-1.14). Women of lower SES and minority race were less likely to travel far for care. Geographic location was a significant predictor of survival, even after controlling for cancer characteristics and treatment received, with Central Valley showing a significant hot spot of high HR that was attenuated after adjusting for race and SES.Conclusions: Geographic location is an independent predictor of OC mortality in CA. Traveling greater distances for care was associated with better survival. Minority women and those of lower SES are disproportionately affected by geographic barriers.