Impact of neighborhood disadvantage on stage at diagnosis and outcomes in women diagnosed with breast cancer.

Abstract

e18603 Background: Low socioeconomic status (SES) has been associated with higher incidence of high-risk breast cancer and poorer outcomes. However, previous studies have inconsistent definitions of SES. Using a more comprehensive definition of SES via the Neighborhood Atlas, which calculates one's relative level of Neighborhood Disadvantage (ND) using 17 variables from American Community Survey and US Census data, may address this problem. We aimed to determine if higher level of ND is associated with later stage breast cancer at diagnosis and/or higher five-year recurrence rates. Methods: We performed a retrospective chart review study of 418 women that were diagnosed with breast cancer in 2015 within the Froedtert Health System in WI. Data collected included: patient characteristics (age, race, BMI, menopausal status, genetic status), tumor characteristics (tumor size, nodal status, metastases, ER/PR/HER2 status, histology, grade, date of diagnosis, self- detected vs. abnormal mammogram), treatments received (surgery, radiation, chemotherapy, endocrine therapy), and outcomes (type of recurrence, date of recurrence, date of death if applicable, last follow-up, etc.). Patient home addresses at the time of diagnoses were geocoded and inputted into the Neighborhood Atlas Mapping Tool to obtain 2015 ND values. Spearman’s rank correlation was used to evaluate the association between ND and TNM stage at diagnosis. Cox regression was used to evaluate the effect of ND on cause-specific hazard of recurrence. Significance value of < 0.05 was utilized. Results: The average age at diagnosis was 60 years and most women were post-menopausal (74%). Most patients had ER and/or PR-positive cancer (82% and 72% respectively) and 15% had HER2-positive disease. The average tumor size was 1.3cm (range 0-10). Node positive disease was seen in 104 women (25%) and 20 women (4.8%) had distant metastatic disease at diagnosis. Only 24 (5.7%) patients experienced disease recurrence within five years of diagnosis. The average National ND value was 36 (range of 4 to 100). ND was not significantly associated with stage at breast cancer diagnosis (p = 0.37) or five-year recurrence rates (p = 0.19). Conclusions: The average National ND value was not significantly associated with stage at breast cancer diagnosis or five-year recurrence rates. It is important to consider our study population demographics in the context of these results. The average National ND value was low at 36 which suggests an unequal representation of those living in disadvantaged areas compared to those living in more advantaged areas. Additionally, we were unable to draw strong conclusions on effects of ND on patient outcomes due to low number of recurrences seen. Future studies with larger sample size are likely needed to capture the possible impact of ND on patient outcomes.