The Impact of living in a food desert on endometrial cancer outcomes

Abstract

Objectives: Living in a food desert, or an area deficient of healthy whole foods, is associated with several poor health outcomes including obesity, cardiovascular disease, and diabetes. An association with gynecologic cancer outcomes has not been investigated. The objective of this study was to compare demographic and clinicopathologic risk factors for endometrial cancer patients living in food deserts to those who do not. We furthermore compare endometrial cancer recurrence rates and overall survival for those residing in areas with different access to healthy foods. Methods: We conducted a retrospective cohort study using the Maryland Cancer Registry with women diagnosed with endometrial adenocarcinoma from January 1, 2016 to December 31, 2017 within the counties of Baltimore City and Baltimore County. Using each patients’ zip code, we created the food desert score (FDS), a metric that reflects the density of convenient-corner stores to whole food grocery stores. Food desert scores were stratified into “quartiles” with Quartile 1 (Q1) indicating the least impoverished healthy food areas and Quartile 4 (Q4) indicating areas with increased severity of the food desert. Demographic and clinicopathologic risk factors were compared with Student's t-test, Chi-square, and one-way ANOVA. Multivariate analysis and Cox proportional hazards were used to evaluate recurrence risk and overall survival. Results: A total of 290 women met inclusion criteria. The mean food desert score was 10.1 (range 1-69). More women with endometrial adenocarcinoma were found to live in the highest FDS quartiles, with Q1 and Q2 at 22.1% and 19.6%, respectively, and Q3 and Q4 at 34.1% and 25.5%, respectively (p<0.005). Women living in food deserts (Q3 and Q4) were more likely to be Black and not married, [OR 3.95 (95% CI 1.26-9.75); OR 3.27 (95% CI 1.09-9.10), respectively]. Those living in the highest FDS quartile (Q4) were also more likely to have positive lymph nodes [OR 7.91 (95% CI 1.94-66.25)] than those living in Q1. There was no significant difference in the likelihood of having positive lymphovascular invasion [OR 2.59 (95% CI 0.92-7.29)] or high grade disease [OR 0.93 (95% CI 0.42-2.08)]. There were only 6 recurrences (2.6%) during the follow-up period of 33 months. Three recurrences were in Q1 and Q2, and 3 were in Q3 and Q4. Overall survival was 28.5, 20.1, 24.6, and 16.5 months for Q1, Q2, Q3, and Q4, respectively. After controlling for confounders, those in Q4 had a significantly lower survival compared to Q1, HR 0.38 (95% CI 0.16-0.89). Conclusions: Among endometrial cancer patients in Baltimore City and Baltimore County, those residing in a food desert were more likely to be Black and not married and were also more likely to be diagnosed with advanced disease (positive lymph nodes). Furthermore, women living in areas with the least availability of healthy food options had a significantly shorter survival than those with healthy food options. To our knowledge, this is the first study demonstrating an association between patient zip code, as it relates to the availability of healthy food options and endometrial cancer risk factors and outcomes. Living in a food desert, or an area deficient of healthy whole foods, is associated with several poor health outcomes including obesity, cardiovascular disease, and diabetes. An association with gynecologic cancer outcomes has not been investigated. The objective of this study was to compare demographic and clinicopathologic risk factors for endometrial cancer patients living in food deserts to those who do not. We furthermore compare endometrial cancer recurrence rates and overall survival for those residing in areas with different access to healthy foods. We conducted a retrospective cohort study using the Maryland Cancer Registry with women diagnosed with endometrial adenocarcinoma from January 1, 2016 to December 31, 2017 within the counties of Baltimore City and Baltimore County. Using each patients’ zip code, we created the food desert score (FDS), a metric that reflects the density of convenient-corner stores to whole food grocery stores. Food desert scores were stratified into “quartiles” with Quartile 1 (Q1) indicating the least impoverished healthy food areas and Quartile 4 (Q4) indicating areas with increased severity of the food desert. Demographic and clinicopathologic risk factors were compared with Student's t-test, Chi-square, and one-way ANOVA. Multivariate analysis and Cox proportional hazards were used to evaluate recurrence risk and overall survival. A total of 290 women met inclusion criteria. The mean food desert score was 10.1 (range 1-69). More women with endometrial adenocarcinoma were found to live in the highest FDS quartiles, with Q1 and Q2 at 22.1% and 19.6%, respectively, and Q3 and Q4 at 34.1% and 25.5%, respectively (p<0.005). Women living in food deserts (Q3 and Q4) were more likely to be Black and not married, [OR 3.95 (95% CI 1.26-9.75); OR 3.27 (95% CI 1.09-9.10), respectively]. Those living in the highest FDS quartile (Q4) were also more likely to have positive lymph nodes [OR 7.91 (95% CI 1.94-66.25)] than those living in Q1. There was no significant difference in the likelihood of having positive lymphovascular invasion [OR 2.59 (95% CI 0.92-7.29)] or high grade disease [OR 0.93 (95% CI 0.42-2.08)]. There were only 6 recurrences (2.6%) during the follow-up period of 33 months. Three recurrences were in Q1 and Q2, and 3 were in Q3 and Q4. Overall survival was 28.5, 20.1, 24.6, and 16.5 months for Q1, Q2, Q3, and Q4, respectively. After controlling for confounders, those in Q4 had a significantly lower survival compared to Q1, HR 0.38 (95% CI 0.16-0.89). Among endometrial cancer patients in Baltimore City and Baltimore County, those residing in a food desert were more likely to be Black and not married and were also more likely to be diagnosed with advanced disease (positive lymph nodes). Furthermore, women living in areas with the least availability of healthy food options had a significantly shorter survival than those with healthy food options. To our knowledge, this is the first study demonstrating an association between patient zip code, as it relates to the availability of healthy food options and endometrial cancer risk factors and outcomes.