Abstract Details

Title Geographic Proximity to Healthy Food is Not Independently Associated with Ischemic Stroke Risk Factors

Topic Cerebrovascular Disease and Interventional Neurology

Presentation(s) P3 - Poster Session 3 (5:30 PM-6:30 PM)

Poster/Presentation
Number 3-006

Objective

To examine the association of food desert residence (FDR), low income (LI), and low vehicle access areas (LVAA) to cardiovascular (CV) risk factors among ischemic stroke (IS) patients.

Background Environmental/socioeconomic barriers may prevent people from obtaining diets rich in healthy foods that can reduce CV risk factors. The association between FDR and burden of CV risk factors has been previously attributed to low income rather than food access but has not been characterized well among IS patients.

Design/Methods We retrospectively collected data on IS patients between 10/1/2014 to 9/30/2015. Demographics, residence, and vascular risk factors were extracted. FDR, LI, and LVAA status were determined using The USDA Food Access Research Atlas. Spearman correlation and adjusted logistic regression were used to examine the relationships between CV risk factors and FDR, LI, and LVAA.

Results We identified 353 patients (mean age 69, 45% black, 56% female, 67% Medicare, 56% FDR, 52% LI, and 40% LVAA). Black race correlated with FDR (r=0.16; p-value=0.003), LI (r=0.29; p-value <0.0001), and LVAA (r=0.51 p-value <0.0001). FDR was not associated with having hypertension (OR=0.65, CI=0.32-1.3, p-value=0.23), dyslipidemia (OR=0.95, CI=0.59-1.5, p-value 0.82), CAD (OR=1.1, CI=0.64-1.7, p-value=0.82), diabetes (OR=0.76, CI=0.45-1.2, p-value=0.22) or obesity (OR=1.1, CI=0.7-1.8, p-value 0.63). LI area residency was found to be associated with hypertension (OR=3.2, CI=1.4-7.4, p-value=0.007). FDR (p-value for interaction =0.05) and LVAA (p-value for interaction =0.04) were found to modify the association between LI and diabetes.

Conclusions Among ischemic stroke patients, food access does not have an association with hypertension, diabetes, dyslipidemia, CAD, or obesity as an independent variable. However, FDR and LVAA modify the association between LI and diabetes. Further studies into FDR, LI, and LVAA should be conducted to examine this interaction in larger data sets.

Authors/Disclosures
Alan D. Howard, Jr. PRESENTER	No disclosure on file
Chigozirim Izeogu, MD (McGovern Medical School/UT Health Houston)	Dr. Izeogu has nothing to disclose.
	No disclosure on file
Michael J. Lyerly, MD, FAAN (University of Alabama At Birmingham)	Dr. Lyerly has nothing to disclose.

��ɫ��

��ɫ��