Area Deprivation Index (ADI) and Median Income Associations with Clinical Outcomes of Endovascular Therapy for ACLVO stroke
Murat Sari1, Vanessa Dwairi2, Rishabh Jain1, Marcelo Rocha1
1Neurology, UPMC, 2UPMC
Objective:
We aimed to see whether or not an association is present between higher ADI scores and lower median income with worse stroke outcomes.
Background:
Poor socioeconomic status has been linked to worse cardiovascular disease outcomes. ADI incorporates housing, education, income, and employment status at the census block group level. Higher scores of ADI represent poorer conditions. We also used zip code based median income as an indicator of socioeconomic status.
Design/Methods:
This single center retrospective study of patients admitted between March 2020 and March 2022 looked at demographics, medical comorbidities (including COVID-19), socioeconomics (ADI v3.2 and median income). Last known well (LKW) and hospital arrival times are used with outcomes measures of TICI grade, hemorrhagic conversion, discharge NIHSS, discharge and 3 months mRS. Spearman correlations for univariate comparisons and regression analysis for independent association between socioeconomic predictors and clinical outcomes are used.
Results:
A total of 606 were included in the study with mean age 71 (sd 15), 53% female. Mean median income was $50,080 (sd $10,480) and mean ADI score was 65 (sd 23). Median income is found independently correlated with TICI score, NIHSS on discharge, mRS at 3 months. (p<0.05)
NIHSS on discharge and TICI grade had significant coefficients for median income (in $10K USD) (−0.637, 95% CI −1.126 - −0.148; p<0.05) and (0.763, 95% CI 0.098 - 1.428; p<0.05) respectively. LKW-to-Arrival time had significant regression coefficient for median income (−0.625, 95% CI −1.239 - −0.011; p<0.05), but there could be many omitted variables for which data wasn’t available.
Conclusions:
Median income was independently associated with stroke patients’ time of treatment and functional outcomes. ADI data did not show a clear relation with outcome variables despite it is a good indicator of socioeconomic status. With less than 10% of total population, COVID-19 patients were not included in final analysis.