ABSTRACT
BACKGROUND: The spatial mismatch hypothesis (SMH) postulates that the discrepancy between where Black workers live and where they have access to jobs can lead to higher unemployment and worse economic outcomes. This gap exists due to structural factors such as redlining and hiring discrimination. As one of the most salient structural factors preventing economic mobility, the SMH provides a novel lens for examining racial disparities during the COVID19 pandemic. This study explores whether there is an association between measures of spatial mismatch and COVID-19 positivity rates by neighborhood racial composition. METHODS: We conducted a retrospective cohort study of patients tested for COVID-19 at an academic medical center and five community-based testing sites in Chicago (March 12-June 25, 2020). Analyses were limited to patients living in Black or White majority neighborhoods, and those with missing data were removed. Each patient's residential address was geocoded to the census block group level and paired with neighborhood race/ethnicity data (majority Black or White) from the 2018 American Community Survey. The dependent variable was COVID-19 positivity, defined by a PCR-positive sample and extracted from the electronic health record. The primary independent variables were neighborhood racial composition and three different measures of SMH at the block group level-commute time, public transportation usage, and neighborhood low-wage job rate. Mixed effects logistic regression models were used to assess COVID-19 positivity as an independent function of block group racial composition and SMH variables, adjusting for patient sociodemographic factors and insurance type. RESULTS: Among 21,285 patients tested for COVID-19, data on 14,488 patients from 1,752 block groups were analyzed. Patients were predominantly non-Hispanic Black (69.2%), female (60.9%), and ages 50-64 (23.8%). There were significant differences in the patterns of neighborhood racial composition and SMH measures. For example, <10% of patients living in a White majority neighborhood (n=347) also lived in a neighborhood with high travel time (>75th percentile) to work. Patients living in a Black majority neighborhood had 2.06 times higher adjusted odds (95% CI, 1.76-2.42) of COVID-19 positivity relative to those in a White majority neighborhood. High travel time (AOR=1.35;95% CI, 1.12-1.64), high public transportation usage (AOR=1.24, 95% CI, 1.01-1.51), and low neighborhood low-wage job rate (AOR=1.32;95% CI, 1.05-1.65) were associated with higher COVID-19 positivity. In a cumulative model, spatial mismatch accounted for 12.6% of the disparity in COVID positivity. CONCLUSIONS: The SMH accounted for a small but significant proportion of the racial disparity in COVID-19 positivity among patients at an academic medical center in Chicago. The impact of spatial mismatch should be explored for other health outcomes, particularly chronic disease, to quantify its contribution to health disparities and better target interventions.
ABSTRACT
STATEMENT OF PROBLEM/QUESTION: How can preclinical medical students be leveraged to address racial and geographic disparities in COVID19 vaccination rates? DESCRIPTION OF PROGRAM/INTERVENTION: As vaccine rollout began in Chicago, communities most affected by the pandemic had the lowest vaccination rates. At our urban academic medical center, eligible patients received vaccine invitations via the patient portal or text message;however, this approach did not effectively reach many elderly patients who were not technologically connected or who had circumstance-specific questions. Due to clinical demands, staff were unable to reach out to individual patients. Preclinical medical students, with more flexible schedules, volunteered to address this gap in access. Targeting patients who lived in high-risk ZIP Codes (per the The COVID-19 Community Vulnerability Index), we aimed to leverage preclinical students to expand the capacity of our vaccine outreach and tackle vaccine hesitancy. MEASURES OF SUCCESS: 1. How many patients were contacted to inform them of their eligibility? 2. How many patients were scheduled for a vaccination? FINDINGS TO DATE: Overall, 34 students contacted 820 patients. Most patients were Black or African American (91.0%). Of the patients that were reached (n=489), 84 (17.2%) were scheduled for vaccine appointments. Additionally, 79 (16.2%) of the patients that were reached were not immediately scheduled but agreed to vaccination, 52 (10.6%) said they were considering vaccination, 193 (39.6%) reached patients had already scheduled or received vaccination elsewhere, and 89 (18.2%) declined the vaccine after some discussion. KEY LESSONS FOR DISSEMINATION: We showed that integrating preclinical medical students into the health system can extend existing outreach efforts and thus is a model that is generalizable across many health-related issues. Beyond the tangible impacts of connecting patients with vaccine information and appointments, we learned several lessons. 1) Trainees' outreach increased healthcare accessibility for many patients. Many patients did not have a primary care physician and/or had previously only been seen in the Emergency Department, which created an opportunity to connect these patients to the healthcare system. 2) Many patients had difficulty independently making an appointment or held misinformed beliefs. As such, direct outreach gave us the opportunity to assist with patient-specific issues. 3) This intervention also benefited clinicians, who have limited time to proactively reach out to thousands of patients. 4) Further, our initiative benefited medical education: preclinical medical students gained experience and confidence speaking to patients, delivering patient education, and using the electronic medical record. Models like ours can address gaps in care beyond COVID-19, this model can be applied effectively to address inequities in healthcare access while leveraging the time, motivation, and skills of preclinical trainees.