RWJF Mockup

Mockup of Content for RWJF scrollie

Author

Ran Li

Published

May 25, 2023

1. Intro

  • Context: Community environment affects health status. Here we quantify the impact of some community level factors on chance of developing severe COVID-19 (severe enough to require a hospitalization); we use two main data sources.
  • Community environment information such as poverty and access to healthcare. Obtained via the Census.
  • Hospitalization records from HealthShare Exchange -a regional health information hub that gathers clinical information from member health care providers within the Philadelphia Region.
  • Cases: all cases of severe COVID-19. severe COVID-19 cases were defined as those that required inpatient care between January and December of 2020.
  • Controls: a random sample of the population in the clinical data repository who had at least one healthcare encounter in 2020 and were not diagnosed with COVID-19.
  • We have data for the five counties around Philadelphia shown below

2. Clinical data is representative of population

  • Before we can do analysis we have to make sure our clinical data is representative of our population.
  • The idea in this slide is to show the geographic distribution of controls (which is a random sample of the population) correlates with population.

3. COVID-19 hospitalization risk varies by community

  • Ratio of cases to control map. The idea is to show geographical distribution of the “risk”.
  • To discuss:
  • cut-offs: filter for zcta with pop > 150; anything below looks strange because of low numbers e.g. very high ratios

4. Vulnerability relates to hospitalization risk

  • SVI map – compare map of SVI with map 2 (ratio cases to controls). This map could then transition to a plot of SVI vs. case to control ratio.

5. Focus on County

  • Possibly focus on Philadelphia or being able to see the different counties in the other maps.

  • Example for Philadelphia we will make interactive for other counties.