Community Vulnerability Index Provides Insights on Pandemic Preparedness
by Joyce Lee-Ibarra, JLI Consulting LLC
If you’ve done even limited reading about COVID-19 in recent weeks, you’ve likely encountered a short, consistent list of risk factors for individuals developing serious illness from the disease: those ages 65 and older, and anyone with underlying medical conditions. But what do we know about community-level risk for COVID-19? How might we identify which communities are more or less resilient to the current pandemic?
A good place to begin exploring answers to these questions can be found through the COVID-19 Community Vulnerability Index (CCVI).
What is the COVID-19 Community Vulnerability Index?
Developed by data scientists at the Surgo Foundation, the COVID-19 Community Vulnerability Index builds on the Social Vulnerability Index (SVI) developed by the Center for Disease Control & Prevention. Where the SVI uses four categories of socioeconomic and demographic factors to gauge community vulnerability to any type of disaster event, the CCVI adds two additional categories of factors—epidemiologic and health care. Together, these six categories represent 34 total factors for analysis of community vulnerability to the current coronavirus pandemic. Overall and thematic category scores for CCVI are calculated for every state in the US, and data can also be viewed at the county and census tract levels.
What insights is CCVI most useful for providing?
Importantly, CCVI does not identify which states or communities are most likely to see spikes in infection rates or large numbers of COVID-19 patients. Rather, the Index helps policy makers and public health officials understand which communities are better or worse equipped to respond to COVID-19’s impacts, and to plan for resource allocation accordingly.
How are the vulnerability categories designated?
CCVI scores range in value from 0 – 1, with higher scores indicating greater vulnerability. A given geographic unit—for example, a census tract or county—is ranked relative to all similar units across the country, based on the six categories of analysis described above. The score generated can then be used to identify a vulnerability designation. Each designation corresponds to a quintile of all of that geographic unit type in the US; i.e., a county score of 0.00 - 0.20 would correspond to “very low vulnerability” compared to all other US counties, a score of 0.21 - 0.40 would correspond to “low vulnerability,” and so on up through the last category of “very high vulnerability” with scores ranging from 0.81 - 1.
Graphic Source: Surgo Foundation, https://precisionforcovid.org/us
Why does Hawaii as a whole have a higher CCVI score than each of its counties?
CCVI is first computed at the census tract level; county- and state-level CCVIs are then calculated using population-weighted averages of those census tract CCVIs, which are then re-ranked from 0 – 1 as described above. As a result, different vulnerability rankings may arise between, for instance, the counties and the state. As Nick Stewart, Research Scientist at Surgo Foundation explained via email:
The index is essentially a ranking system comparing each geographic unit with all others across the US (i.e. one county versus all other counties). For instance, the CCVI score in Honolulu county (0.535) is its numerical representation of vulnerability compared to ALL other counties in the entire US.
When aggregating the index to a higher geographic level (e.g. from county to state), we first take the mean [average] of the CCVI for all counties in a state and weight that mean by the population. However, we then need to “re-rank” (by taking the percentile rank of the weighted mean) this metric by comparing it to all other states across the US.
The “rank” of the weighted average of all counties in Hawaii is therefore higher than most of the averages from the other 50 states, generating a state CCVI for Hawaii of 0.78. For instance, other (larger) states may have several counties with a CCVI = 0.7 and many more counties with a CCVI of only 0.2. So, the weighted average of these counties will be lower than that of Hawaii.
How might decision makers at all levels use the CCVI?
At a national level, CCVI can be used by policy makers and leadership of national or regional organizations to identify the states and regions less equipped to deal with the effects of COVID-19. Surgo Foundation’s analyses, for example, find that communities in the southeastern US are the most vulnerable to the coronavirus’ impacts, and that they also have the highest proportion of vulnerable counties. This finding can inform decision makers at the federal and state levels to plan accordingly.
Here in Hawaii, the more granular data undergirding CCVI can help local decision makers determine appropriate COVID-19 response strategies. Close analysis of census tract vulnerability, as well as consideration of the scores for each subfactor theme (e.g., socioeconomic status, housing type & transportation, etc.), can further guide local, tailored responses to community need.
Joyce Lee-Ibarra is principal of JLI Consulting LLC and a consultant to the Hawaii Data Collaborative.
