Harvard Global Health Institute, Harvard’s Edmond J. Safra Center for Ethics and a number of other researchers and public health experts across the country together launched a new data framework for identifying COVID-19 risk levels for entire states and their counties.
Identified regions fall into one of four risk level categories -- red, orange, yellow or green -- depending how high their increase of COVID-19 cases is each day. Regions’ risk levels don’t necessarily change each day, but rather are based on the daily changes in data.
Regions labeled green are “on track for containment” and are reporting fewer than 1 new COVID-19 case per 100,000 people each day. Regions labeled yellow are experiencing “community spread” and are reporting 1-9 new cases per 100,000 people each day. Regions labeled orange are experiencing an “accelerated spread” and are reporting 10-24 new cases per 100,000 people each day.
Researchers say regions labeled red are at a “tipping point” and are reporting at least 25 new cases per 100,000 people each day.
Hawaii and Vermont are the only two states labeled at a green risk level. Majority of the states, 33, are labeled yellow and 13 are labeled orange. Arizona, Florida and Mississippi are the only three states labeled at a red risk level, as they are seeing significant spikes in COVID-19 cases.
When broken down further, there are far more counties experiencing a red risk level than there are states. A majority of those counties are concentrated in southern states labeled at a red risk level, but some are spread throughout the rest of the country in Washington, Iowa, South Dakota and even in Michigan.
The entire state of Michigan has been labeled yellow, or experiencing “community spread,” by the research initiative. Most Michigan counties are split among yellow and green risk levels, with 45 and 35 counties in each category, respectively.
Oscoda and St. Joseph counties are both experiencing an orange COVID-19 risk level. Oceana County is the only Michigan county labeled red, reporting more than 47 COVID-19 cases per 100,000 people a day, according to the research.
The state of Michigan has also categorized counties by COVID-19 risk level under the MI Safe Start Map, which shows most counties are at a medium or low risk. The state’s data assesses a region’s increase of COVID-19 cases per million people, whereas Harvard’s new research assesses the case increase per 100,000 people.
Researchers that collaborated on the new data framework with Harvard wanted to use a smaller number to more evenly gauge COVID-19 case increases across all counties and cities, regardless of population size.
“It allows you to compare a rural area in upstate New York compared to New York City and have an apples-to-apples comparison for relative impact and relative caseload,” Ellie Graeden of Talus Analytics and the Georgetown University’s Center for Global Health, Science and Security told NPR.
The county-focused data is meant to help residents make informed decisions before traveling to a different part of the state, and maybe reconsider visiting counties at higher risk levels.
The data is also meant to help policymakers as they respond to -- and try to prevent -- COVID-19 outbreaks.
Each of the four risk levels are accompanied by suggested actions to help a region mitigate or suppress the spread of COVID-19. Green regions are encouraged to maintain widespread testing and contact tracing, while yellow and orange regions are encouraged to rigorously test and conduct contract tracing, and maybe even implement stay-at-home orders.
Red regions, or those labeled at the “tipping point,” are strongly advised to implement stay-at-home orders -- which the researchers say is necessary at that stage of spread.
Regardless of their risk level, researchers are encouraging all counties and states to continue taking steps to contain the spread of COVID-19.
“Local leaders need and deserve a unified approach for suppressing COVID-19, with common metrics so that they can begin to anticipate and get ahead of the virus, rather than reacting to uncontrolled community spread”, said Beth Cameron, Vice President for Global Biological Policy and Programs at the Nuclear Threat Initiative and a member of the COVID-Local.org team. “Unless and until there is a whole of government response, with measurable progress communicated similarly and regularly across every state and locality, U.S. leaders will be left to react to the chaos of the virus -- rather than being able to more effectively target interventions to suppress it. "
Researchers say this new framework is intended to create a baseline for COVID-19 risk data that can be used by people and lawmakers across the country, since most public data available differs from state to state.
The group that launched the framework Wednesday is comprised of members from Rockefeller Foundation, CovidActNow, Covid-Local, Talus Analytics, Google Research, Microsoft Research, Apple University and more.
If the maps above look familiar that’s because they were crafted with help from CovidActNow, who previously launched a similar COVID-19 “warning system” that identifies each state’s risk for virus spread. Michigan was labeled as one of only three states “on track to contain the virus” in mid-June, but has since taken a step backward and is now experiencing “controlled disease growth.”