What’s the value of a strong, coordinated response when faced with an unprecedented global pandemic? Turns out, in the 10-county area, there’s a strong correlation between a decrease in government response and an increase in Covid-19 cases.

Inspired by the University of Oxford’s Coronavirus Government Response Tracker (OxCGRT), the Atlanta Regional Commission published the Local Government Response Index (LGRI) dashboard. By visualizing the relationship between local policy decisions, residents’ mobility, and disease outcomes, it provides a systematic means to measure the effects of local governments’ response to COVID-19 across the 10-county region from the beginning of the pandemic through July 30. The goal is to explore whether increasing the stringency of local government responses interacts with a jurisdiction’s infection rate and/or mobility changes, as well as to identify correlates of  more or less stringent responses. 

The dashboard shows COVID-19 daily new cases per 100,000 population, the LGRI, and the staying at home index at both the 10-county and county level, with data charted on a seven-day rolling average to minimize data noise. The LGRI and the staying at home index are scaled from 0 to 1. The LGRI score is based on a composite of seven policy indicators: school closures, cancellation/prohibition of public gatherings, workplace and business closures, shelter in place orders, public transportation and travel restrictions, public facility operations, and public messaging. 

The infographic below provides highlights from the dashboard.

* Please click here or on the image to access to the full dashboard. Hover over the dashboard to get daily data for the region and each county.

Before the statewide shelter in place order:

Prior to the shelter-in-place order, we can see the LGRI and Staying at Home index and both increasing at a steep upward trend, while cases per 100,000 are rising slowly and steadily.  

• Certain local government responses to COVID-19 began as early as March 1st, with most county public schools closing by March 16th.  
• From March 12th to April 2nd, the stay at home index dramatically increases indicating a significant decrease in mobility across the region. This could be due to a reduction in students/parents commuting to school and/or a move to working remotely for those who could. The Atlanta Regional Commission, for instance, moved its workforce to work remotely by March 12th, a shift that many downtown companies made at that time.
• As school shutdowns continued throughout our counties, local governments began passing ordinances dealing with shelter in place orders, public facility closures, business closures (restaurants, bars, etc.), and public messaging.
• By March 18th , Georgia governor Kemp ordered all public elementary, secondary, and post-secondary schools in Georgia to close. 

After the statewide shelter in place order:

By April 2nd Georgia’s statewide shelter-in-place order is enacted, superseding all local government orders.

• Depending on previous county orders, the April 2nd state shelter in place order either meets the county’s LGRI score or increases it by a reasonable amount (<=) 0.1. These state orders affect five of our seven indicators impacting policy stringency uniformly across the 10 counties.  
• April 2nd to April 24th saw its peak Staying at Home index values, indicating that mobility across the region had decreased dramatically to its lowest mobility rates. During this period cases continue to increase overall but at a slower pace and with a dip in cases occurring in mid-April.

After Georgia reopens:

After the state reopened, policy stringency decreased significantly and residents began staying home less. 

• By April 23rd the Governor released a plan to reopen Georgia and reduced shelter-in-place and travel restrictions to just vulnerable populations.
• While COVID-19 cases hit a peak during the shelter-in-place order, the significant and steady decrease in new cases beginning in early May through Memorial day indicates the early peak may be related to the 14-day lag between infection, experiencing symptoms, and testing positive. 
• The significant decrease in the LGRI score and the steady decrease in the Stay at Home index occurring between May 7th until late June coincides with Memorial Day and the Fourth of July. In the weeks following, we can see a region-wide increase in cases, with a sharp increase shortly following both holidays.  

When it comes to the impact of government response, staying at home and an increase in cases, the 10-county area offers a twist on our usual method of thinking through the impact of stringency: If you run the calculations for the entire time period, there is no correlation between the increase in cases, LGRI and the Staying at Home index. If, however, you narrow your range of view to June 2 through July 30, when we saw our massive spike in new cases, there is a strong negative correlation between the LGRI score and an increase in cases — basically the absence of policy has a strong statistical relationship with an increase in cases.

This may seem counterintuitive, but it’s in line with research examining differences between states’ responses and the spread of COVID-19. In a working paper, researchers from Oxford found that states like (and including) Georgia, where policymakers relaxed their response quickly and never tightened it back up, experienced an upward trend. In other states, such as Delaware, Connecticut, Minnesota, New Hampshire, Vermont, and West Virginia, they found that stringency enacted after a bump in cases actually did impact how many new cases they experienced.

Another way to think about it is that the absence of a government response when faced with an increase in cases gives COVID-19 the foothold it needs to begin its exponential increase.

– Developed by: Allie Orrego, GIS & data analyst for the Metropolitan North Georgia Water Planning District; Ryan Schlom, planner and researcher for the Atlanta Regional Commission; and Hae Seung Sung, senior data analyst at the Atlanta Regional Commission