Tagcovid

We visualized COVID’s spread across every U.S. state and county. Here’s what we discovered:

Just by looking at numbers and daily updates, it can be hard to decipher a clear narrative of how COVID has affected the U.S. and its diverse regions over time. 

To give you a better understanding, our team at Covid Act Now took all of the data we’ve been collecting and created a time-lapse of COVID’s spread across 3,000+ U.S. counties since March. Watch what happens:

Notice anything interesting? Here are 5 key patterns that our team saw:

1. More movement = higher COVID incidence.

Mobility data, or data relating to transportation, show a striking similarity between the percentage of people staying home and cases per 100,000 people (also known as incidence) during the month of August (source: U.S. Bureau of Transportation Statistics).

Especially in the Southeast (which includes states like Alabama, Arkansas, Florida, Georgia, Kentucky, Louisiana, and Mississippi), only 5 percent to 10 percent of people are consistently staying home, and this is reflected in their red case density values:

2.Masks and other non-pharmaceutical interventions (NPIs) play a powerful role in controlling the spread of the virus.

As America’s viral epicenter, New York pioneered NPIs for COVID prevention. Starting on April 17, Gov. Andrew Cuomo enacted a mask requirement, and this, in addition to other NPIs in New York State, contributed to New York’s eventual containment of COVID (source: New York State). Some of New York’s NPIs include:

March 15: Schools close for the first time.March 28: New York State’s 2020 democratic primary is postponed to limit in-person social contact.April 16: Stay-at-home order and school closures are extended through May 15.April 22: Cuomo announces plans to begin a localized contact tracing program throughout New York State.

From this animation of New York, we can see the success of these policies in limiting the virus from March to June.

3. NPIs not only reduce the spread of COVID within states but also between them.

Throughout the animation, it is clear that COVID spreads rapidly within states, often going from a few red counties to an entire red state within weeks, not months.

The border between Arizona and New Mexico, however, is an example of how COVID oftentimes meets a metaphorical wall at state borders. With the Arizona side of the border solidly red in late July, New Mexico was able to maintain a handle on the virus.

The difference in COVID cases in the two states can be attributed to differing statewide policies:

Masks:New Mexico institutes a statewide mask requirement on May 16, with harsh monetary penalties for mask infractions (source: State of New Mexico).In contrast, Arizona, despite having some of the highest case density in the country, still only “recommends” mask wearing. In fact, until June 17,  Gov. Doug Ducey does not allow local governments to introduce mask mandates (source: Office of the Governor of Arizona).Quick response to outbreaks:It takes until July 9, after Arizona’s cases reach record highs and hospitals reach capacity, for Gov.Ducey to reduce restaurant dine-in services to 50 percent capacity.In comparison, indoor dining is not allowed in New Mexico until June 1, and Gov. Michelle Lujan Grisham quickly passes an executive order in July as Arizona’s cases spike, outlawing indoor dining, closing state parks to out-of-state visitors, and requiring masks while exercising.

The disparity in cases is well reflected in July, when Arizona (left) is almost entirely red, while New Mexico (right) has much more yellow and green:

4. Relaxing NPIs too early leads to flare-ups of COVID.

Let’s look at Georgia as an example here. Although Georgia initially implemented NPIs, such as stay-at-home orders and school closures, several premature reopening policies may have led to further outbreaks in the state. Let’s look at the timeline:

April 24-27: Gov. Brian Kemp allows the reopening of restaurants (including indoor dining), movie theaters, gyms, and more (Georgia ACLU).August 15: Gov. Kemp does not allow local governments (such as Atlanta) to create mask mandates until August 15. Up until this point, there are no mask mandates on either the state or local level. However, only cities and counties that have 19 or more cases per 100,000 people may enact mask requirements for public property. For reference, at 19 or more cases per 100,000, our animation displays dark orange. August 20: University of Georgia and Georgia Tech reopen for in-person classes.

These failures to maintain effective NPIs are reflected in the difference in Georgia between April (left) and August (right) where you can see the number of counties in the red go from approximately 25 percent to 90 percent:

5. When critical safety measures are taken, large, outdoor gatherings may not result in a spike in cases.

While many were concerned that the recent protests and gatherings for social justice would lead to a spike in COVID cases, our map shows that this was in fact not the case. The New York Times produced this map on June 6, when protesting peaked, showing where protests happened geographically:

Via the New York Times, “Black Lives Matter May Be the Largest Movement in U.S. History”

Regions that saw protests include California’s Bay Area and Los Angeles County, New York and Washington DC, and Minnesota/Wisconsin.

If the protests had caused outbreaks, the animation would have displayed meaningful color changes in regions that hosted large protests and gatherings. The lack of outbreaks demonstrates that if we take proper precautions, large gatherings that take place outdoors may not be as dangerous as we once thought.

Symptoms of COVID usually show up two to 14 days after infection (source: CDC). If the protests caused a spike in cases, we would have detected the increase in about two weeks. However, in all of the regions listed above, there was no spike after June 6. See screenshots from June 5 and two weeks later below:

By comparing state and local policies (including mask mandates, school closures, restrictions on indoor dining and other NPIs) to the geographic spread of COVID, a clear picture emerges of how COVID can be contained when society takes action. What’s more, our video depicts visually how states that lift stay-at-home mandates and other restrictions prematurely experience rapid flare-ups of COVID. In contrast, states that keep NPIs in place are even able to stop COVID from crossing their state border.

Check out our  Youtube Channel for our educational videos.

To learn more about Covid Act Now, visit our about page. For more information please reach out on our contact page or by email at info@covidactnow.org.

Sign up to our alerts to stay up-to-date on the COVID risk level in your area.

CAN Compare: See COVID’s Spread Across the U.S.

By: Lexie Kaplan

CAN Compare is our newest feature to give you a quick way to compare states and counties. Specifically, you can: 

Compare states by overall risk level and by each key indicator, e.g. incidence, infection growth rate, test positivity rate, ICU headroom, and tracers hiredCompare counties within a state or nationallyCompare a county to its neighboring countiesCompare just counties within metro areas or just counties outside of metro areasView which counties have significant college student populations

Let’s dive into a few of the more granular features within CAN Compare:

Metro vs. non-metro counties

With CAN Compare, you can look specifically at counties in metropolitan (metro) or non-metropolitan (non-metro) areas. 

So what’s the difference between metro and non-metro counties? We follow U.S. Census definition of a Metro Statistical Area (MSA) as one or more “urbanized areas” of 50,000 persons or more, as well as outlying areas that have strong economic ties to these central hubs. Outlying areas are deemed to be part of a metro area if 25 percent or more of their employed workforce commute to the central county, or if 25 percent or more of the outlying county’s employed labor force is made up of commuters from the central city.In CAN Compare, we define “metro” counties as those belonging to MSAs and “non-metro” counties as those not within an MSA. We think this filter allows for more meaningful comparisons, since metro counties with densely populated urban cores tend to experience COVID outbreaks differently than non-metro counties. 

College counties

In September, as our team was using CAN Compare, we noticed something in common among eight of the metro counties with the highest case incidence: On September 2, six were home to a large college or university that had recently reopened. Of those six, all had seen massive spikes in the past three weeks, coinciding with when many students were arriving on campus. So we decided to display tags for counties that have high college student populations (over 5 percent of the local population). The tags provide an extra layer of visibility to that trend. 

If you are a university administrator, you may want to know if there is a direct correlation between universities opening and local outbreaks so that you can implement more stringent policies for students and faculty living on campus. Likewise, if you are a college student or parent, you may want to know about this correlation so that you can make an informed decision to withdraw from in-person class and seek online alternatives.We’ve thus made it easy for you to see which counties are home to colleges. We hope that this helps you better understand the relationship between COVID spikes and college towns so that you can make more informed decisions for both yourself and your community. 

Nearby counties

In addition to understanding how a given county is performing in comparison to all the counties in the state, we’ve made it easy to rank and compare counties that are adjacent. This helps you quickly discover what areas around you are more or less impacted by COVID outbreaks. 

In September, we tested this feature for ourselves. When we looked at Bergen County, New Jersey, a county just outside of New York City, we used the “Nearby” filter to better understand how Bergen County was performing in relation to New York City and the other nearby suburbs that are likely to have similar conditions.

We found that of the adjacent counties, on September 17 Bergen County had the 11th highest case incidence, performing better than the other suburban counties, Westchester and Passaic, and city counties like Bronx County. 

Try out CAN Compare 

Our hope is that CAN Compare will help decision makers and the public contextualize the COVID risk scores and key indicators. For instance, local leaders may want to understand how their county’s current risk score compares to other similar counties in the state in order to enact appropriate policies. Similarly, residents can better understand how their state or county compares to others on key COVID indicators to know what behaviors or policies might be appropriate. 

You can try out CAN Compare here.

And if you want to get really deep into the data, we recommend you try CAN Analytics, a feature that lets you analyze raw COVID data. 

Introducing a Critical COVID Metric: Daily New Cases Per 100K Population

Today, Covid Act Now is adding an important fifth metric to our COVID warning system: “daily new cases per 100K population” (also referred to by epidemiologists as “incidence”). The addition of this metric rounds out our warning system by incorporating a measure of how much COVID there is in each community today. Our previous metrics focused on direction of change and preparedness, but incidence corresponds to a person’s actual chances of being infected and suggests how many people will likely be infected in the near future. Learn more here.

Nine state risk scores have changed.

This is what the risk level map of the United States looked like on July 21 with the original four metrics:

This is what it looked like on July 22, with the case incidence metric added:

350 county risk scores have changed. 1800 new counties now have risk scores.

Adding the incidence metric has changed the scores of 350 counties. It has also allowed us to expand our coverage to more counties. Previously, many counties did not have enough data for us to calculate a risk score. CAN has always prioritized providing actionable data, and, since a lot of COVID decision-making by policy makers and residents will happen at the local—rather than state or federal—level, we believe it is critical to provide a county-level view of COVID. 

We will now grade every county with a green case incidence score (less than one new case per day per 100K people) as green overall. If case incidence is not green, our normal grading system applies, whereby a state’s overall color reflects the highest risk color for any one of its metrics. Counties that have not reported how many cases they have will show up as grey. 

Here is what the Covid Act Now map of counties looked like before: 

Here is what it looks like now: 

What does this change mean for my community?

We understand that taking into account Daily New Cases per 100K Population has increased the risk score for many states and counties. This change may be disheartening, but we believe it is important for our COVID risk score to reflect risk as accurately as possible and adding this metric improves our ability to do so. 

In addition to helping our overall risk scoring be comprehensive, this new metric also has an important intuitive interpretation. As incidence increases, so does the risk that you’ll run into an infected individual on your trip to the grocery store or at a barbeque.

We also made a minor change to our contact tracing metric. Read more here.

Check out our Youtube Channel for our educational videos.

To learn more about Covid Act Now, visit our about page. For more information please reach out on our contact page or by email at info@covidactnow.org.

Sign up to our alerts to stay up-to-date on the COVID risk level in your area.

Calculating better Infection Growth Rates (Rt) for more communities

Today we are making a change to the way we calculate Rt to better serve lower population regions and regions with lower case counts as well as to improve the timeliness of our Rt metric. We want to be more timely in letting people know when COVID is growing or shrinking in their communities, and this hopefully helps people understand how policies and actions are able to achieve different outcomes.

Our Newest Metric: Contact Tracing

Today, Covid Act Now is excited to announce a fourth metric added to our COVID warning system: contact tracing. We will layout how we calculate whether states have sufficient contact tracing capacity, and why we think it is an important metric to assess reopening.

Why Does Contact Tracing Matter?

When people contract the virus, they do not show symptoms right away. Even as states begin to reopen, people will need to quarantine themselves if they have been silently exposed to someone with COVID.

How will they know? That’s where contact tracing comes in.

Because of this problem, it is critical that enough tracing capacity exists to rapidly trace the contacts of individuals who test positive for COVID. Those contacts can be tested, quarantined (if necessary), and asked about whom else they have come into contact with. Because exposed individuals begin infecting others, it is critical that this process be completed in less than 48 hours. If this routine of testing and tracing is done quickly and completely, it can contain COVID, as we have seen in South Korea and Taiwan, and without the need for costly lockdowns.

The White House Coronavirus Task Force’s Guidelines say that contact tracing is a “core responsibility” of states in order to be prepared to reopen. As of May 21, 27 states (CA, CT, DE, FL, HA, IL, IN, KS, KY, ME, MD, MI, MO, MT, NE, NV, NM, NY, NC, ND, OH, PA, RI, SD, WA, WV, and WI) call for contact tracing in their reopening plans. The American Enterprise Institute’s roadmap to reopening says states must massively scale up contact tracing and isolation/quarantine of traced contacts.

How Should We Measure Contact Tracing?

So how do we calculate contact tracing capacity? Experts recommend tracing contacts of someone who tests positive for COVID within 24 hours, to contain the potential of transmission. Based on conversations with practitioners and public health experts, our metric assumes that tracing all contacts for each new positive COVID case requires an average of five full-time contact tracers. 

Therefore, our contact tracing metric measures the percentage of new cases for which all contacts can be traced within 48 hours relative to available contact tracing staff in the state (assuming 1:5 new-positive-COVID-case:contact-tracing-staff ratio).

We use green if greater than 90% of the contacts can be traced within 48 hours, yellow if between 20% and 90% of the contacts can be traced within 48 hours, orange if between 7% and 20% of the contacts can be traced within 48 hours, and red if fewer than 7% of the contacts can be traced within 48 hours.

Here is an example from Wyoming:

As of June 13, Wyoming has an average of 12 new cases per day. If Wyoming needs 5 contact tracers per case, that would be 60 contact tracers necessary to trace all cases in 48 hours. Since Wyoming has 50 contact tracers, that is enough to trace 82% of cases.

What Should The Goals Be?

How did we choose our targets? Research estimates that the infection rate can be driven below 1.0 if 70-90% of cases are identified and 70-90% of those contacts are traced and isolated within 48 hours or less. Therefore, we chose 90% as the cut-off between green and yellow.

The boundaries between yellow, orange, and red are trickier. When less than 90% of positive cases have their contacts traced within 48 hours, contact tracing will likely be insufficient to contain COVID. We use 90% as the boundary between green and yellow. In the absence of expert consensus, we have set inclusive lower thresholds for yellow and orange. We peg the cut-off between yellow and orange at 20% — the number required for there to be one contact tracer per active case per 48 hours — and the cut-off between orange and red at 7%. Every state currently coded red is either currently experiencing a new outbreak or effectively has no tracing capacity.

A state can become green either by increasing the number of contact tracers, or by decreasing the number of new daily COVID infections. We hope that this new metric will help states factor contact tracing capacity into their reopening decisions.

Check out our Youtube Channel for our educational videos.

To learn more about Covid Act Now, visit our about page. For more information please reach out on our contact page or by email at info@covidactnow.org.

A Dashboard to Help America Reopen Safely

Note: This blog post is a higher-level description of Covid Act Now‘s four COVID metrics. For a more technical, “kitchen sink” description of assumptions and methodology, please see our references and assumptions document, along with our data sources presentation.

Fighting COVID is hard, but understanding where we stand against COVID just got easier.

A policy consensus is emerging how to reopen safely and several metrics are key to understand when and how.

The four key metrics are:

COVID case growth: Are the number of cases increasing, stabilizing, or decreasing?

Testing capacity: Are we testing broadly enough to track disease spread?

ICU safety margin: Is there enough ICU hospital capacity to absorb a possible wave of new COVID hospitalizations?

Contact tracing: Are we finding and isolating new cases before COVID spreads?

As of today, we track these four metrics for all 50 states and 93 of the most populous counties in the United States. An additional 1,727 counties have at least one metric, and more counties will be added to the tally in the coming days. 

Why We Built The Dashboard

The dashboard is based on feedback from federal, state, and local decision-makers. The feedback was clear: without actionable data, it is incredibly challenging to know whether, when, and how to reopen.

Our goal building the dashboard was to solve this problem by providing easy-to-understand, actionable, real-time metrics relevant to COVID.

Data-driven decisions help us better manage our response to COVID. The better we manage our response to COVID, the less our economy will be harmed and the more lives we can help save.

Case Growth

In any given state or county, cases of COVID can be increasing, stabilizing, stable, or decreasing. We color the number red if the cases are increasing (R(t) > 1.4), orange if the cases are stabilizing (R(t) 1.1 – 1.4), yellow if the cases are stable (R(t) 0.9-1.1), and green if cases are decreasing (R(t) < 0.9). In order for a state to reopen safely, the number of COVID cases should be clearly decreasing. In other words, R(t), the important epidemiological metric that we described previously, should be less than .90 For example, above is a chart of the infection growth rate in Montana (as of May 12). The R(t) of 0.96 means that each person with COVID is, on average, infecting 0.96 other people. Therefore, the total number of cases is (at least in the present moment) decreasing. Because of how we weight our data (Gaussian smoothing), and because of potential reporting delays and errors in the incoming case data, we need 10 preceding days of data before we can calculate a final R(t) value. Therefore, we notate preliminary R(t) values — values for which we don’t yet have 10 preceding days of data — with the dotted line. To calculate R(t), we use new positive cases and COVID death data sourced from The New York Times. An emerging policy consensus suggests that states should wait to reopen until cases have steadily decreased (an R(t) less than 1.0) for at least two consecutive weeks. Johns Hopkins University and American Enterprise Institute both include two weeks of decreasing cases as a metric for reopening in their policy guidelines, as do (as of 12 May) fourteen states (CN, DE, IN, NV, NY, RI, SD, WI, VA, KY, KS, HI, DC, and CO.). The Center for American Progress also calls for a sustained decline in positive cases in its reopening recommendations, and Duke University’s reopening plan cites AEI’s two-week figure. Testing Capacity In order to consider opening up, states and counties should be testing a sufficient percentage of the population to ensure that there are no undetected infection clusters that could lead to a second wave of disease. How do you know if you’re testing enough? One way to measure testing capacity is to look at the percentage of positive tests. As Michael Ryan, executive director of the WHO Health Emergencies Program, puts it, “If 80-90% of the people test positive, you are probably missing a lot of cases.” The White House Coronavirus Task Force’s reopening guidelines also point to a reduction of positive tests as a percentage of total tests as a key metric for reopening. So, low is good. How low is low enough? We have created four categories for the percentage of positive tests: above 20% is critical (in red), between 10% and 20% is high (in orange), between 3% and 10% is medium (in yellow), and below 3% is low (in green). How did we arrive at those standards? 10% — the threshold between high and medium — is the World Health Organization’s recommendation for maximum test positivity (i.e., you want to be under 10%). Harvard University epidemiologist William Hanague and Admiral Brett Giroir, a member of the White House Coronavirus Task Force, have also both cited the 10% figure. 20%, the threshold between critical and high, is twice that number. The 3% cutoff between medium and low is roughly the rate achieved by the countries successfully containing COVID. South Korea, for instance, has a test positivity rate of 1.6% (as of 12 May). ICU Safety Margin If or when there’s a new wave of COVID infections, are our hospitals ready to absorb a possible surge of patients? This question is answered by our third metric: COVID ICU capacity. The White House’s reopening plan, AEI’s plan, and Harvard University’s Safra Center for Ethic’s plan all cite health system capacity as an essential metric for reopening. But how do we measure it? It’s a bit wonky, so let us explain by example: Here is a chart of ICU headroom in New York as of June 4. New York has 4,121 ICU beds. We estimate that 1359 are occupied. We make that calculation by taking the typical number of occupied beds and applying a 30% “decompensation factor” to account for deferral of elective procedures and/or surged ICU capacity.  Of the remaining 2,762 ICU beds, we estimate that 1,037 are occupied by COVID cases, which amounts to 38% of ICU headroom. The 38% statistic is indicative of the healthcare system’s ability to absorb a surge of new COVID hospitalizations, should a new wave of infections occur. This metric is very much an approximation of hospital and ICU surge capacity; it just gives a rough sense. We use 50% available ICU headroom as the threshold between low and medium per Resolve to Save Live’s policy recommendation. They suggest that in order for states to consider reopening, COVID cases in the ICU should be able to double without overwhelming hospital systems. Contact Tracing When people contract the virus, they do not show symptoms right away. Even as states begin to reopen, people will need to quarantine themselves if they have been silently exposed to someone with COVID. How will they know? That’s where contact tracing comes in. Because of this problem, it is critical that enough tracing capacity exists to rapidly trace the contacts of individuals who test positive for COVID. Those contacts can be tested, quarantined (if necessary), and asked about whom else they have come into contact with. Because exposed individuals begin infecting others, it is critical that this process be completed in less than 48 hours. If this routine of testing and tracing is done quickly and completely, it can contain COVID, as we have seen in South Korea and Taiwan, and without the need for costly lockdowns. The White House Coronavirus Task Force’s Guidelines say that contact tracing is a “core responsibility” of states in order to be prepared to reopen. As of May 21, 27 states (CA, CT, DE, FL, HA, IL, IN, KS, KY, ME, MD, MI, MO, MT, NE, NV, NM, NY, NC, ND, OH, PA, RI, SD, WA, WV, and WI) call for contact tracing in their reopening plans. The American Enterprise Institute’s roadmap to reopening says states must massively scale up contact tracing and isolation/quarantine of traced contacts. So how do we calculate contact tracing capacity? Experts recommend tracing contacts of someone who tests positive for COVID within 24 hours, to contain the potential of transmission. Based on conversations with practitioners and public health experts, our metric assumes that tracing all contacts for each new positive COVID case requires an average of five full-time contact tracers.  Therefore, our contact tracing metric measures the percentage of new cases for which all contacts can be traced within 48 hours relative to available contact tracing staff in the state (assuming 1:5 new-positive-COVID-case:contact-tracing-staff ratio). We use green if greater than 90% of the contacts can be traced within 48 hours, yellow if between 20% and 90% of the contacts can be traced within 48 hours, orange if between 7% and 20% of the contacts can be traced within 48 hours, and red if fewer than 7% of the contacts can be traced within 48 hours. Here is an example from Wyoming: As of June 13, Wyoming has an average of 12 new cases per day. If Wyoming needs 5 contact tracers per case, that would be 60 contact tracers necessary to trace all cases in 48 hours. Since Wyoming has 50 contact tracers, that is enough to trace 82% of cases. How did we choose our targets? Research estimates that the infection rate can be driven below 1.0 if 70-90% of cases are identified and 70-90% of those contacts are traced and isolated within 48 hours or less. Therefore, we chose 90% as the cut-off between green and yellow. The boundaries between yellow, orange, and red are trickier. When less than 90% of positive cases have their contacts traced within 48 hours, contact tracing will likely be insufficient to contain COVID. We use 90% as the boundary between green and yellow. In the absence of expert consensus, we have set inclusive lower thresholds for yellow and orange. We peg the cut-off between yellow and orange at 20% — the number required for there to be one contact tracer per active case per 48 hours — and the cut-off between orange and red at 7%. Every state currently coded red is either currently experiencing a new outbreak or effectively has no tracing capacity. A state can become green either by increasing the number of contact tracers, or by decreasing the number of new daily COVID infections. Are These Four Metrics Enough to Assess Reopening? No. These four metrics are not the be-all, end-all that tells us whether a state or county is able to safely reopen. It’s a complicated case-by-case decision and includes other factors such as PPE availability, ability to trace positive cases, and much, much more. Nevertheless, these four metrics provide important benchmarks as we figure out how to reopen our country safely. Our hope is to create a common, data-driven language for discussing our COVID response and to provide a foundation that healthcare and policy decision-makers can build on. Check out our Youtube Channel for our educational videos. To learn more about Covid Act Now, visit our about page. For more information please reach out on our contact page or by email at info@covidactnow.org. Sign up to our alerts to stay up-to-date on the COVID risk level in your area.

Exporting Covid Act Now Data to a Spreadsheet

If you’re more familiar with editing or manipulating a spreadsheet than you are with code, we have good news! You can now export the data from Covid Act Now‘s model to your favorite spreadsheet, such as Google Sheets or Microsoft Excel.

The Covid Act Now API, which we launched yesterday, exports in CSV (comma separated value) format. This format can be easily imported into a spreadsheet.

We offer forward projections on the effects of COVID-19 for 50 U.S. States and more than 2,000 U.S. Counties. The projections are available by intervention type. (For more on “inference projections,” see this blog post.)

No Action TakenSocial DistancingShelter-in-Place“Inference Projections”U.S. StatesDownloadDownloadDownloadDownloadU.S. CountiesDownloadDownloadDownloadDownload

To import into Microsoft Excel, first open Excel and select the File Menu. Then, choose Open (File => Open). Navigate and select the CSV you just downloaded.

To import into Google Sheets, first create a new blank Google Sheet. Then choose File => Open. Choose Upload and drag-and-drop the CSV file you just downloaded.

The data is updated at least every three days, and we include a “last updated” field in the download so you can ensure your data is fresh.

If you are in the government using our data to plan your response to COVID, you can reach out to us at gov@covidactnow.org. Or if you’d like to provide general feedback, thoughts, or questions, you can email us at info@covidactnow.org.

Announcing the CovidActNow.org API

Covid Act Now (CAN) is a non-profit organization of technologists, epidemiologists, and medical professionals working to model how COVID-19 will spread in each U.S. state and county. 

We published the first version of our model on March 20. Since then, 10+ million Americans have viewed the model and we’ve engaged with dozens of government officials, including the U.S. Military and White House, to assist with response planning.

Today, we are launching an API to make our data programmatically available to everyone.

Our API exposes:

Reported Data: State and county level data for confirmed cases, deaths, and hospital bed capacity. The data is collected from a number of sources, including The New York Times, and is updated daily.Forward Projections: State and county level projections for hospitalizations and deaths based on several possible interventions. This data is generated from our model.

By launching a public API, we are making the data that powers CovidActNow.org available to anyone, free of charge, under the Creative Commons 4.0 license. The data is updated daily around midnight U.S. Pacific Time, and is available in both JSON and CSV format.

You can view the documentation on GitHub.

By making this data available, we are hoping it will be a helpful input into COVID efforts such as response planning, reopening initiatives, data visualization, and the creation of new tools.

Background

Since Covid Act Now launched on March 20, our team has spent significant time refining our model. It is originally based on a traditional SEIR model by Dr. Alison Hill at Harvard, and now Dr. Rebecca Katz and her team at the Georgetown Center for Global Health Science and Security audits our work. In the past few weeks we’ve made several improvements, including adding hospitalization and severity rates, data sources, and inference projections.

Going forward, our organization will be focusing on providing data to leaders to inform their decisions around reopening safely. 

Our work is being done in the open, and you can find our model open-sourced on GitHub.

Using the API

To get data from our API, you’ll need to construct a URL with the state or county, the intervention type, and append .json or .csv. For example:

+ [ STATE CODE (CA, PA, NV) ] + [ .INTERVENTION_OPTION ] + [ .timeseries.json ]

JSON objects can be easily manipulated with code, while CSV (comma separated values) can be easily imported into Google Sheets or Excel.

The intervention choices are:

API ParameterIntervention TypeR AssumptionsNO_INTERVENTIONNone3.7WEAK_INTERVENTIONSocial Distancing1.7STRONG_INTERVENTIONShelter in Place1.1 for 4 weeks, 1.0 for 4 weeks, 0.8 for 4 weeks.OBSERVED_INTERVENTION–A dynamic forecast based on the observed data in a given U.S. state

Two notes on interventions:

OBERSERVED_INTERVENTION infers an R(t) value from recent cases, hospitalizations, and deaths in each state.To see more details on each intervention and sources for our R assumptions, please see our model references and assumptions.

For example, this link returns an aggregated list of how COVID will spread in every U.S. State based on observed data, in CSV format:

And this is Dakota County, Minnesota’s data for how COVID would spread if it reopened everything tomorrow, in JSON format. (Note that we’re constructing the URL with the FIPS code for the county.)

Open this link in your browser now to see the data:

Hopefully these examples give you a taste of how to use the API. For all the details, see the documentation on GitHub.

In the future, we intend to make more data available, including:

Additional file formats like shapefiles for GIS systems.Integrations with data visualization products, like Tableau.

Please Play Around!

In order to accelerate decision making, safely re-open the country, and, ultimately, save lives, a massive, unprecedented collaboration across government, business, and the general public is needed. 

We hope this data plays a small part in that collaboration, and we’re excited to see how it is used, visualized, integrated, and transformed to make a difference.

To learn more about Covid Act Now, visit our about page. For more information please reach out on our contact page or by email at info@covidactnow.org.

Happy coding!

Thanks to Addy Osmani, Chad Arimura, Ilya Voldarsky, Josh Dzielak, Max Lynch, Paul Irish, and Steve Wilmott for reading drafts of this post and giving feedback on the API.

Return to covidactnow.org

LINKS

Recent Posts