COVID-19 Analysis: Model of Cases Averted And Lives Saved By Social Distancing

As COVID-19 shot up to become the #1 Killer in the US this month, people were changing behaviors to #StayHome, wash hands, wear masks, otherwise social distance. This behavior change, partially government mandated and partially good citizenship is flattening the curve, averting nearly 1.7 million cases, and saving over 25,000 lives -- just from our behaviors in April.

Don't get me wrong, we are still growing our case count above the 20% weekly growth rate that would deprive the coronavirus of hosts and cause it to burn out. But each day, that weekly rate of growth is lowering. As of yesterday, it is 36% versus 58% a week earlier. If you can, tweet your governor and give them a word of encouragement because we need a couple of weeks below 20% to keep the coronavirus from performing a second act on us. As always, let me explain the math and the model and point you to the data for your state.

The model is based on confirmed cases for the US, by state and county from John Hopkins GitHub Repository. The Social Distancing data is from Geopath and Intermx. I analyzed the data up through April 15 and used April 16th through 20th as a hold out to check the model forecast accuracy (which is within 1% of actual). I broke down the time series into the "pre" social distancing/government #StayHome initiatives (grey line) and a "post" (yellow line). I fit a polynomial curve to forecast, and calculated the difference between the trend we were on before social distancing to the curve after.

I built individual models for each of the 50 states, and compared the total of the individual models to the overall model, and they converge with the overall US model. I applied a 1.5% Infection Fatality Rate (IFR), which is conservative. The current Case Fatality Rate (CFR) is 4.4% in the US, but varies greatly from state to state, likely indicating that some states are doing a better job at testing and finding infections than others. If I had used CFR, the total lives saved would be 78,653. The IFR is based on my research from the Diamond Princess Cruise, which is one of few closed population studies where everyone was tested, and we know the denominator and outcomes after sixty days of infection. This analysis only considers lives saved based on cases averted in April and uses the more conservative IFR rate. The death would have occurred between two weeks to two months after infection. Cases averted would have also spread the disease to more people, however the analysis only focuses on the initial cases averted and deaths averted.

Why This Analysis Matters Now:

Consider these two charts. The first one expands on my analysis back in February, where I examined what if NYC was like Wuhan, and how COVID-19 would become the leading cause of death. As of last week, COVID-19 became the #1 killer in the US. Business Insider did an effective chart showing how it compared to other leading causes of death. I've overlaid each states current rate of death per million. Unlike most causes of death, which are predictable and relatively stable over time, COVID-19 is increasing at an alarming rate -- 92% overall in the US.

The second chart is my forecast of what happens if we don't lower the weekly rate of spread below 20% and keep it there for a few weeks. Consider the relatively small bump we are experiencing now is based on less than 5% of the US infected with COVID-19. If we lack the will to make our stand now, I doubt we will have the will come the second wave -- and the math shows it will be much worse as we get up toward 20% infected. Consider the red COVID Death Rate line in the chart above corresponds to the small bump in the chart below. Consider the implications if we have far more cases come the fall. Help encourage others that #StayHome is the smart and patriotic action in this moment. Celebrate the success of Governors and Citizens in saving over 25,000 lives, just based on what we've done in April.

The Briggs family goes far back in Texas history and family legend says we took a last stand in the Alamo. I don't know if that is true. I suspect if you added up all the families in Texas that count a relative among the Texas brave, we would have had far superior numbers to Santa Ana. Yet, I'd like to think that the spirit of the Alamo, determination in the face of overwhelming odds, is with us all as a country.

What we are being asked to do with #StayHome is inconvenient for some and devastating to others. The economy has been shredded, people are restless, but going back a few weeks too early can't and won't reverse much in the economy. It can only make things worse for 2020.

Rebuilding the economy will be our next monumental challenge. Let's take a moment to be grateful for the cooperation in #StayHome and the over 25,000 lives saved.

Want to see how your state compares? I created a website with a deeper look at this analysis and a state-by-state breakdown. No one will confuse me with a web developer. I picked up a few courses at the start of this year to deepen my appreciation for the work of the fine developers in my life. Forgive any rough edges.