US COVID-19 Trend Projections

Interactively compare trends of COVID-19 cases on a dateline.

The projections are based on exponential growth as viral infections grow exponentially up to the peak of infection, after which, recovery and immunity reduces the infection. For comparison purposes, we take the peak as half the population.

How to I make selections? Double-click the entry in the legend. Then click the current and worst trend entry, or any other entry. Double-click on your entry to bring everything back up.

Projected Total Cases on Log scale

What is a log graph? Values on the vertical axis increase in proportional chunks - 1, 10, 100, ...

Why log graphs? In a log graph, exponential growth appears as a straight line. Thus we can project the trend by extending the straight line, at least, to the peak of the infection, after which it will stop growing exponentially.

What is the worst projection? This is based on the 7-day period that gave the fastest growth.

What is the current projection? This is based on the last 5 days. If it is less steep than the slope of the worst projection, the curve is flattening, and we have delayed the peak of the infection.

Projected Total Cases

To see the exponential nature of the growth rates, the projections have been plotted on a normal graph.

What is important here? Both the worst and current projections will reach the same peak of infection but at different times. The difference in the delay of the peak will affect the number of New Cases per day, especially at the peak, and this is explored in the next graph.

Projected New Cases

Why is this important? New cases per day directly relates to how many people will be hospitalized per day (~20%). By seeing the peak of new cases, we can see an estimate of the maximum burden that our hospitals will be placed under.

New cases are also exponential! New cases are calculated from the rate of growth of the Total Cases graph above. If you haven't studied this math before, you may be suprised to see that New Cases are also exponential.

What is important here? The difference we want to see is the reduction in the peak of New Cases between the worst and current projections. If it diminishes signficantly then the peak number of New Cases will have been reduced, allowing our hospitals to treat more sick patients effectively.

Projected Total Cases Per 100,000 on Log scale

Cases divided by the population, multiplied by 100,000

By plotting the projections per capita, one make a fairer comparison. In this graph, a peak of half the population corresponds to 50,000 cases per 100,000 people

Projected Total Cases Per 100,000

Cases divided by the population, multiplied by 100,000.

Projected New Cases Per 100,000

New Cases divided by the population, multiplied by 100,000

What is important here? One can get a sense of the comparative size of the initial outbreak, by the height of the worst projected peak. The effectiveness of the lockdown can be seen by the comparative reduction in the peak in the current projection compared to the worst projection.

Bosco Ho,
Plotly graphs,
Johns Hopkins CSSE COVID-19 data,
John Snow Labs country population data