In which I share a bunch of charts.
We’ve been comparing the US data on testing, infectivity, and case fatality with those of other countries. So now we’ll look at how this looks across the US. (There are probably much prettier graphs and maps out there, but these are mine)
Not surprisingly, the New York region (390), including New Jersey (210), Connecticut (88), and Massachusetts (95) has the highest number of cases per 100,000 in the population. Louisiana sits at number 3 with 113 cases per 100,00. But does that reflect real infection rates?
Who is doing the most testing? New York, of course, has tested the most. Many other states have tested more than 500 per 100,000 and are finding lower infection rates.
For some states, the low number of cases only reflects the low rate of testing. If you don’t look for it, you won’t find it.
The California data are misleading. The state has reported 6932 cases but with a population of 39.6 million, that’s a pretty low proportion of people. Most importantly, they also have a backlog of more than 50,000 tests. If we include that backlog, they are testing 322 people per every 100,000, but that means little without results. That hotspot in Southern California is where my family lives, so you can bet I’m paying attention. More about California’s testing crisis here.
Interestingly, the rate of hospitalization doesn’t correlate with the number of cases by state. This may be misleading as well, though. Oklahoma has the highest rate of hospitalization (31.3%) followed by Ohio (26.6%). Oklahoma has tested the smallest proportion of its population in the country, and Ohio is far down on the list as well. They are likely testing their sickest patients, so the data are skewed to present a worse scenario. While New York is struggling under the shear volume of cases, particularly in NYC, they rank 12th in hospitalizations.
A couple of weeks ago, Washington had a very high case-fatality rate, but has now been surpassed by Louisiana (4.6%), Vermont (4.4%), and Oklahoma (4.1%). All except Oklahoma have ramped up their testing, so we would have expected the case fatality rate to drop. That it hasn’t suggests that the populations impacted are already the most high risk. Maybe some people from those states can chime in to describe the local situation.
We use the data we have to better understand the epidemic.
Pandemics typically happen in waves. In the 2009 H1N1 pandemic, the second wave was actually bigger than the first.
We want to prevent a repeat of the Second Wave of H1N1.
While places like NYC are overwhelmed, other states are still well ahead of the curve. They can limit that first peak and flatten the curve by incorporating T3: Testing, Tracing, and Tracking into their prevention and control measures.
If current lock down efforts have been effective, contact tracing now will mostly identify only household members. We already presume household members to also be positive. So why bother?
Because there will be the odd case that has contacts outside the home. Identifying and isolating these contacts becomes critical. This is particularly important if they are essential workers, such as grocery store staff.
Most importantly, we can’t let up. It may seem that as we reach the lower threshold, in that “valley,” we can stop T3. This is when it will be most critical. There will still be cases. We will not get to zero anytime in the next year. As cities attempt to re-open for business, which must happen, testing, tracing, and tracking with isolation will be the only way to limit that second wave.
If it seems like I’ve said that before, it’s because I have. And I’ll keep saying it. While places like the greater New York region struggle to gain control of this outbreak, the rest of us can help by limiting its spread in our own communities. A pandemic doesn’t end in two months. We must utilize our long term strategy.