The Tragedy of Arizona: We Can Crush the Curve, but It Takes Time

Despite Vichy Republicans giving up on fighting COVID-19, there’s no reason to give up. We still can turn and fight this, though it will take a lot longer now. And no state illustrates this more than Arizona.

In the two weeks preceding July 4, Arizona has had 42,163 cases of COVID, resulting in a two-week prevalence of 0.58% (number of cases over two weeks divided by the total population), but that’s probably a massive underestimate, as the percent positive tests during that time frame is 24.3%, meaning there are lots of people who need tests and aren’t getting them (news reports bear that out). A charitable two-week prevalence for Arizona is probably one percent (though it could be higher than that, easily).

For Arizona to decline to the prevalence threshold Germany uses to throw the emergency brake for a region, 0.05% prevalence (yes, that’s how bad it is in Arizona), Arizona likely needs a twenty-fold reduction just to reach the ‘oh shit! Break the glass!’ threshold Germany uses. This can be done, because D.C. did the same thing. It took about three and a half months of strict measures, but D.C. did do this.

We can do some back of the envelope calculations. Like I noted in a post about prevalence, these numbers, while precise, should not be necessarily seen as accurate, but they do indicate roughly the scope of the problem. When we look at the Rt (the number of new infections arising from a single infection; greater than one means it’s spreading, less than one means it’s receding) for states that have managed to mash things down, Rt is usually around 0.75 (though many of the heavily affected states have never had an Rt that low). If we assume that it takes six days for the next group of people to be infected, we can dummy out what this looks like.

(technical note: Obviously, it will be more complex. To start with, I’m discretizing a continuous process. The ability to infect begins at different times for different people, and is some kind of curve, not a simple function. Rt won’t remain constant, and is hard to estimate when case numbers are small. Likewise, six days might be too long an interval (or too short). Plus many other complications. But as a toy example it illustrates the key point, which we’re about to get to).

Suppose we’re stating at 1,000 new cases a day, and we’re shooting for 50 new cases per day (i.e., 20-fold decrease). If Rt equals 0.75, then in six days, the number of daily new cases will be 750. Six days after that, 563 new cases per day. Six days after that, 422 cases per day, and on, and on (and on…). If you’re patient you can reach your target, but it’s going to take a long time, somewhere in the ballpark of two months.

For Arizona, to get to the German threshold, it could take around two months, provided, of course, that Arizonans can get their shit together and do the fucking job. To reach what I would call ‘protective prevalence‘, where it’s low enough to survive the occasional bout of stupidity, would probably take another five to six weeks after that.

Again, these are ballpark estimates, though they worked pretty well for D.C., and they give an indication of how long it will take–much longer than fourteen days.

So, why did I refer to the tragedy of Arizona? In the last two weeks of April, Arizona had a two week prevalence of 0.04%. To crush this to 0.01% or lower would have taken a few weeks, and they obviously would be in a much better position than they are now.

Instead, Arizona’s leadership got* stupid, and now COVID-19 rages out of control. And there’s no reason to think ‘what happens in Arizona, stays in Arizona.’ As deputy CDC director Anne Schuchat put it, “We are not even beginning to be over this… As much as we’ve studied [the 1918 flu pandemic], I think what we’re experiencing as a global community is really bad and it’s similar to that 1918 transformational experience.”

It’s going to be a long summer and fall, and it didn’t have to be. Remember that when you vote.

*Or continued wallowing in its stupidity, depending on one’s political predilictions.

This entry was posted in COVID-19. Bookmark the permalink.

3 Responses to The Tragedy of Arizona: We Can Crush the Curve, but It Takes Time

  1. Dana says:

    The math looks good but it doesn’t take into account that Arizona people vacation in California (San Diego!) and would have brought it back. It would be hard to keep it low even if you got it low to start. That’s why I expect the 50 states to continue infecting each other in an endless cycle… Now if we had real contact tracing that might help keep it down.

  2. coloncancercommunity says:

    Even if AZ was willing to do what it took to get the prevalence down to the 0.01% level, the investment required in the form of contact tracing and support for those put in isolation would be something the federal government would have to help take on. I don’t see ANY WAY that you could get the funding required. The other requirement would be MASSIVE stimulus packages for every.single.resident to keep them financially whole why they were forced to stay home.

    Even if you could get all that, getting the cooperation of a willfully ignorant population to go along with it, would be well-nigh impossible.

    Providing you had the support described, you MIGHT have regions like the northeast and states like CA and WA willing to do the work to get there. But those areas would then be in a perpetual state of revolving restrictions (even lockdowns) when people from out-of-state would come in and undo all that hard work by reinfecting the population.

    In a country where a good part of the population has a meltdown over simply wearing a MASK in public, I don’t see that happening.

    In the end, it isn’t rocket science. But the ability of large swathes of our population to understand and cooperate with such a plan is nearly nil. You have it some regions, but you would have to drag everyone else kicking and screaming out this pandemic.

    • dr2chase says:

      Massachusetts did a good job till late June, we’ve since gone flat. If you shut down every indoor spreading behavior that you can (bars, restaurants, casinos, gyms, theaters, churches, and unfortunately, schools) , get everyone wearing masks, and then add test, trace, and quarantine, it comes down at a steady pace. And this was with not-perfect Masshole mask use, about 10% of the people in the grocery story wearing masks with exhalation valves (there ought to be standards), some clearly get the seal wrong (one, with the angle right, I could see daylight between mask and face), and of course all the nose-free-or-die people.

      And yes, there’s quarantines after movement into the state.

Comments are closed.