by Hamilton E. Davis
A kind of miasma is hanging over Vermont as this is being written on Sunday afternoon. We’ve been pretty much locked down for a couple of weeks now, which actually feels like a couple of months. Many of the lockees have little to do but stew about the coming surge, expected over the next couple of weeks, and worry on a longer reach about when we can get to a more normal place. A tsunami of information from the web is washing over us, but we somehow don’t feel that we know very much.
In fact, however, there is a lot of movement under the surface. One of the most important developments, entirely unremarked as far as I know, is that Vermont state government has embarked on an effort to move its virus management strategy closer to that elusive target—the best case scenario. The vehicle for that effort is something called “contact tracing,” which I described in an earlier Virus Paper as “tracking.” Contact tracing is the policy scalpel South Korea used to carve the virus right out of the guts of their country.
The idea is simple enough, although the execution is fraught. It works like this: when a person gets the virus you isolate and treat him or her. But then, you interview that patient to determine who he or she has been in contact with in the previous couple of weeks. Then you seek out and test each one of those contacts. If there are 15 and three of those tests positive—with or without symptoms—you isolate them and treat them till they test negative; meanwhile you interview the three positives to determine whether they have the virus and repeat the whole process.
Managing the virus across the state like this amounts to a huge effort, but it enabled China and South Korea to stop the beast in its tracks. China isn’t a model for us because it is an authoritarian country, but there is nothing to prevent Vermont from following the track laid down by South Korea. The Scott administration has produced a model of how it might work, and the potential benefit if it does, even partially:
The top line in red looks like Mt. Philo in Chittenden County, and it’s real scary—it blows right past our medical resources over the next two or three weeks, by a factor of four or five. The capacity in this view of the situation in this graph is shown by the horizontal line labeled 622. Fortunately, we’re not going there. The putative capacity is 622 beds and there are also caps on the available Intensive Care Units and ventilators; there has been some speculation that the latter two caps could be breached. The capacity numbers, however, have a fat margin of error because they do not include any of the resources available at Dartmouth-Hitchcock Medical Center (DH) in nearby New Hampshire. I’ll get to the DH in another Paper, but in normal times 40 percent of DH’s traffic comes from Vermont, and it is unlikely that DH’s management would ignore that…
In any event, the shallow green curve immediately below the horizontal is the track we are on now; it is the product of a function called an epidemiological curve, and it is dramatically different from the exponential curve that we have been riding on since early March. Over that three-week stretch, the number of confirmed cases were doubling every three days or so. I am working with Easton White, Ph.D, a researcher on biological monitoring at UVM, and White spotted the shift in function on March 27, when the doubling period itself doubled, from just under three days to just under six.
In Mathland, the early exponential track is well within the capacity of a high school student, but the epi curve is anything but. Even my tiny corps of brilliant readers will be forgiven from taking a pass on that; it will take the corps from high school math, a cinch for them, to the cold mists north of ordinary differential and integral calculus. I have looked at the generic epi formulas; and, trust me, one of those could wreck your whole day. Even the pros, like White and the team of experts now working for the state, aren’t running these calculations on the backs of envelopes—they are dumping their data into computer programs that can do things like evaluate an integral that would make your head hurt.
Anyway, it’s that middle curve that will control our lives until we get by the virus—specifically, it will tell us when it is safe to begin re-igniting our economy. At this point, the curve shows us getting close to normal by June. However:
It is absolutely critical to note that not a single point on the green curve is guaranteed past April 2, the day it was published. Vermont’s performance has been solidly within the relatively optimistic range, but that performance depends on continuing the rigorous passive defense we have mounted so far—shutting down human contact as far as possible, closing businesses, etc. If we take our collective foot off the brakes, however, there is nothing to prevent moving the green line in the wrong direction, to the unforgiving red slopes of Mt. Philo.
The sermon for today though is about a cautious reason for real optimism, the potential to move the what’s-actually-happening green line to the lower blue line, the best case scenario. The full potential of such a move can be seen in the numbers listed on the two lines. At the peak, the green curve shows us needing 521 hospital beds. The blue curve at the same time shows us needing just 170, a three-fold improvement. And the slope of the blue line moves asymptotically to normal quite rapidly. Just looking at that curve evokes for me the recent headline on a New York Times piece by Maureen Dowd, “Let’s ‘Kick Corona Virus’s Ass’.”
Can Vermont Actually Do That?
Well, the Scotties are taking a shot at it, to their great credit. The story of the American response to the virus is a chronicle of blundering and mismanagement, from the patent absurdities coming daily out of Washington to the buffoonish maunderings of the nitwit governor of Florida. So, Vermont is punching well over its weight on the Covid challenge, but that doesn’t guarantee anything—just escaping a train wreck will be hard enough.
In fact, I was flabbergasted that they are trying. I asked the Covid team about it in a recent press conference and I thought it was pretty much a throw-away question. I swim in the ocean of crap that floods the web every day and I had seen no sign of tracking in other states. But the Scott team jumped right on it.
“Yes,” said Health Commissioner Mark Levine, without hesitation. (paraphrasing slightly.) “We are working on it now.”
I asked the obvious questions—can you do enough testing, and do you have the people necessary to do the interviewing?
Levine didn’t say yes exactly to either one, but the replies amounted in my view to a real effort. The state has considerably more testing reagent than it had earlier, Levine said, which means that it can go well beyond the earlier constraints that led to testing only people who had all the symptoms. Okay, that’s good. How good? Good enough apparently to begin testing non-obvious people, based on the interviewing of people with confirmed cases. That allows the managers to get deep into the sizeable potential that exists in the whole population. How deep? Impossible to say, but it’s reasonable to expect some inflection of the green line in the direction of the blue.
In fact, a few days later came what looked like a spike—I think it was 70 some new cases in one day. That was almost certainly an artifact of increased testing, rather than a movement back toward exponential increase. And I found it noteworthy that the World Health Organization (WHO) had recommend a 90 percent cold rate for testing: If you test 100 people and get more than 10 positives you aren’t testing enough. My calculations are that in recent days, the Vermont rate has been right around 10 percent, pretty much on the WHO target. In any event, Levine sounded determined to ramp up testing significantly. The testing isn’t that hard, however; you just need reagent—testing kits in the earlier vernacular. But the second half of contact tracing seemed intractable to me—who would do all that seeking out of exposed individuals, the interviewing.
Levine didn’t even slow down for that. The work would be done by personnel from the Department of Public Safety, AKA the Vermont State Police. Levine, I think, said they had 40 cops ready to go. (You can’t easily check anything in Lockdown Land). OK, perfect. Crime and traffic should be down, and state cops could basically do what they always do—go out and talk to people and find stuff out. They can worry less about burglaries and guys testing out the Porsche at 140 on I-89 east of Brookfield, and more about who is walking around undetected infecting other Vermonters.
Will it work? Nobody can say for sure, including the Scotties team. But it’s a gutsy effort, and every Vermonter ought to wish them well. They deserve it. And every increment that the green line can be leveraged in the direction of the blue will redound to the benefit of the whole state.
I am recommending that every one of my tiny corps of brilliant readers go out on his or her front porch and launch a cheer into the clear air in favor of contact tracing. It’s the most hopeful tactic we have.