Underneath the Panic
Recently, the Dallas Police Department apparently (seen in a video someone took from their TV where local news was making the claim, but I’ve not found it on their site) announced they were going to stop responding to certain calls, blaming SARS-CoV-2 and COVID-19 for it. But is it wise, and it is justified?
First, we have to consider the wisdom of the PD announcing to the public they won’t respond anymore if your car gets broken into, someone is shoplifting, or acts of vandalism. Leaving aside whether they have such a policy for a moment, let us question the wisdom of telling criminals and would-be criminals they won’t go after them.
In a word, it is unwise at best, and blatantly stupid more accurately. Most of these types of crimes are crimes of opportunity. This means that given the opportunity the perpetrator will consider the chance and the chance of getting caught. It doesn’t take a criminologist to know that if you tell people the police won’t even respond if the store calls the cops because they caught you stealing that almost all risk associated with that opportunity is gone and you can expect the rate of that to rise.
Will we ever know, however? Will they record the calls and mark them as “not responded to” so an analysis could be done later? Doubtful. Further, when people know their call won’t do any good what will they do? This would decrease the call volume and thus skew the data, if it were collected, quite heavily.
This also brings to mind the question of what will happen when someone catches someone doing those things, knowing the police won’t be bothered to respond. I suspect an increase in vigilantism to make up for it, at least partially. Though I do wonder if calling it vigilantism is accurate when the law enforcement straight up abdicates that role.
So there is certainly no wisdom in the announcement or policy from a policing standpoint. But what about the panic aspect? Is the action justified? To answer that you have to ask a few questions:
- What is the risk absent the action?
- What is the reduction in said risk?
- How does that reduction weigh versus the increased harmful effects of the policy and announcement?
Let us look at the first question: what is the risk absent the action?
To answer that you have to look at the odds of the police contracting and spreading the disease, and the effects of doing so. To get there we look at some basic numbers. The population of Dallas is around 1,350,000 people. The current number of cases of COVID-19 there is around 38 — according to the Texas State Health Service.
So presently so the current known case load in that population is .0028% of the population. So, statistically speaking — and if those cases were evenly distributed — you would have to come into closer personal contact with 40,000 to have a chance at being exposed. Even if you want to say “but there may be more people who have it and are not tested” it doesn’t significantly increase that risk. Now there are two, so far, cases of a DPD officer with the virus. But they were isolated and their contacts have not shown to have got the virus.
That last bit may be surprising to anyone only following main stream news, politicians, and celebrities. After all, we are repeatedly told it is “highly contagious/infectious”. Yet it isn’t. While the formula is not complex, determining it is, yet we do have well tested means to determine it within reason.
To figure it out you take the probability of infection from a given contact, multiply it by the rate of contact and multiply that by how long the infectious period is. Obviously if any of those numbers is high, the end result will be. But what do we have for this virus? 1.2–2.2. How does that compare?
In short, it isn’t as infectious as most things we encounter and isn’t significantly more infectious than seasonal influenza, SARS, H1N1 from 2009, and rhinovirus (one of the two main causes of common cold — the other one being corona virus 229e in the U.S.). The “base reproduction rate” (known as “r0” value) is in the range of 1.2 to 2.2 — and even lower in some populations — and significantly higher in some tight populations.
This is the average number of people someone with it will infect while they have it. Now, if this is your first time with this value that may seem high. After all infecting 1.5 other people seems pretty contagious. But first you have to realize that is out of everyone that person comes into contact with over the duration of them being contagious. Think about how many people you come into contact with over the course of a cold — say 2–3 weeks. Now you can begin to see how low that number is. But how does it compare to other viruses? Not especially high.
A value of 2 — such as Ebola has — is considered low by actual experts in the field. Seasonal influenza (made up of many strains) runs in the 1.4 to 2.4 range depending on the season and study. You can get a decent visual comparison of well known infections here:
The summary: the common cold is about 3–5 times as contagious, influenza is slightly more contagious, and SARS was slightly more contagious, while “swine flu” (H1N1–2009) was about as contagious as SARS-CoV_2 (the virus that causes COVID-19). So, the hysteria over it being highly contagious is just that — hysteria. The actual data shows it to be overblown.
In another example, say you don’t trust the data from China, you can look to studies done in the U.S., and find the first case: the person in Illinois. They tracked all known contacts for the person — who got it while travelling — over the course of it. This was hundreds of people and included medical providers and a spouse. The only person who got it was the spouse.
Or look to the cruise ships. The infection rate among the cruise ships was also really low. Indeed when you did into them, outside of the first ship the numbers of cases onboard are really low — and almost all of them did not get it from someone on the ship. While the cruise ships are enormous, you can’t exactly say people are not in close contact with each other. If it were highly contagious you’d expect most of those people to have it. They don’t.
Or even imagine that one person brought it to Dallas and ALL of the remaining cases were confirmed via lab tests and a result of “community spread”. The first “presumptive confirmed” cases, more about the quoted it in a moment, were a person who had been travelling extensively, and someone in close personal contact. So imagine that that was the source of all cases in the Dallas area. That means that over the course of two weeks The DFW area has had about 250 cases out of almost 7 million people. Not exactly a fast spread there either — and we know it isn’t from just one pair of people.
So in reality the risk of exposure is lower than those who profit from the panic would have you believe (news, politicians, and celebrities and would-be celebs).
Now for a brief sidebar on that “presumptive confirmed.” While there was much talk about China’s changing definitions, we aren’t really that much better. Presumptive confirmed seems to be the same a “clinically confirmed” — which is not “lab confirmed”. Most of us likely thing “confirmed” means you took a test and the lab results said you had it. And this is where the media is falling down and failing. Because “clinically confirmed” is what they almost always use but don’t tell you it just means you went to a doctor and they judged your symptoms — no actual test confirmation, just that your symptoms matched. So take “confirmed case” counts that do not specify as probably larger than actual.
But, what about the fatality rate?
Again, we have a hard time overall because we’re looking at early data — which is always worse than overall — and because we’re being too broad. Even in China’s numbers you can see the familiar pattern of a novel virus outbreak wreaking havoc where it pops up initially then being much less so. Did you know that while the reported Case Fatality Rate in Huwai province (where Wuhan is) is about 2% but under 0.4% outside of it but still in China? Not likely
Mortality rate is hard to determine in the early stages and the media screw it up badly. They’ll look at “confirmed cases” (I’ll get to that) and divide out the deaths. But that isn’t reasonable nor is it accurate. You have to take cases with a resolution rather than all cases.
Obviously, when most of your cases are still active, that number won’t be very reasonable — most of the deaths are “front-loaded”. People who die as a result of the infection tend to die early and quickly while those who recover do so slowly. And here you get into differences in population. Most of you probably know that if you’re 75 and get it you are much more likely to die than if you’re 25 and in the same general state of health. This isn’t that different from influenza — other than influenza has a higher rate among kids than sars-cov-2 does.
But overall the data so far, while trending down as expected, is looking like anywhere from 0.4% to 1.4% depending on age range among adults. And here is where we diverge from influenza and where the panic mongering sets in. The paniceers will do math like “if 5% of the world gets it and 1% die, that is XX million people”.
But they do this with all of them. And none of them have ever come to that. Part of the problem with comparing it to influenza on mortality is that you usually see influenza given as a “per 100,000 people in the population” mortality rate but we don’t yet have that kind of data for this one.
So instead we could look at that broader rate, or at least try to. China has been hit pretty hard, I think we can agree on that. With a population of nearly 1.5 billion, and the source point for the virus, what portion of their population has died from it? just under 3500 — for about 0.0002%. The vast majority of their cases have been resolved. Now this is with a CFR of about 3%. If we took that same calculation for U.S. deaths of now, despite it being early, we’d see our percent of population at 0.0001 with a raw CFR of 100%.
No, that isn’t a typo. As of the time I am writing this we have 499 deaths and 0 recovered. Therefore the “current” fatality rate of concluded cases is 100%. That alone should tell you how small and useless most of the broad data for the U.S. is.
So this is where the panic mongering comes into play. They focus on “if XXX million people get it, X million WILL DIE” but never tell you that the pattern they claim has never held up. It makes it rough for people to compare it to influenza because millions of people do get that and tens of thousands die.
Which brings us to the risk-abatement issue. In a situation where you’re in a lockdown or “shelter in place” situation, not responding to calls isn’t going to be a significant risk abatement method. Especially when proper personal sanitation and care about where/when you touch your face is so much more effective. And that is just regarding risk of infection.
Risk of death goes down dramatically when you’re on top of the illness — this is the underlying reason for the low mortality rate of seasonal influenza. In both influenza induced illness and COVID-19 the main complication is respiratory. Speaking as someone who has had severe life threatening respiratory issues early detection and treatment can eliminate 99% of the problem.
This is one of the reasons the lethality of a virus goes down over time. It is rare the actual cause of death is the virus but the pneumonia and similar respiratory problems that can come from it. How does this apply to risk abatement in the case of this policy? The police document who they come into contact with, meaning they know who might have infected them or become infected by them. Thus this would be highly valuable in reducing the risk of death from an infection — if it happens.
So overall the risk abatement of this policy is miniscule at best. With a miniscule risk abatement benefit I doubt comparing that to the risks of increased crimes and possible increased cases of people taking the law into their hands would be in favor of the policy. Thus I judge this move to be unwise, and potentially more dangerous.
