COVID-19 — Personal Notes

Off the Cuff

— Personal comments on various topics —

— Less Testing = More Future Cases!!! —

”If we did half the testing we would have half the cases.”
                — a foolish dangerous lie spread repeatedly by
                    Donald Trump (most recently on 2020 Jul 15)

The U.S. ruler’s most recent repeated excuse for his failed leadership against the COVID-19 pandemic is not merely a matter of opinion or interpretation, nor as his chief of staff (note, not a knowledgeable expert on these matters) suggests merely a matter of which numbers to focus on. Trump’s statement is simply false as it stands, a fool’s outrageous misunderstanding of basic math. And since he keeps repeating it, it goes beyond simply being an incorrect answer on a fifth grade math test: it now rises to the level of being a blatant lie. And since its deception has already led other foolish leaders and too many citizens looking for an excuse to point the finger of blame elsewhere, thereby neglecting basic individual and social responsibility, it becomes a very dangerous lie.

The number of current COVID-19 cases is completely independent of the number of current tests being conducted. For the sake of this explanation, presume complete reliability of the tests, presume instantaneous test results, and presume that those being tested are representative of the full population of any given region (the realities of lack of complete confidence in test results, of delays in delivery of test results and of how severe symptoms must be in order to be eligible for testing only make Trump’s foolishness and lie all the more severe.) Say that out of 1 million people, 100 thousand are currently positive for COVID-19, i.e., a real positivity rate of 10%. If we test 5,000 people and if our assumptions are met, we would then know of 500 people with COVID-19 on the basis of that test. But how many cases do we have? Testing does not infect people, as would be suggested by belief in our ruler’s lie; rather, we still have 100 thousand currently positive — we know of 500 of those, and if our assumptions are met we can be concerned of 495,500 existing cases whom we have not yet identified. If we foolishly buy into our ruler’s ignorance and only conduct 2,500 tests, giving us only 250 cases, we have not changed the 100,000 current case count by one single person; all we have done is increase the number of cases about which we remain ignorant. No no a million times no, Mr. Ruler, half the testing does not give you half the cases; you still have the same number of cases and have done nothing but increase your ignorance of how high that number is.

And in the end, if you cut the number of tests in half, you will materially increase the number of future cases, since you will now have a greater number of infected people spreading the coronavirus to others.

Really, this is rather basic mathematics, so basic that one ought not need to ask an actuary. But our ruler cynically relies on misleading the public with something even more obvious than the danger of injecting bleach, trying to pretend that the surge shows how well the Administration is doing. The virus isn’t so easily fooled. Cut your tests in half, and it will spread out of control.

— Marking the Peak —

Still Not Reached the Peak as of April 17 — Based on an extraordinarily superficial and faulty appraisal, the U.S. COVID-19 team claimed on April 15 that the U.S. was “past the peak.” By any reasonable understanding of “peak,” that moment has not been reached by Friday, April 17, nor is the peak likely to be seen within the coming week. And falsely publicizing the peak will only push it out further away, as many workers and other citizens will relax distancing and other safeguards too early, thereby aiding persistence of the pandemic, if not resurgence. [ARL 2020 Apr 17 CDST 3:34]

Count of States with “Peak” of “New Cases” — Not the Best Gauge — Counting the states thought to have reached “peaks” of “new cases” is not a reliable basis on which to formulate polity of protection, mitigation and recovery. On April 15, the U.S. COVID-19 task force claimed that evidence supported a contention that the U.S. was past the peak of new cases. The first problem with that message was how quickly it got muddled: almost as soon as the words were spoken, the media — particularly media outlets that are politically prejudiced in favor of the current Administration — were abbreviating the claim, falsely raising hopes that the country had passed the worst of the pandemic. But more importantly, the measure itself is of little to no value in evaluating the current state or future of the pandemic.

Counting states thought to have reached the peak of new COVID-19 cases is akin to gauging property damage from a major hurricane by counting widespread hospitals that seem to have felt a lull in ambulances showing up at the ER door. What appears to be a majority of hospitals that overall are smaller and in less populated districts is easily overshadowed by the deluge of cases still coming through the halls of the heavily populated hospitals, even if the count of those hospitals is in the minority. And what’s coming through the doors doesn’t yet tell us what is still being collected by ambulance crews overwhelmed with patients still out in the field. And most importantly, even reaching any real and lasting peak of incoming patients doesn’t help if the pace of patients waiting for empty hospital beds continues to exceed the number dying or walking out the door.

Or picture a bathtub being filled with 50 faucets, some large, some small. After turning the flow in the faucets higher and higher, we turn the flow back from the max on the smaller faucets, but even for those smaller faucets the water coming in exceeds the rate at which the water can drain, while for too many of the larger faucets we continue to increase the incoming flow. We can’t measure any supposed peak by counting the number of faucets where we turned the flow back for perhaps a spell. We need to look at the rising level of water in the tub.

Not to mention all the additional variables an actuary would take into the analysis, ranging from potentially adverse factors such as the availability and quality of testing to potentially positive factors such as the possibility of improved medical procedures, maybe even a vaccine. None of which can be accurately incorporated into the flimsy indicator being touted by the U.S. COVID-19 team.

All things considered, the aggregate number of active cases provides the best leading indicator of any peak. “Leading,” since the period between identification of a new COVID-19 case and resolution can be at least two weeks, hence any actual peak in aggregate active cases might presage a delay in the peaks for medical needs, absences from employment, or deaths. The clear inadequacies of the peak basis cited by the U.S. COVID-19 task force can be readily seen in clear evidence that active cases continue to climb up to and well after the April 15 claim, further straining scarce medical resources and continuing to feed the death toll. And while it is hardly surprising that the political emphasis driving the task force would lean on perhaps the only blink of a glimmer in a still-looming storm of dark reality, it is disappointing that the scientists on the panel fail to season the outlook with more credible analysis. Maybe they need an actuary on the team. [ARL 2020 Apr 17 CDST 8:30]

— Death Toll —

I’m still projecting that by the end of April, the U.S. death toll will reach the rosy 60,000 projection most recently suggested by the U.S. Administration’s COVID-19 response team to be “most likely.” And that the U.S. will pass that milestone at a pace that will still have my early projection of 130,000 quite well in its sights. [ARL 2020 Apr 17 CDST 8:54]

— Mortality —

[Preliminary Note — Here – and in morbidity and in projections and in reserves and in employee benefits, all of which I will stretch to before I feel satisfied with this project – we come to one of the central themes of an actuary’s interest in studying COVID-19. And as of April 6, 2020, I might be almost ready to begin adding comment and opinion on this particular theme. I’m deepening my studies in that direction by poring through a recent actuarial study of mortality for specific causes of death, looking closely at data from death certificates collected by the National Center for Health Statistics, and stitching in data and information from several other sources. Watch for it….]

Extremely rough, very early, not-yet-settled back-of-the-envelope estimate of a mortality rate in the U.S., based on the proportion of closed cases that have ended in death contrasted with those that have ended in recovery, suggests that the eventual death count in the U.S. could exceed 130,000. Drop that down on the basis of a fair slice of those coming from New York City and other areas strained by the early crush, with hopes that flattening the curve for most other regions might permit hospitals to reduce the mortality rate in coming days. Increase it by the fact that we have not tested all those who already might be infected, plus those who will become exposed in the future. Drop it to the degree that social distancing and other measures might cut into future incidence and severity. Increase it by reliance on face cures and the potential for early relaxation of reasonable health measures in deference to pseudo-economic considerations. All things considered — but still pending much further and much more extensive analysis — I would currently call this estimate of 130,000 U.S. deaths to be a somewhat optimistic goal, one we are more likely to exceed than not. [2020 Apr 6]

Even Dr. Fauci now states that the number of U.S. deaths due to COVID-19 is “more likely 60,000” versus their 100k-240k range given 10-11 days ago. For now, I’m sticking with my first armchair estimate of 130k. Hoping for anything less than 100k at this stage is way too rosy, in my opinion, so much so that I feel somewhat disappointed to find myself suspecting the “more likely” possibility to be that Dr. Fauci has finally bent to the pressure of the political and economic members of the Administration’s team. [2020 Apr 10]

— Low Ten Shouldn’t Guide High Forty —

At its mid-April news conference foreshadowing policy aimed at re-opening the economy, one of the scientists on the Administration’s COVID-19 team lent support for premature relaxation of reasonable health standards by noting that up to that point 9 states still had less than 1,000 COVID-19 cases. Sad, dangerously sad, that a reputable scientist can abandon the simplest scientific principles and common sense in favor of yielding to political pressure.

The federal government has the power to set forth bifurcated standards that properly distinguish the factors driving the pandemic – one set for regions with lower risk, with a different set for higher-risk regions – in much the same manner that providing hurricane relief to a hard-hit Gulf state need not necessitate handing out at that same time equal amounts to states in the Rockies. But if a uniform standard is to be established for this national health emergency, then the “low ten” cannot govern the policy that needs to be in force for the “high forty.”

Using the same rule of thumb set forth by the Administration’s team, as of midnight starting April 15, the lowest ranked ten states – obviously, 20% of the number of states, as in 20% of the senators in the U.S. Senate – held nearly that same percentage in land mass, with nearly 17% of the acreage of the nation. But those states accounted for less than 5% of the country’s population. And the fact that those states had so far experienced an even lower portion of the country’s COVID-19 cases – about 1% – can be attributed in large part to the lower average population density for those ten states – about 10 persons per square kilometer, versus the national average population density of about 36.

Perhaps the point of these comparisons need to be explained to a ruler who relies on the will of an electoral college over that of the majority of voters; but it is incomprehensible how a reputable doctor and scientist can ignore the point. The coronavirus will not be so accommodating. Having been unleashed with a vengeance upon New York City, COVID-19 does not turn to Maine and Vermont and mete out the same headcount of cases. The virus spreads person to person, not state to state. Moreover, it ought to have been well-known and accepted by now – although the Administration team seems to forget the point when it’s politically expedient to paint the rosiest picture – that this virus runs rampant in areas of high population density. So telling us that the low-density states have been seeing lower transmission shouldn’t be guiding us on how to treat the high-density states, nor even the high-density cities within those lucky low ten. Last but not least in the practical sense of the desired economic impact, those low ten states contribute proportionately low figures to our damaged GNP.

But watch: set our policy for the factories and the cities and the beaches on the basis of the wide open plains of the Midwest or the mountains of New England, and COVID-19 will be quite unforgiving in the penalty it will impose. Which in the end will be quite self-defeating: the economy cannot be resurrected by sending workers home sick or killing them off. [ARL 2020 Apr 17 CDST 18:30]

Disclaimer — Any analysis, comment, conclusions or other content herein is completely of a personal nature intended for the private use of the author. It should not be represented nor used as actuarial or other professional advice. Links to other websites do not connote endorsement of the content of those sources.
not the future expected

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this:
search previous next tag category expand menu location phone mail time cart zoom edit close