“We should act incrementally as prudent risk minimizers and pursue any effective no-regrets options. We do not have to wait for the formulation and acceptance of grand strategies, for the emergence of global consensual understanding, or for the universal adoption of more rational approaches.”
— Vaclav Smil (Global Catastrophes and Trends: the Next Fifty Years)
This post is an attempt at synthesis; those just grazing in (Midwesterners don’t surf) are directed to Reopening — I (Practice) and Reopening — II (Theory) for accounts of my earlier action and contemplation, respectively. For my third installment, I can do no better than lead off with a quadrant diagram of my own devising:
The above came to me after realizing that the so-called NPIs (non-pharmaceutical interventions) so widely recommended to fight COVID-19 were, at least by contrast with a well-targeted vaccine for SARS-CoV-2, relatively non-specific. So I decided to make “specificity” the independent variable, which puts it on the x-axis, and the obvious dependent variable f(x) would be effectiveness, on the y-axis … and recall from Taleb that unless you have the luxury of doing pure research, you need to concentrate on f(x). To a good first approximation, then, the general population should be adopting, or avidly pursuing, the things in quadrants I and II, and ignoring—in some cases simply violating—the contents of quadrants III and IV.
Priorities readily emerge from this analysis …
- A subtle but desperately necessary point is that individual differences must be respected, individual perceptions of risk in particular. “Then lest us stop judging each other; rather, use your judgment to keep from offending your brother or putting a stumbling block in his way.” (Romans 14:13, Lattimore)
- Low-specificity, high-effectiveness tactics should be strongly preferred, for their versatility in managing future incidents (about which more in the next entry in this series). They are not necessarily cheaper from the perspective of an individual consumer or employee, but their yield across decades and generations is unsurpassed.
- High-specificity, high-effectiveness tactics are, by definition, relatively unique, but include the gold standard of a well-targeted vaccine, which is vitally necessary in environments otherwise lacking in institutional functionality and financial margin, which is to say the Third World (and marginalized locales in wealthier nations).
- Everything else. Individual experimentation should of course be allowed, but quadrant III and IV tactics—very much including general-population lockdowns—should not be imposed as a matter of law, except perhaps as a very local response on the scale of a neighborhood or a small town.
- Air purification is quite simply the single best institutional response to this pathogen (and many others). Libertarian though I am, I would seriously consider advocating requiring that all new construction include it and that all existing public accommodations retrofit it within the next few years. Its hierarchy of effectiveness looks like this:
- Dilution ventilation with outdoor air, basically running HVAC blowers continuously—and/or opening windows, wherever possible—for more air changes per hour (ACH); 4.5 ACH gives 90% relative risk reduction, and 6 ACH = 95%. A more strictly quantified recommendation is 7-10 L s⁻¹ pers⁻¹. In the case of the venue I’ve helped, the total volume of air in the auditorium is ≈9 × 10⁵ L, and a reasonable maximum number of people present simultaneously is 100. At 10 L s⁻¹ pers⁻¹, that’s 4 ACH. The total cross-sectional area of the supply ducts is ≈0.5 m², requiring air movement of only 2 m sec⁻¹ (≈ 4.5 mph) to meet the criterion. We are actually getting 6 ACH and thereby 95% protection from this technique alone. Not to overlook the obvious, outdoor events are essentially perfectly safe at anything like normal areal densities of crowds (~1 pers m⁻²), even with almost no air movement. Also, “energy-efficient” buildings with no openable windows are more or less maladaptive.
- Inactivation of pathogens, usually with irradiation or reactive oxygen, possibly with humidification.
- Irradiation, specifically ultraviolet germicidal irradiation (UVGI), means UV-C, generally λ = 200-280nm, with peak effectiveness at λ = 265nm but most commonly in the form of bare (that is, no phosphor) mercury-vapor lamps at λ = 254nm. KrCl excimer lamps with λ = 222nm are becoming available. UVGI is phenomenally effective, sometimes reaching 20 ACH equivalent, but also phenomenally expensive, $~100 m⁻². Hospitals and some other very high-traffic facilities use it, but they have large budgets. The very good news is that this will be largely meliorated, starting as early as next year, with the ramp-up of production of UV-C LED germicidal lamps with λ = 265-280nm. I expect not only lots of these to be mounted inside HVAC ducts but to be widely used in the form of “puck lights” for surface decontamination, especially in kitchens and bathrooms.
- This is just to separately note that UV-B (λ = 280-320nm) is only ~20% as effective as peak UV-C, but it’s a lot better than nothing, and it’s free in the form of sunlight. One more reason to do things outdoors and/or open windows whenever possible.
- Reactive oxygen species-based techniques, in the order of familiarity to me: bipolar ionization, dry hydrogen peroxide (DHP), and photocatalytic oxidation (PCO). Bipolar ionization as installed in the venue where I have assisted is providing an additional ~80% protection, taking our 95% risk reduction from 6 ACH up to 99%; it is also currently about an order of magnitude cheaper than UVGI. DHP has an if-you-have-to-ask-you-can’t-afford it quality, seemingly suitable only for very large, high-traffic buildings. PCO is of imprecisely determined but probably significant effectiveness against viruses; its cost appears to be greater than that of bipolar ionization but less than UVGI or DHP.
- Humidification is of lesser, and variable, effectiveness but I again note that it often comes free with outdoor air, so get outside or open those windows!
- Filtration is being communicated as a simple matter of upgrading to MERV 13 or higher filters to trap more virus. MERV is an acronym for Minimum Efficiency Reporting Value; MERV 13 filters will capture nearly half of flu virions and reduce relative risk (RR) of infection to 64% of its baseline, unmanaged value, even in the absence of any other measures. They will also, however, substantially increase the static pressure load on HVAC systems. Some large fraction of existing installations will be damaged over time by reflexively installing these filters. (HVAC is going to be a very good business to be in for the foreseeable future.) In the effort I helped manage, our decision was to upgrade from MERV 7 to MERV 10, which gives us an RR of 72%—and in combination with always-on HVAC and bipolar ionization, pushes our overall risk below 1% of what it would be with no mitigation at all.
- Vitamin D supplementation is supported by numerous studies finding that likelihood of developing a severe, or even noticeable, case of COVID-19 plummets as blood concentrations of vitamin D rise above 30 ng mL⁻¹.
- Darker-skinned people, elderly shut-ins, and most of the population in wintertime are susceptible to deficiency due to lack of UV-B sunlight, less efficient synthesis in the lower layers of skin epidermis, or both. This is a significant element in the present growth of case counts, with increases in severe cases at the margin and attendant drawdown of medical resources. I remain astonished that supplementation is not receiving greater emphasis by public-health authorities, especially since most blacks and Hispanics in the US are at least borderline deficient.
- Essentially everyone in the population should take 4000 IU (100 mcg) of vitamin D₃ per day, and many people should take 10,000 IU (250 mcg). Toxicity does not begin until far above this level, probably 30-40,000 IU per day, and this vitamin has numerous health benefits. Minority residents of Kansas City are at around triple the risk of death from COVID-19 of the general population, and the age distribution nationwide notoriously skews older, to the point that nearly half of all deaths have been in nursing homes.
- The combination of better ventilation/air purification and vitamin D supplementation alone would almost certainly drive the R₀ value of SARS-CoV-2 below 1 and snuff the disease out in a couple of months. The messaging on this has been as fragmentary and inept as the vaccine development in Project Warp Speed has been focused and successful.
- Masks don’t work well in many of the environments in which they are “required,” but there are a relative handful of situations where they are highly effective, and in those situations, failing to require them is seriously negligent.
- As noted above, outdoor air provides so much protection that with the possible exception of mosh-pit-level density, say 2.5 pers m⁻², combined with lots of talking, shouting, or singing, there’s no point in wearing masks outside.
- Most office environments are low-density, ≤0.1 pers m⁻², and mask-wearing offers only about 90% RR. Grocery stores appear similar; I note that 2-meter spacing = 0.25 pers m⁻², and stores are rarely if ever so crowded (nor are the customers talking with one another in large groups, much less singing or shouting). And their HVAC is certainly running nonstop.
- Properly worn N95 respirators are far more effective, providing 10% RR, but they are relatively uncommon, and very few people are adequately trained in fitting them.
- Restaurants and entertainment venues with seating densities of ~1 pers m⁻² should both address their ventilation and require masks whenever possible, by way of providing constructive alternatives to the disastrous policy of social distancing, which has devastated small businesses and caused a secondary pandemic of mental illness. We should pretty much be willing to wear moon suits to get rid of social distancing.
- The issue for churches is similar, especially for group singing; measurements of hymn-singing with and without masks indicate a reduction of over 90% in aerosol-droplet production, ie RR < 10%. Bluntly, any church not requiring masks is run by very careless people.
- Vaccines—and prioritization thereof—will be getting plenty of attention over the next few months. The success of Operation Warp Speed has been astonishing, and the new technologies utilizing mRNA are among the greatest blessings of our time (and accordingly are hysterically opposed by conspiracy theorists, but I’m saving a SWOT analysis of what project managers call “risk events,” both positive and negative, for my next post in this series).
- Front-line medical workers seem likely to get the first wave of available doses, which is as it should be, but I suspect that many medical people quite far from the front line of this struggle will also be included, and human nature being what it is, various ostensibly important or essential political figures as well.
- The next priority then ought to be the high-risk populations already mentioned: any combination of dark-skinned, elderly, the severely obese, those with histories of recent infections or major surgeries, etc. I am less confident that this will actually be done, and not at all confident that if done, it will be well-received. The list of what might be called Durkheimian sacralized entities in American society is not, to put it mildly, universally agreed upon. Giving the first available doses to, say, EMTs is one thing; moving somebody to near the front of the line because they have a BMI of 40 or compromised liver function from decades of alcohol abuse is something else, even if they are (say) a teacher in an inner-city school district. But this is another one for the SWOT analysis.
Bibliography of sources (other than my own experiences as recounted in the earlier entries in this series) I have found useful in compiling the above: