Just when it looked like we’d turned a corner, COVID has returned with a vengeance – or so it seems. Most of the recent media attention has focused on rising infection rates in Florida, Texas, Arizona, and California. Three of those four (FL TX, CA) top the list of most populous states, so their raw case numbers should be no surprise. The fourth most populous state is New York, which second only to New Jersey, leads not just the US but the entire world in accumulated per capita infections and mortality.
What’s going on?
According to the New York Times, new cases are on the rise in 38 states and the Virgin Islands. They are mostly steady in 11 states and Guam, including the hardest hit states of NY, NJ, MA, and CT. Cases are declining in only 3: NH, ME, and VT. (All are based on recent seven-day averages, so are sensitive to daily fluctuation and can be misleading). These “record” case counts should be taken with a grain of salt. Testing was so scarce in the early stages we have no idea what the actual counts should have been in March and April. Increasing hospital admissions confirm that the disease is truly on the upswing in a few populous regions, though probably not as dramatically as the simple case counts might suggest.
In the context of the present resurgence, this is a good time to assess where things stand and how our previous calls have stood up. A number of the following points have already been made on this site, but are re-introduced both for new readers and as a reminder to others.
- New York mostly failed to flatten the curve. It saw a massive spike in infections and deaths, followed by a continuous decline to now very low levels. New cases, hospitalizations, and deaths in New York City have dropped to almost nothing. This follows the pattern seen in the worst-affected nations, including Belgium, United Kingdom, Sweden, Italy, Spain, and France. NJ flattened it only slightly. Both NY and NJ suffered per-capita fatality rates double that of Belgium and triple that of Sweden. Here we see the infection curve for New York:
- Some states did flatten the curve, seeing moderate infection rates spread out of many weeks. (IL, IN, MD). Most of them now have steadily declining fatality rates. Illinois, for example:
- Many states totally squashed the curve. Instead of spreading out the cases over time, their shutdowns had the unfortunate consequence of merely pushing the curve, hence the crisis, forward in time. Until quite recently, FL, TX, AZ, and CA had infection rates far, far below the national averages. Presently, their per capita fatality rates remain less than 1/2 the national average, and around 1/10 the death rates of NY and NJ. This was predicted by Kissler, et al, from the Harvard Department of Epidemiology, whose recent paper in Science warned:
“Strong, temporary social distancing can lead to especially large resurgences”Projecting the transmission dynamics of SARS-CoV-2 through the postpandemic period, Science, 22 May 2020: Vol. 368, Issue 6493, pp. 860-868
“Under all scenarios, there was a resurgence of infection when the simulated social distancing measures were lifted. However, longer and more stringent temporary social distancing did not always correlate with greater reductions in pandemic peak size.”
- The virus is incredibly infectious and easily spread by asymptomatic victims. A recently discovered variant that has become dominant is even more infectious, though not more deadly. (The original, less infectious strain that circulated through Asia may partially account for their early success in suppression). There was, and is, zero possibility of containment. From the very outset, Harvard epidemiologist Mark Lipsitch predicted that 40 to 70 percent of the global population would be infected [possibly much lower; read on]. Masks and social distancing can slow the spread. In so doing, they accomplish the original purpose of “flattening the curve”: to prevent hospitals from being overwhelmed. Unfortunately, many got the message that through such efforts we could stop the pandemic in its tracks. That became impossible about six months ago. The overwhelming percentage of cases are now known to be mild or asymptomatic, something we have been reporting from the outset.
- The infection fatality rate is confirmed at a mean value of around 0.4%, or about four times worse than the common flu. This is in the range initially predicted by internationally renowned researcher John Ioannidis in mid-March. By May 4, the evidence had become sufficiently compelling that we made that call on this blog. This is now the official position of the CDC. [Note: On July 13, 2020 this was revised upward to 0.65%] The local death rate may be higher (Bergamo, Italy) or lower (Singapore) depending upon the age distribution of its victims.
- The data increasingly suggests that the Swedish approach (prohibit large gatherings, no closures or lockdown) has been equally effective with far less social and economic disruption. Swedish COVID fatalities have declined precipitously since a mid-April peak. (Their death rate could have and should have been much lower, but for a policy of malign neglect toward the elderly having nothing to do with the decision not to shut down). As of today, the per capita fatality rate in Sweden remains lower than nine US states and the District of Columbia – all of which locked down.
- “Herd immunity” may kick in at much lower levels than popular wisdom asserts, and is by far the best explanation for the dramatically declining rates of the nations and states listed in #1. The widely circulated estimate that herd immunity requires a 60-70% infection rate is based on a simple formula with simplistic assumptions. The weakest of these assumptions is that all people are equally susceptible. Now, there are four widely circulating coronaviruses that cause the common cold. A series of reports in the last two months have reported evidence for pre-existing T-cell immunity to the SARS-Cov-2 virus in unexposed individuals, linked to at least two of the more benign coronaviruses. The implication is that a large proportion of the population already has partial immunity that would never show up on antibody tests. One international team of researchers, including representatives of Oxford and the NIH, projects that herd immunity could occur at levels as low as 20% after taking into account variations in individual susceptibility.
One highly significant observation is that several weeks after reopening in most of Western Europe, all-cause mortality has declined back to baseline (i.e., the death rate is exactly what it ought to be for this time of year).
[for a further explanation, check out A New Understanding of Herd Immunity, The Atlantic, July 13, 2020]
- A vaccine is not inevitable. Every report, commentator, and column that speaks of a vaccine in terms of “when” or “until” is either naïve or disingenuous. Dr. Francis Collins, head of the NIH, stated in an interview that a vaccine has never been developed in less than three years. After forty years, there is still no vaccine against HIV, and that is clearly not from lack of trying.
It would be inaccurate to declare our containment measures (masks, social distancing) are futile. It is mathematically possible that fewer will get sick or die because of these, yet the long-term net reductions may be minimal. It also wouldn’t be fair to trivialize the benefit of dying six months later rather than now. Risk is an unavoidable fact of human existence. We’d save many more lives by the abolition of cigarettes, mandatory flu vaccines, confiscating all firearms, and raising the driving age to 21 – but there’s no popular support for such measures.
Much criticism has been targeted toward young people going to the beach and having parties rather than conforming to rigid social distancing, yet there is little evidence for any long-term benefit. I have tremendous sympathy for young people forever missing out on once-in-a-lifetime opportunities like high school graduation, or in the case of my own son, live stage performances with summer theater. We are all paying a price, but the young and poor are paying the most. Every policymaker must be called to answer one simple, overriding question: what is the endgame, here?