How can we see into the future? The simple answer is that we can’t. We are left only with guesses as to what may be. Some of these are educated and informed; others are wild speculation. It is the former that I will focus on today; informed predictions of what may happen with the global pandemic of COVID-19 in the coming weeks and months. At present, we have no vaccines for SARS-CoV-2 and no evidence-based therapeutics for COVID-19. What we do have are public health measures, such as “social distancing,” for prevention and medical interventions consisting of supportive care. Leveraging our public health tools, however, buys time for the development and testing of vaccines and therapeutics.
A very thoughtful and thorough projection is available, based on sophisticated modelling by Kissler and colleagues.1 Their efforts take into account the possible contributions of seasonality, duration of immunity, and cross-protection imparted by prior infection with the two other betacoronaviruses in common circulation (HKU1 and OC43). Then, they provide a variety of scenarios that simultaneously assess the effects of the length (4 weeks to indefinite) and strength (0–60% reductions in Ro) of social distancing. First some basics:
- Seasonal coronaviruses are seasonal; they circulated primarily in the late autumn, winter and early spring months in temperate regions;
- Immunity to HKU1 and OC43 wanes fairly rapidly, over the course of about a year;
- Some cross-protection exists between these two viruses, and perhaps, extends to SARS-CoV-2;
- Both of these seasonal coronaviruses are less infectious than SARS-CoV-2.
The modeling efforts lead to some interesting and some uncomfortable conclusions:
- SARS-CoV-2 can proliferate at any time of the year (as we are seeing now across the globe);
- If immunity is not permanent, SARS-CoV-2 will eventually enter into regular circulation as our fifth seasonal coronavirus;
- If immunity is permanent (or very long-lasting), SARS-CoV-2 could disappear after a few years;
- High levels of seasonality will lead to a smaller initial peak, but larger wintertime outbreaks;
- Low levels of cross-protection from OC43 and HKU1 might allow resurgence of SARS-CoV-2 following a period of low activity lasting a few years.
When social distancing is added without seasonality, the following scenarios emerge:
- Short durations of social distancing do little more than displace the cases into the near future;
- Longer durations of higher-intensity social distancing effectively reduce case burden in the near term, but result in significant outbreaks during autumn and winter;
- Permanent social distancing of moderate to high intensity works well to keep SARS-CoV-2 at bay (but would be unpalatable to almost all of us).
Finally, and what I think are the most likely projections, are of those with social distancing added into a seasonal world:
- Short durations of social distancing slightly delay the peaks of COVID-19, but result in high overall infection rates;
- Longer durations of social distancing push the peaks into the winter months and increase the overall infection rate;
- Intermittent social distancing, based on good surveillance, may be needed to keep case load in check until vaccines are available or a sufficient percentage of the population has been infected, become immune, and herd effects take over.
I suspect our best choices here require a Faustian deal, buying time now at the expense of a future catastrophe, in the hope that effective therapeutics and vaccines become available and that critical data emerge regarding the extent of population immunity, duration of immunity, and its rate of decline. Perhaps the last and best words for COVID-19 predictions are those of Ebenezer Scrooge, “Are these the shadows of the things that Will be, or are they shadows of things that May be?”2 Only time will tell.