One key overarching assumption that is easily forgotten is that the spread of the disease is inevitable until herd immunity or a vaccine is reached. It is important to understand that this is an assumption, not a conclusion or prediction. It is literally built into the model. Its corresponding but inverse cousin is the assumption that once you have recovered from the disease, you can neither get it again nor spread it to others, at least for some period of time. Without that assumption, there is no herd immunity and vaccines may be of little use. Both of these assumptions are reasonable. However, that is not the same as guaranteed. Throughout human history since long before vaccines were created and even since that time in some cases, disease outbreaks of varying scales have come to an end before herd immunity was reached in the broader population by segregating the hot zones strictly until the disease ran its course within that area. There is a possibility, still uncertain, that China may be experiencing that result through its extraordinarily severe restrictions in Wuhan and the surrounding area. Notwithstanding concerns about the reliability of the data released by Chinese authorities, if they had experienced infections and deaths at the per capita rates experienced by Europe and the U.S., it is hard to imagine them having successfully hidden that magnitude of impact. The results observed in China, even if adjusted for bias, point either to a termination of the current outbreak through extreme measures, or to a special case in the standard SIR models in which especially aggressive measures temporarily reduce R 0 to near zero giving the appearance of eradication, only to have a follow on spike and spread until herd immunity is reached like other areas. It is too early to tell which is true, although watching the rate of infections in China will continue to be highly instructive for the next couple of months. If the assumption of eventual immunity, at least partial or time-limited, proves to be false, which is also true for some diseases even in the modern era, then public and personal health responses are forced to transition on an indefinite basis to improvements in treatment and protective measures. This does not mean never leaving the house again. It does, however, mean that protective measures will need to evolve to a level that the average person has reason to believe they can proceed with daily life at a risk level for COVID-19 that is commensurate with other accepted risk factors. There is one other model that has been well publicized, the IHME model out of Seattle, that appears to use a different methodology. While the specifics of the methodology have not been publicized, it is described as a hybrid approach using a “mixed effects non-linear regression framework.” Essentially, it appears that the model is designed to attempt to fit actual statistical trajectories from countries that experienced the virus earlier to the initial experience in the U.S. and the changes in that experience over time. An advantage of such an approach is the reliance at its base on reported data that can be tracked rather than assumptions regarding inputs and relationships. However, the IHME model remains vulnerable to the accuracy and time lag problems of the underlying statistics from which it is derived. In particular, reliance on reported data with fewer assumptions about the underlying drivers of observed interactions can miss unique and potentially important characteristics of different areas that could lead to differing predictive results if properly considered.
∴ PROGNOSIS
27
Made with FlippingBook - Online catalogs