Our Goal

With the intense media scrutiny and partisan bias on social distancing policies in the US, we believe we should open up the country, but also address the polarizing views from our research standpoint. We want to generate and share reliable and transparent evidence on the impact of the COVID-19 epidemic, and to critically examine the institutional policies made in response to the epidemic around the globe.


We believe the primary goal of any interventions is to ensure the health of the entire public, including those at low risk of death or severe disease. Unfortunately, mandated social distancing does not accomplish this goal; instead, it prioritizes the lives of very few at the expense of many healthy individuals.

For example --
  • Loss of employment and income is associated with excess mortality.
  • 70% of child abuse cases are reported by school teachers; a wide body of evidence exists directly linking adverse childhood experiences (ACEs) to poorer employment, education, and health outcomes.
  • Mental health issues have grown exponentially as measured by call volumes to mental health and suicide hotlines.
  • Domestic violence reports have increased exponentially.
  • Social distancing measures have been implemented without solid, good-quality peer-reviewed evidence from scientific journals. In fact, a 2012 paper implied that such measures must be done swiftly and immediately, or not at all in order to balance the severe economic consequences with halting disease spread.
As a result, our primary goal is to provide research that supports optimizing the health of the entire public as a whole, vs. saving lives in the immediate future. By doing so, we can save hundreds of thousands of more lives in the long run, as well as decrease the very serious adverse effects resulting from mandated lock downs. We must end all lockdowns ASAP to avoid an economic depression far worse than the Great Depression of the 1920s and convince the public that the threat of COVID-19 is far less severe than the panic and hysteria being conveyed in the mainstream media and by many clinicians and politicians. If and only then will the costs of the pandemic not outweigh the economic and public health costs.


If COVID19 is not as deadly as it seems, why is there so much panic and hysteria?

From a psychological standpoint, people have difficulty understanding scale. There is also a fear of the unknown, and fear when not in control. For example, car accidents kill 40k Americans per year, yet millions of Americans drive daily - we do not prevent people from driving; we just penalize high risk drivers (from increasing insurance premiums to suspending driver’s license). Many people fear flying because of airplane accidents, yet fatal car accidents are far more likely - this is driven by false sense of security when in control of the situation. The media preys on these fears for attention with their dashing headlines. Why aren't speed limits universally set to 10 mph if driving is so dangerous?

In addition, COVID-19 is a novel disease.

Most models are not a perfect reflection of reality; as George E.P. Box famously stated: “All models are wrong, but some are useful”: blindly following models without a good understanding of the assumptions can lead to false conclusions. What is the right answer?

First, we don't have a thorough understanding of the assumptions, and a transparent, thorough justification of the parameters used, including assessment of the sensitivity of results to input parameters. Models should always be used as a guide, PARTICULARLY when the assumptions are flawed and there is no data to support these assumptions. This is especially clear in the IHME/Bill Gates model and the Imperial College of London models, both of which grossly overestimated the number of potential deaths resulting from COVID-19 and led to much of the hysteria. Further, these models never once considered the costs of these lockdowns, rending their models useless for public policy.

For more information on perspective - cognitive biases and panicked thinking: article on panicked thinking

What are the flaws in current modeling techniques?

In addition to lack of thorough assumptions in most government health models, using standard SEIR (susceptible, exposed, infected, recovered) models is flawed because those models are highly dependent on reported cases/deaths which are severely based on testing availability and who we are testing (i.e. testing in the early stages was reserved only for symptomatic individuals). Without accurate data on people who were NOT tested, the models don't represent the general population, rather just a subset that was tested.

In fact, SEIR models are only useful AFTER the fact, when an epidemic has ended. Using SEIR models to project anything useful about the COVID-19 pandemic is useless at best, and harmful at worst - and this is exactly what has played out in the public policy arena.

It is far more important to model trends (daily changes in rates) vs. absolute levels because this provides greater clarity into what is actually going on. Doing so also nets out time-invariant biases; although we have since seen that reporting biases are NOT time-invariant as we had originally thought. Time-invariance is important because it implies that certain biases remain constant over time (e.g. Country X consistently reports cases and deaths one way that differs from Country Y, but this reporting bias is consistent over time for that country).

Unfortunately, on 3/24, the CDC changed its coding rules , thus undermining the assumption of time-invariance. Nonetheless, modeling daily changes is still far more accurate than modeling reported levels, for the reasons described above.

What causes miscoding/inflation of deaths?

In addition to the inaccurate number of tests per amount of people who have the virus, the case numbers are likely to be inflated, especially in regards to deaths.

Due to the CARES ACT, there is flexibility and incentive to code a death as COVID-19 based on better reimbursement and loose CDC coding guidelines. In cases where a definite diagnosis of COVID-19 cannot be made, but is suspected or likely (e.g., the circumstances are compelling within a reasonable degree of certainty), it is acceptable to report COVID-19 on a death certificate as “probable” or “presumed.”

Furthermore, CMS is paying 20% on top of typical reimbursement for all COVID-19 related cases, creating an incentive to code all medical care as COVID-19. And that reimbursement is even greater for patients who require ventilator use (which we have since realized actually kills more patients than it helps). As a result, it is probable that hospitals are coding most deaths as COVID19 without a definite diagnosis.

For more information, please visit official policies outlined by the CDC:

Modeling (see above re: focusing on rates of change and time invariance) [TECHNICAL]

We have chosen to model the epidemic in a different way from other research groups, which are employing standard SEIR models. However, because SEIR models rely on absolute reported levels, and these levels change daily, it is impossible for them to develop an accurate outlook for the future.

We employ an hv-block cross-validation methodology, using Gompertz and logistic models to track mortality and case loads across countries, under the justifiable assumption that the logistic S-curve describes biological growth phenomena, including viral spread.

  • The Gompertz distribution is a specialized form of the logistic S-curve that restricts all estimates to be >0
  • Hv-block K-Fold cross validation is a machine learning technique that allows us to identify key coefficients necessary to estimate the number of deaths by country, using other countries that are further along in the epidemic (e.g. China, Korea, Iran, Italy) to trend the change in number of deaths over time
  • Relatedness of the incidence decay with exponential adjustment (IDEA) model,“Farr's law”and SIR compartmental difference equation models
Other sources of data include datasets retrieved from government databases and correspondence with other researchers across the country:

Population Health

Sweden’s situation

When we look at other countries with lax policies, Sweden and Japan are notable examples that are doing well. Sweden hasn't had any SICs except for gatherings of over 50 people. Yet, their death rate is lower than Spain, Italy, UK, France, Netherlands, Belgium, and many other European countries. Moreover, Swedes now have herd immunity, so they are confident that they can remain open for business indefinitely while other countries struggle with whether they have to re-SIC for the “second wave”!

With regards to herd immunity, their goal is relatively feasible that Anders Tegnell, current state epidemiologist of Sweden, believes “We are starting to see so many immune people in the population in Stockholm that it is starting to have an effect on the spread of the infection”. He also said that he believed that stricter lock downs "only serve to flatten the curve and flattening the curve doesn't mean that cases disappear -- they are just moved in time." "And as long as the healthcare system reasonably can cope with and give good care to the ones that need care, it's not clear that having the cases later in time is better."

Sweden does have a higher death rate per million than the US; this easily explained due to the more elderly population in Sweden (i.e. higher % of population >85 years old). Because the Swedes age better and longer than Americans, they are more susceptible to dying from this virus!

Japan’s situation

Japan is extremely similar to Sweden, even with measures less extreme than the US; despite less testing and less strict measures, cases are falling. It is estimated that cases are much higher than reported, implying that the death rate is much lower than reported. This is also in contrast to South Korea and Taiwan who have strict policies but the same level of movement within the country.

For more information on Sweden & Japan, please visit:

Social Distancing/Contact Tracing/Post SIC Life?

Given the conversation revolving around antibody tests and immediate contact tracing/removal of those infected, there are several flaws in the reasoning. First, Infection Fatality Rates (IFRs) are totally flawed, because NO ONE knows the denominator. There are no accurate tests, either serology or PCR. So if the serology/PCR is negative, clearly that doesn’t mean that the subject hasn’t been exposed. It doesn’t mean that they have. And we don’t know from testing how infectious someone is. So testing is totally misleading for managing this epidemic.

Furthermore, you have to look at broad population totals: Sweden, India, Korea, US prisoners, etc. all prove that the denominator is probably the entire population. That makes C19 much less lethal than flu, if you apply a flu IFR of 0.04%. Again, what does flu infection fatality rate mean? Some people have no symptoms, some have minor symptoms, some have moderate symptoms, etc. etc. and some die.

Economic Impact

With regards to economic impact, According to McKinsey data, the economy, particularly these sectors, will suffer a contraction that hasn’t been seen since the pre WW2 - post WW2 era. The economic consequences are drastic and as stated above, far outweigh the costs.

The five hardest hit sectors include Commercial Aerospace, Air & Travel, Oil & Gas, Insurance Carriers, and Automotive with each suffering a contraction from 20%-40%. These industries are responsible for the jobs and livelihoods of millions of Americans and it is imperative they get back to work. This is not a shareholder/big company problem, rather everyone down to the average citizen is impacted.

The economic downturn is only bad for wall street and big companies, or also for the average American?

Based on our research, the mortality rates of those impacted far outweigh the mortality rates caused by COVID19, especially considering the unknown circulation of the virus in many populations already. For example, about >30 million Americans have filed unemployment claims in the last five weeks. Economists project that US GDP data, scheduled for release on April 29, will indicate a 3.5% decline (annualized) in the first quarter of 2020. [Bain & Co]

There is direct positive correlation between the drop in the economy and an increase in mortality rates, specifically in populations of low socioeconomic status. “Epidemiologic studies using a variety of SES (Socioeconomic Status) measures have consistently shown that, in the general population, mortality risk increases as SES decreases: "The economic shutdown is putting more members of this population under more duress and will cause greater harm than COVID19 itself would. "

These findings are in line with those of previous research by Rodgers (1979), who stressed the positive association between high levels of socioeconomic inequality and elevated mortality. The pandemic is exacerbating these socioeconomic inequalities through long term unemployment and loss of livelihood. As a result, these crises are often accompanied by rising income inequality, a deterioration in the social safety net, and material hardship; which in turn lead to increasing levels of psychosocial stress, and temporary periods of stagnating and even declining life expectancy.

How do we balance the economic and value of human life lost during the pandemic?

The US government values lives at ~$10M per life (see FAA and EPA policies, this is a validated dollar value) - the economic devastation puts the value of lives lost significantly higher than this. The VSL (value of a statistical life) is also heavily backed by a vast economic literature (see Viscusi et al; Kneisner et al).

The CARES relief package represented $2 Trillion in relief funds; divided by $10M = 200,000 lives; given that the number of US deaths is significantly lower than this, the stimulus package highly overestimates the value of each life at a far higher price than the US government’s own stated policies.

For more information on economic impact/socioeconomic status:


There is absolutely no reason for the economy to have locked down. We therefore believe everything should be opened immediately, with the following precautions:
  • Maintain infection precaution measures for high-risk individuals (e.g. nursing home residents, immunocompromised, elderly with comorbidities), including appropriate PPE, handwashing, and avoidance of large groups.
  • Reopen all schools, colleges and universities as quickly as possible. As of May 1, 2020, only 9 children under the age of 15 and only 50 under the age of 25 were reported to have died in the US. We suggest that effective immediately, all schools and daycares open for children. In turn, this will allow working parents the opportunity to return to work.
  • Allow any and all businesses to open completely if they are willing to do so and decide what social distancing measures they would like to implement.
Other suggestions have been described:

Other Common Questions:

Why go against Dr. Fauci?

The leading epidemiologist on the corona virus task force has often been wrong in his previous works. During the HIV/AIDS epidemic about transmission and spread he believed that “First, it is possible that AIDS can be vertically transmitted. Perhaps even more important is the possibility that routine close contact, as within a family household, can spread the disease.” Without undermining Fauci’s expertise and contribution to epidemiology, however, his stance on the COVID19 pandemic contradicts empirical research we have seen in other countries and other situations. The models that the US is relying on right now are overblown and inaccurate as shown by our models.

Social Distancing = Less Hospital Overload? How do we prevent hospital overload?

One of the main justifications for the global lock downs was to prevent a crush of patients overwhelming hospital intensive care units. But as with everything C19 Sweden has proved that shutting down economies and locking down the young and healthy is completely unnecessary to avoid ICU overcrowding. Despite no lock downs or other social isolation controls except prohibitions on gatherings over 50, the Swedish hospital system never experienced anything remotely like the crush of ICU patients in Italy, Spain or New York City. Sweden’s ICU C19 patient census (updated nationwide daily) peaked in early April with about 50 new admissions daily. Now it is gradually declining to about 35 new ICU cases daily.

For more information Fauci/AIDS response:


Joel W. Hay, PhD
1362 US HWY 395, Ste 102
Gardnerville, NV 89410
United States of America