As the CCP Virus (Coronavirus) continues to infect the world, infection and death rates are going up at an exponential rate. As of writing this, the United States has seen over 7,400 deaths in nearly two weeks. However, as Americans continue to “shelter in place” the future remains unclear. When will this be over, and how damaging will the virus truly be?
To answer the question about how bad this virus will be, we are looking at estimates. The current estimates for the total number of deaths are between 100 ,000 – 240,000. However, many people have been questioning the various models and projections coming out of CDC. For example, Mike Cernovich has rightly been pressing the issue of model accuracy while maintaining that this virus is dangerous.
Additionally Neil Ferguson, an Epidemiologist from Imperial College London, has come under fire for “revising” his estimates about the projected coronavirus deaths. Though Ferguson maintains that his revisions were still inline with his previous estimates, assuming countries (UK) took extreme mitigation efforts.
Experts, business people, techies, and more are beginning to ask hard and serious questions about the CCP Virus models. These models are currently holding hundreds of millions of people hostage, and we have a right know what’s going on, but how much trust should place in these models?
All the talk of models and assumptions made me curious about the flu data: exactly how accurate are annual flu infections and deaths? This year many people wrote off the potential damage from the CCP Virus based off “60,000 people died last year from the flu, it’s no big deal.” Considering the impact coronavirus is having in America, and that the death toll in the U.S. is officially 7,400, did 60,000 people really die of the flu last year?
According to the CDC’s website , there were an ESTIMATED 34,157 deaths from the flu between 2018 – 2019. There is a range of death estimates from 26,000 – 52,000. I capitalized “estimated”, because these are not recorded deaths. In fact, the recorded deaths from the flu are likely considerably lower, since the according to the CDC, they use a multiplier to account for non-documented deaths that “should” be attributed to the flu. Before going into death rates, lets dig in to how the flu hospitalizations and illnesses are tabulated.
Below is how the CDC calculates hospitalizations. The CDC “assumes under-detection of influenza” and therefore must make an “adjustment” to find the accurate numbers. The under-detection has various reasons, one of which is the “average sensitivity of influenza tests.” We will come back to this later. Though it should be noted: influenza tests may not be sensitive enough to pick up all influenza infections.
In addition to test in-sensitivities and limited geographical flu testing, actual flu data for a given year can lag up to two years. Therefore the CDC uses data from earlier seasons to make estimates for current seasons. While I cannot debate the efficacy of the CDC flu estimates by directly looking at their models, these “assumptions” , “estimates”, and lack of “test sensitivity” make me very suspicious about the accuracy of CDC flu numbers.
Lets move on to how the CDC estimates flu illnesses in the America.
The CDC extrapolates total illness based on hospitalizations and how many people sought “medical care for an influenza-like illness.” based on a 2010 Behavioral Risk Factor Surveillance Survey. What on earth is an “influenza-like illness?” Is it wise to base flu illness off of a survey? Surveys require people to use their memories to answer questions, we know memories are not good guides to providing hard and reliable data. Additionally, people can experience fevers, coughs, congestion, headaches, etc. for many illnesses besides the flu, this survey seems weak, and should be investigated.
When looking at flu deaths, the CDC uses a model to estimate a ratio of deaths to hospitalizations(which is already estimated), After making “adjustments” they look at death certificates and estimate flu “related” deaths because they don’t all occur in the hospital. The CDC may also include data from people who died of pneumonia and other respiratory and circulatory causes. Once again, these estimates seem a bit suspect in their accuracy.
Lets see why the CDC doesn’t just count death certificates of people who died from influenza.
The CDC may count people who died of pneumonia, congestive heart failure, and chronic obstructive pulmonary disease in their flu death estimates. Many deaths in America are being estimated as flu-related, yet these estimates cannot be verified . The CDC makes these estimates because they are confident the flu is under-reported. How does the CDC know the flu is under-reported? Because the “virus is only detectable for a limited number of days after infection.” , thus many people who actually have the flu may come up negative for the flu infection. This is why the CDC discusses “test sensitivity” in one of their many reasons for using various “estimates” and “adjustments” to calculate flu illness and hospitalization.
Finally we can look at admitted limitations of “Influenza Burden Estimates”, according to CDC.
First off, the CDC acknowledges that hospitalizations are based on reporting, and hospitalization reports are refined over-time as more data is collected.
Second, in-hospital flu deaths are adjusted based on flu testing rates and the test sensitivities( which are suspect). However, during current seasons CDC isn’t aware of “testing practices”, therefore they use data from prior seasons, then CDC “updates” their estimates when more data arrives about “testing practices”.
Third, CDC estimates of illness, medical visits, and hospitalizations may not be accurate ” if patterns of care-seeking have changed.” But I’m sure the CDC is constantly searching the United States to examine peoples behavior, so nothing to worry about.
Fourth, CDC deaths for current seasons may be estimated from prior seasons if current data isn’t available. And their models use “frequency of influenza-associated deaths that have cause of death related to pneumonia or influenza (P&I), other respiratory or cardiovascular (other R&C), or other non-respiratory, non-cardiovascular (non-R&C) to account for deaths occurring outside of a hospital by cause of death.”
Fifth, currently CDC uses statistical methods to calculate the influenza burden that cannot be compared to years before 2009. So you cannot compare current flu illness, visits, etc… to years prior to 2009, because the models have changed.
At this point the limitations of flu data coming from the CDC should be obvious. There are many assumptions about how flu illness, medical visits, hospitalizations and death are calculated, these assumption should be constantly reviewed and updated, are they being constantly reviewed? The underlying assumptions about the flu are extremely important today, as many people used flu statistics to dismiss the danger of the CCP Virus. In addition to dismissing CCP Virus dangers, many people are currently confused about how deaths from the CCP Virus are calculated, and skeptics believe we are over counting deaths because some patience already had underlying conditions(possibly severe). However as noted above, many people with heart and lung problems, who have never tested positive for the flu, are counted as flu-related deaths every year. Therefore while questioning the validity of data and modeling about the CCP Virus, we should also take a look back at the flu, and maybe even CDC.