Friday, March 27, 2020

CDC March 21, 2020 Update: 24,000 Influenza Deaths Including 155 Children This Season

From Centers for Disease Control and Prevention, Weekly U.S. Influenza Surveillance Report, "Key Updates for Week 12, ending March 21, 2020:"

Key Points
    ***
  • Pneumonia and influenza mortality levels have been low, but 155 influenza-associated deaths in children have been reported so far this season. This number is higher than recorded at the same time in every season since reporting began in 2004-05, except for the 2009 pandemic.
  • CDC estimates that so far this season there have been at least 39 million flu illnesses, 400,000 hospitalizations and 24,000 deaths from flu.

At the time of publishing this post, there were 100,392 Coronavirus cases and 1,543 deaths in the US. Let's hope deaths from the Coronavirus wil not reach the level of deaths from the Flu.

For the current numbers of Coronavirus cases and deaths:
https://www.worldometers.info/coronavirus/country/us/

Tuesday, March 24, 2020

Unanswered Coronavirus Question

The coronavirus may be a substitution of one cause of death of older, severely ill people for another cause on a death certificate. Stopping the virus might not decrease the mortality rate of the most likely affected individuals.

So far, most of the people dying from the coronavirus are older, have multiple serious health problems and more likely male. On average, an 85 year old male has almost a 10 percent one year mortality rate, a 10 percent chance of dying in one year when not adjusted for the health of the individual, A 90 year old male has a 16 percent chance of dying in one year. Individuals who have multiple serious health problems have much higher mortality rates than the above average actuarial numbers.

The important unanswered question about the coronavirus is whether or not the virus is increasing the mortality rate of severely ill older people? Whether the virus is causing excess deaths in epidemiology terms? If someone is so ill that they would have likely died from heart failure, flu, pneumonia or some other non Covid-19 cause in the next few weeks, dying from the coronavirus and listing it as the cause of death is misleading. It gives the public the false impression that but for the coronavirus the individual would have an extended life and not have died from other causes around the same time.

If stopping the virus does not increase the expected life of infected ill individuals, is it worth shutting down the economy?

Addendum
As more insurance companies modify their existing health insurance policy coverage to fully pay for coronavirus outpatient and in-hospital testing and treatment, doctors will have incentives to list more coronavirus diagnosis as the primary cause of death on medical records and death certificates in place of other causes. The number of deaths due to coronavirus will certainly be inflated by some amount and we should expect to see a sharp increase in proportional deaths around the time of expanded health care insurance coverage.

Wednesday, March 18, 2020

Are We Misreading COVID-19 Statistics? Where Is A Statistical Epidemiologist When You Really Need One? Total Deaths Vs Excess Deaths: Coincident Vs Causal: Contagion Rate Vs Morbidity And Mortality: Aged Adjusted Death Rate

Since death is a natural phenomenon of life, fear of death from a disease, such as the Coronavirus COVID-19, is rational and a major health concern requiring emergency measures, if the virus does the following two things:
  1. Does the virus increase the number of deaths in an age group above the normal mortality rate? In epidemiology terminology: Does the Coronavirus cause Excess Deaths?

  2. Does the virus cause death from various health factors to occur sooner, i.e. shorten life expectancy of an ill person? In epidemiology terminology: Does the Coronavirus cause Premature Deaths?

Excess Deaths
The average US death rate from the known cases of Coronavirus is currently around 1.5 percent, (127 deaths in 8055 Coronavirus cases at the time of publishing this blog post). With many undiagnosed and recovered US Coronavirus cases still unknown, the death rate is likely very much lower than the current 1.5 percent average rate.

Looking at the Social Security Administration Actuarial Life Table for 2016, which is not adjusted for the health of individuals, males have to be 65 years old or older to have a one year mortality rate greater than the COVID-19 average mortality rate of 1.5 percent. Females have to be 70 years old or older to have a one year mortality rate greater than the COVID-19 average mortality rate of 1.5 percent. A 70 year old male has a 2.3 percent one year mortality rate.

At 85 years old, the one year mortality rate is 9.6 and 7.3 percents, respectively. By 90 years old, the one year mortality rates jump to 16 and 13 percents, respectively.

Most of the COVID-19 deaths are occurring in older people who have serious pre-existing health conditions. Their mortality rates are higher, probably much higher, due to their illness than the general population average Social Security actuarial mortality rates.

Absent more detailed contrary statistics from a Statistical Epidemiologist that includes disease mortality rates by age, illness and COVID-19 infection presence, the average Coronavirus data is not showing an excess US population death rate from the virus.

Coincidence Coincident Vs Causality
It appears that the news and our politicians are confusing coincidence coincident with causality.

Imagine a hospital with patients in two wings, Wing A and Wing B, with each section containing severely ill patients with a 50 percent chance of dying within a month. Suppose by chance, the patients in one of the wings, Wing B, get infected with a virus. The staff diagnoses the infection and reports it to health authorities. A few weeks later, the hospital reports to authorities that half the patients in Wing B with the virus died and the others in that wing did not die or get worse.

The news headline would read, "50 percent of hospital patients with the virus died." Without a comparison to the identical, but virus free population, the 50 percent death rate is meaningless. If 50 percent of the patients in the other wing who were not infected with the virus also died, then there are no excess deaths due to the virus. The diagnoses of Wing B patients having the virus infection is a coincident factor. It it something that happened at the same time, but it is not a causal factor, a cause, of any deaths in Wing B.

Coincidence Coincident, and not causality, seems to be the state of our current COVID-19 virus data. We know the virus was present. We do not know it caused the deaths.

Premature Deaths
Looking at the SSA Actuarial Table, one can see that about 10 percent of all 85 year old males and 7 percent of all 85 year old females will die within a year. For a 90 year old, the one year death rate jumps to 16 and 13 percent.

Do we have any reliable data showing that COVID-19 infection has decreased very old senior citizen life expectancy with their already high mortality rates? Probably not.

Since many of the seniors who have died, purportedly from Coronavirus, have been seriously ill, if there were any shortening of remaining life, it may have been, if it occurred at all, for a short time of a few days or a few weeks.

Contagion Vs Morbidity VS Mortality
A high contagion rate is often mentioned about the Coronavirus COVID-19. A highly infectious disease, a high contagion rate disease, means that the disease is easily transmitted from person to person and easily spreads throughout a population. The fact that a disease is easily transmitted, does not indicate that a disease is dangerous nor does it say anything about the effect the disease will have on an individual. The virus infection could have low morbidity. The infection could be innocuous and even unnoticeable with no short or long term effects.

The fact that the Coronavirus is Contagious and has some Morbidity, symptoms, is not an indication that the disease affects Mortality rates. The common cold is highly infectious. It can cause discomfort and generally has a low mortality rate, except for those who have a very high risk for infection, such as immunosuppressed individuals or those who have certain pre-existing health conditions.

Aged Adjusted Death Rates
Comparisons are often made to other countries that are infected with the virus. The percentage numbers will be misleading unless one adjusts for the different percentages in each age group in each country. On the surface, the virus seems to affect different age segments differently. To be able to accurately compare the US to other areas of the world, the numbers for each country's infection and death rate need to be adjusted for different percentages of each population age segment in each country. The process will result in Age Adjusted Death Rates that can be used to accurately compare country death rates.

In a few years, after the studies appear, it will interesting to see if all the extra precautions put into place were warranted or an extreme over-reaction,. We will eventually find out whether the economic harm to the economy and individuals through business and government closings and self-isolation exceeded any harm that may have come form the COVID-19 Coronavirus.