Coronavirus Pandemic

Posts relating to the current pandemic

Unpacking Mortality from COVID-19

The severity of any disease, especially a new pathogen like SARS-COV-2 is important to measure, especially it’s ability to cause death, which is the ultimate measure of severity. There are two ways to measure mortality from any infection.

Fatility Infection Ratio (IFR)

The proportion of deaths among all infected individuals. To measure IFR one has to know accurately the total number of infections as well as all deaths caused by, the disease. In the midst of a pandemic, with testing variably available and deaths often occurring at home it is impossible to accurately measure IFR.

Case Fatality Rate (CFR)

The proportion of deaths among identified cases. In the early stages of the pandemic, most cases are identified by surveillance and often only the most severe cases are tested. This leads to wide variation in estimates of CFR ranging from 0.1% to as much as 25%.

In fact, it is only possible to accurately measure either one of these fatality rates in retrospect, long after the initial stages of the epidemic. The number of deaths attributed to COVID-19 is almost certainly an underestimate. On the other hand, the number of people who have been infected is also certainly an underestimate. People with mild or asymptomatic infection are unlikely to get tested.

Another problem is that fatality rates from COVID19 are not uniform. Certain groups of people have an increased risk of mortality from COVID-19, so mortality is not uniform across people who are infected.

Excess Mortality

One way to deal with the first problem is to look at excess total mortality rates compared to historical mortality rates. It is very likely that excess mortality during the pandemic reflects the impact on mortality of COVID-19. Even if all of these deaths are not directly attributable to COVID-19, some may reflect unavailability of care at hospitals overwhelmed by COVID patients.

Excess mortality statistics are available through the first 30 months of 2020, which takes us through July 25 of this year. Because of the delay in reporting of death certificates, data for August and September are incomplete. For the US as a whole, there were 207,000 excess deaths for the first 30 months of 2020. This figure suggests that we passed 200,000 deaths from COVID-19 by the end of July whereas the number of reported deaths from COVID-19 at the end of July was 150,000. This was clearly an underestimate. It is also clear from these excess mortality numbers that the mortality from COVID-19 is much higher than from influenza.

Here is a graph of weekly recorded deaths from all causes for the first 30 weeks of 2020. The dark line represent excess mortality for 2020. The gray lines underneath are death rates from the previous 5 years. The spike in the beginning of the top gray line represents the H1N1 influenza epidemic. You can clearly see that even this spike is dwarfed by the excess mortality for the first 30 months of 2020. Here is a link to the website which has these data

Mortality by race/ethnicity

Another way to look at mortality data is to look at mortality by race and ethnicity. The mortality rate for african americans from COVID-19 is twice as high as for non hispanic whites. For native americans the death rate is 1.4 times as high and for hispanics the death rate is 1.1 times as high. These increased death rates by race and ethnicity have nothing to do with genetics. People of color have all sorts of socioeconomic factors that increase their risk of underlying conditions as well as living in crowded housing that make social distancing difficult or impossible. Many have low wage jobs that increase risk of contact with multiple people.

Mortality by age

Risk of dying from COVID-19 increases dramatically with age. Taking age 18-29 as the reference group, here are data from the CDC.

30-39 Risk of death four times higher

40-49 Risk of death ten times higher

50-64 Risk of death thirty times higher

65-74 Risk of death 90 times higher

75-84 Risk of death 220 times higher

85+ Risk of death 630 times higher

Mortality by underlying condition

According to the CDC, 94% of COVID-19 deaths had at least one underlying health condition. These include obesity, chronic lung disease, diabetes, poorly controlled high blood pressure, asthma as well as conditions or medicines that suppress the immune system.

Bottom Line

Mortality from COVID-19 is complicated. Traditional measures of mortality are impossible to obtain in the midst of the pandemic. Excess mortality is the best way to assess the impact of COVID-19 on death rates. Measurements of mortality are also not uniform and are markedly increased in older people, people of color, and those with underlying conditions. Excess mortality statistics clearly demonstrate that the death rate caused directly or indirectly by COVID-19 far exceeds influenza epidemics in recent years. The only pandemic that had similar or worse mortality was the 1918 influenza epidemic.

Basic Epidemiology II: How Superspreading Events Work

In a previous post I talked about R, the reproduction number for an infectious agent such as SARS-COV2. R is the average number of other people a person with infection infects. The basic idea is that if R is less than one, the infection dies out. This works for infections like influenza which spreads pretty much evenly in susceptible populations.

There is another epidemiologic number called k, which is the dispersion factor. The math is somewhat technical, but if k is substantially less than one that means that the infection is spread primarily by a small number of superspreader events. It is estimated that k for SARS-COV2 is about 0.1

As we learn more about how this virus spreads, we have discovered that about 80% of people who are infected are unlikely to infect anyone else. The virus is spread primarily by “superspreader events” where one infected person is responsible for as many to as 50 or 100 infections. In other words, about 20% of infected people are responsible for the majority of the spread of the pandemic.

Superspreader events are relatively rare, so if we could focus on preventing the conditions that lead to super spreader events, then we could get the pandemic under much better control without having to shut down the whole economy.

There are some well known superspreader events. The Biogen leadership conference in Boston led to infection of 99 people in Massachusetts alone. A funeral in Albany Georgia led to infection of more than 100 people. Infection in a single person in a choir practice in Washington state led to 52 infected people, two of whom died. In Arkansas an infected pastor and his wife infected 30 church member at services over several days. Three of them died. An additional 26 people were infected from church members, one of whom died.

There are common factors in each of these events. They all happened in large groups of people who were close together, mostly in poorly ventilated indoor spaces. They involved people who were talking, often loudly, or singing or eating (you can’t wear a mask while you are eating or drinking). Talking or singing produces lots of particles and are likely to aerosolize (tiny droplets hanging in the air) the virus and make it much more likely to spread. It also turns out that exercising vigorously in a closed space, such as a gym also can cause superspreader events. For example fitness dance classes in South Korea with up to 22 people in a room led to 65 new cases. On the other hand, yoga classes were not associated with new cases.

So how do we prevent superspreader events? Obviously by avoiding situations where large groups are close together in enclosed spaces eating talking or singing. That means no indoor dining in most restaurants, no exercising in gyms in groups larger than 5-10 people, and no church services indoors. It also means avoiding indoor bars. Alcohol disinhibits people and makes them less careful. It means prohibiting crowds of spectators at sporting events or concerts.

Large indoor retail spaces where it is possible to keep social distancing and wear masks are unlikely sources of super spreading events. Time of contact is important too. If you are in contact with an infected person for less than 5 minutes, you are unlikely to become infected. That means passing people briefly indoors or out is unlikely to create a superspreader event.

People have also worried about using public bathrooms. Although it is true that public bathrooms are for the most part poorly ventilated enclosed spaces, no one stays there very long. Even though flushing a toilet creates a plume of particles, they don’t hang in the air very long. As far as we know, there has been no documented case of infection due to using a public bathroom.

An excellent article that goes into more detail about super spreading events and their importance can be found in an article in Scientific American. It is written in non technical language and is easy to understand. Here is a link to that article.

Bottom Line

COVID-19 infections are primarily transmitted by superspreader events. Eighty percent of infected people do not infect others. The vast majority of infections that maintain the spread of infection are produced by twenty percent of infected people. Superspreader events are associated with large groups of people in closed indoor spaces where talking, singing, eating and drinking, and vigorous exercise are occurring.

SARS-COV-2 Vaccines: Hype vs Reality

There have been predictions by the Trump administration and by some news media as well that it may be possible to have an effective preventive vaccine for the SARS-COV-2 virus by this fall. Every even slightly positive report of phase 1 and phase 2 trials have been breathlessly reported in the media. The purpose of this post is to give some reasonable estimates of when and if there could be an effective and safe vaccine for prevention of SARS-COV-2.

How vaccines are tested and developed

Exploratory stage

In this stage academic and federally funded scientists identify antigens (molecules that stimulate the immune system) that might help or prevent a disease such as COVID-19. This stage normally takes two to four years, but because of the tremendous financial resources and number of scientists involved in the exploratory stage, there are many potential vaccine antigen candidates already identified. In fact there are currently 165 vaccine candidates who have completed the exploratory phase.

Animal Trials

Vaccine candidates are often tested in animal models. If the vaccine protects animals from infection then it is more likely to lead to human trials. Vaccine protection in animals though, does not mean the vaccine will work or will be safe in humans. The two coronavirus vaccines that are farthest along in development, the one by Moderna and the one by Oxford have shown protection from pneumonia in mice and monkeys, but the noses of the animals showed just as much virus as monkeys that were not vaccinated.

Phase 1 Trials

In phase 1 trials, a small group of people are given a vaccine to see if the vaccine induces antibodies and if there are any adverse reactions. These involve only 15-20 people. The production of antibodies does not prove the vaccine will be effecting in preventing or reducing severity of infection. Twenty eight coronavirus vaccines have completed phase 1 trials.

Phase 2 Trials

In phase 2 trials the vaccine is given to hundreds of healthy people in different age and risk groups to see if the vaccine acts differently in different groups of people. Phase 2 trials provide more information about antibody production, but do not provide evidence about the effectiveness of the vaccine in the real world. Fifteen vaccines have completed or are currently in phase 2 trials.

Phase 3 Trials

In phase 3 trials, the vaccine is given to thousands of people. Half the people get a placebo injection while the other half get the vaccine injection. Both groups are followed over time to see how many in each group get infection and how severe the infection is. Each group is also monitored for side effects. A vaccine that is safe and effective will show decreased infection in the vaccine group compared to the control group and side effects in the vaccine group should be minimal and not serious. To count as effective, a vaccine would need to be at least as effective as the flu vaccine, which usually has about 50% protection. Nine vaccine candidates are in phase 3 trials. There are no shortcuts for phase 3 trials. Depending on the prevalence of the virus in the population being tested, phase 3 trials are likely to last from 1 to 4 years.

Chances of success

The success rate for vaccines that make it to phase 3 trials is low, around 30%. The good news is that there are a very large number of vaccines in various stages of testing. It is likely that at least two or three of these will turn out to be safe and effective. There is no guarantee that any of the vaccine currently in phase 3 trials will turn out to be safe and effective, and most will likely fail. Although it is possible that we will have a safe and effective vaccine by early in 2021, it is very unlikely.

Risks of emergency approval of vaccines

Russia and China have given emergency approval of vaccines that have not yet completed phase 3 trials. This is very dangerous. The bar for safety for giving a vaccine to millions of healthy people is very high. Short cutting the phase 3 trials has the potential to harm thousands of people with a vaccine that we don’t even know will be effective.

Hopefully the U.S. will not follow suit and the FDA will not approve a vaccine until phase 3 trials are completed and a vaccine is shown to be safe and effective.

Testing for SARS-COV-2

Many people want to be tested for the SARS-COV-2 virus, not because they are sick, but to make sure they are not infectious. This is important for people to know in certain work situations where they may be at higher risk of infection, to know if it is safe to visit their grandchildren, grandparents or any other person who might be vulnerable to severe disease from COVID-19. It also may be important for young people returning to in person education in school or college settings.

In this post  I will outline the tests that are available now as well as some that may be available soon. I will discuss how sensitive and specific each one is as well as how many days after infection a test is a reliable indicator of the presence of infection with COVID-19.

Incubation Period

Incubation period is the time from exposure to the virus to development of symptoms. This ranges from two to fourteen days, but the median (middle) is 4-5 days. Because of this variable incubation period, it is difficult to predict how soon a test for the virus will be positive. It could be as soon as one or two days after exposure, or as much as 7 days after exposure. People shed the most virus and are therefore most infectious 48 to 72 hours before they get symptoms.

What does a negative test for SARS-COV-2 mean?

A negative test for the virus could mean one of two things:

  1. You are not infected with the virus
  2. You are infected with the virus but are still in the incubation period, so your test is not yet positive.

If you know the date you were exposed, then a negative test 7 days later is good evidence that you are not infected. If you don’t know whether you have been exposed, then one negative test does not distinguish between 1 and 2. In order to be sure that you are not infected, you would have to have a second negative test a week later. If you are having symptoms of COVID-19 and have a negative test, that is good evidence that your symptoms are not due to COVID-19.

PCR Tests

PCR stands for polymerase chain reaction. These tests use a tiny amount of virus recovered from a swab and make millions of copies of part of the viral RNA, which  is then enough to measure. This is the most sensitive test for detecting SARS-COV-2. False negatives and false positives are rare (as long as you are not still in the incubation period).

The standard PCR test requires a nasopharyngeal swab. This means that a swab is put in the nose and pushed all the way to the top of the throat. Not only is this not very pleasant, but it has to be collected by someone wearing full PPE (personal protective equipment) because obtaining the swab can induce coughing and therefore has the potential to release coronavirus particles into the air. The testing is done by large machines at reference labs like Quest and Labcorp. These machines can run multiple samples at a time and give results in about an hour. The commercial labs have been overwhelmed by the huge number of swabs they have received from places like Arizona, Texas and Florida. Delay in obtaining test results is common and can be up to 10-14 days in some instances.

Point of Care PCR Tests

Abbot ID Now. It has received emergency approval from FDA. The machine is about the size of a toaster and is small enough to use in doctor’s offices and pharmacies. It works with nasal or throat swabs which can be obtained by the patient, and thus does not require a technician wearing full PPE. This is the machine that is used for testing president Trump and the people around him. Results are available in about 5 minutes for a positive test and about 13 minutes for a negative test. This is a PCR test.

Mesa Biotech Accula. It has also received emergency approval from FDA, This is a very small machine that fits in the palm of your hand. It is too complicated to use as an at-home test, but would be suitable for clinics, pharmacies and perhaps schools. Test results are available in about 30 minutes. This is also a PCR test.  This test uses just a nasal swab. This can be done by the person being tested, so does not require PPE.

Point of Care Antigen Tests

Antigen tests are used to test for the presence of proteins from SARS-COV-2. This is the same technology used for the rapid strep test and for rapid flu tests. These are rapid tests, with results in a few minutes. This is a much less sensitive method than tests that use PCR. Antigen tests miss up to 20% of infected people, that is, there are a high number of false negatives. False positive antigen tests are rare, so a positive result means a high likelihood of current infection. If you have symptoms that might be due to COVID-19 and an antigen test is negative, that negative should be confirmed with a PCR test.

You cannot safely use a negative antigen test as proof that you are not infected and are safe to interact with other people, especially vulnerable people who are older or who have underling conditions.

At Home Tests

The holy grail of COVID-19 testing would be a test that people could do at home. Lots of companies are working on such a test, which would be similar to a home pregnancy test. It should work with a simple nasal swab or a saliva sample. No such test has yet received emergency approval from the FDA, but stay tuned. It is likely that such a test will soon be available.

At present there are some tests that use saliva or a nasal swab collected at home, but the sample than has to be mailed to a commercial lab. That does not really count as a home test., although it does mean that no one wearing full PPE has to collect the sample. That is definitely a plus.

Antibody Tests

As I reported in a previous post, antibody tests are used to detect previous exposure to SARS-COV-2.  Unless the population being tested has had a high percentage of people infected, positives with this test are likely to be false positives. Antibody tests are therefore not of much use to individuals and are mostly used by epidemiologists to estimate the percentage of a population that has been infected with SARS-COV-2.

Bottom Line

Lets say you want to visit your grandchildren or your elderly parents and you want to be sure that you do not have COVID-19. In that case, you would need two negative PCR tests a week apart.

Antigen tests are much faster and cheaper. If you get a positive test, you are very likely infected. If you get a negative test, it does not tell you much.

Home tests are not yet available, but likely will be reasonably soon.

 

 

 

Back to School in the time of COVID-19

Parents all over the country face the difficult decision about whether and how to send their children back to school. Advice from the president and the CDC is conflicting and confusing. The purpose of this post is to look at the evidence (what there is of it) and try to present it in such a way that it will be useful to parents trying to make this decision as well as to help teachers and other school employees who are understandably nervous about returning to in person education.

Risk of Infection in Children

It is clear that children are much less likely to have symptoms from COVID-19, but it is not clear whether children are less likely to become infected when they are exposed to people with infection. One problem is that there has been very little testing of people without symptoms using the nasal swab PCR test. Since children are much less likely to have symptoms, little testing has been done on children.

What evidence we do have in children comes from antibody testing, which tests whether people have been infected with SARS-COV-2 in the past. A study in Barcelona, Spain, which was an early epicenter of infection, showed that children were infected at the same rate as adults. That study did not separate age of children, however. Here is a link to that study. A study of antibody testing in Switzerland, however found that children ages 5-9 had a much lower risk of infection than older children or adults. The study only involved 114 children ages 5-9. Although the results were significant, this is still a small sample of younger children. Here is a link to that study.

Although there is not much evidence, what evidence we have suggests that children nine or younger have a lower risk of infection when exposed to SARS-COV-2. How much lower  is far from certain.

Risk of Children Infecting Others

A very large study of household infection rate in South Korea looked at that children that tested positive for COVID-19.  Children age 0-9 who tested positive were half as likely to infect other family members as older children and adults.  Again, the numbers of children 0-9 were small. There were only 29 children 0-9 that tested positive for COVID-19 and contact tracing showed that they infected 5.3% of their household contacts. Children 10 and over infected household contacts at the same rate as adults. Here is a link to that study.

Experience of Opening Schools in Other Countries

Germany, Finland and Norway, Japan and other countries have opened schools. All started with primary school first. All have required either mask wearing or social distancing or both. Bubbles (keeping groups of students together with no interaction with other groups was also a common practice. Community transmission was at a low level in all of these countries. So far, none of these countries has had a significant school outbreak. All of them have much lower community transmission rates than in many areas of the United States

Israel and New Zealand, on the other hand have had serious school outbreaks and had to re-close schools. In both instances there was less adherence to mask wearing and social distancing.

What does all this mean for opening schools in the United States?

The evidence so far, although somewhat meager, suggests that elementary schools can be opened reasonably safely in most states. States and or counties with high transmission rates (most of the south and west) will need to maintain social distancing and make use of small “bubbles” (cohorts of children who stay together). Counties with high transmission rates will inevitably have some school infections. Using bubbles means that children and or staff who become infected can be isolated without having to close the whole school. The evidence so far suggests that middle school and high school students transmit COVID-19 just as easily as adults. Places with high community transmission rates probably need to maintain remote learning for middle school and high school students until they get community transmission under better control.

States and/or counties with low community transmission rates (most of the northeast and midwest) can probably safely open middle and high schools as well as elementary schools. Middle and high schools will require strict attention to bubbles, social distancing and mask wearing. Smaller bubbles may be accomplished by having only half or one quarter (depending on the size of the school) of the students present on a given day. This may require a hybrid model with some in person and some online teaching.

Opening schools in a safe way as outlined above will be expensive. The economic fallout of the pandemic has left states and school districts facing the possibility of budget cuts that may require teacher layoffs. Without extensive federal monetary support for states, it will be impossible for most districts to open safely.

 

Why We Should be Worried About Recent COVID-19 Spikes in Young People

Recent spikes in COVID-19 cases in the south and west have primarily involved people in their 20’s and 30’s. The governors of those states have stated that such spikes are of less concern because they are predominantly among young people. Younger people have a much lower risk of hospitalization and death than people over 65 or those with underlying conditions. Although hospitalizations have risen in most of these states, hospital capacity has not been exceeded. Governors and health departments have admonished young people to be more careful about mask wearing and social distancing, but see no reason to not continue relaxing restrictions. Not to worry, right?

I don’t think so. It is very likely that this pattern of infection among younger people went on for weeks in places like New York City and New Jersey. We did not detect this pattern there because the shortage of tests at the beginning of the pandemic meant that we were only able to test people sick enough to be hospitalized.

There is no way to know for sure if this pattern among younger people preceded the terrible illness and death rate in New York City and New Jersey, but what we now know about how SARS-COV-2 spreads makes this very likely. Antibody testing in New York City suggested that as much as 20% of the population had been infected with SARS-COV-2.

It is clear that community spread was happening silently for weeks or months before people started coming to emergency departments in the northeast. We now know that from 30% to 50% of people who have COVID-19 have no symptoms, but can spread the virus to others.

As the people with no or mild symptoms interacted with the more vulnerable populations, hospitalizations, ICU admissions and deaths went through the roof. New York City hospitals just barely managed to avoid the situation that is now happening in India, where the health system and hospitals are completely overwhelmed and turning sick people away because there are no ICU beds or ventilators.

Could this happen in states like Florida, Texas, Arkansas, Mississippi and California? Unfortunately, I think the answer is yes. All of these states have large vulnerable populations. Many of these live in rural areas with small hospitals that would be overwhelmed by only a few very sick COVID-19 patients.

There are two things that might mitigate or prevent this impending disaster. It is probably too late to do one of them, but the other is still possible.

  1. Contact tracing for cases with isolation of cases and contacts. This is doable if the number of new cases is reasonably small. The maximum number of new cases where this will be feasible is 1-500 per month. In places like Florida, Arizona and Texas where new cases are exceeding 1000 per day, this approach will not be feasible.
  2. Universal mask wearing in public and maintaining at least 6 feet social distancing. We know this works to limit community transmission. Unfortunately, there is growing resistance to these measures.

People are emotionally and economically traumatized by the initial draconian measures that were necessary to get virus community transmission down to a manageable level. They just want to forget about the virus and get back to normal life.

Unfortunately, SARS-COV-2 has gone nowhere. It is still among us and will reproduce itself in any human host where it gains access through mouth, nose or eyes.

People who are upset and angry about not being able to return to normal life should listen to an old Rolling Stones song. The lyrics go “You can’t always get what you want. You can’t always get what you want. But if you try, sometimes you get what you need.”

What we need is to learn to live with mask wearing and social distancing for a long time to come. We also have to learn to live without sports, concerts or any other venue that requires large gatherings of people indoors. Outdoor gatherings may be possible, but only with mask wearing and social distancing.

We may have an effective vaccine that will provide long enough immunity that we can return to Pre-COVID life. There is no guarantee of that and if it does happen it is very likely to be next summer or later.

That’s all the good news I have for today! Hopefully my next post can be more upbeat. It would be especially wonderful if my predictions are wrong.

Risk and Probability Made Simple

There has been a lot of conversation about risk lately. What is the risk of having COVID-19? How much is that risk increased by not wearing a mask or not maintaining social distancing? How is that risk affected by where I live and the number of cases in my state or more importantly in my county?

Underlying all these questions is how we understand the concept of risk. What does risk really mean? What does it mean for an individual to have 0.2 risk of being infected with COVID-19?

People’s eyes glaze over when someone talks about probability theory, but I’m going to give some real world examples to show how to understand risk and probability.

The first thing that is very important to understand is that risk (which describes probability of something bad happening) is calculated for a population of people, not for an individual. If we calculate that the chance for some bad outcome is 1%, what we really mean is that if we take a large number of people who are similar in some way (age or diabetes or obesity, etc.) 10 people out of every 1000 are likely to have this bad outcome. The larger the number of people, the closer the proportion will be to 1%.

Probability has nothing to say about what is going to happen to any one individual in that population. For an individual, the risk of a bad outcome is either zero or 100%.

Lets make that a little clearer by using the example of a coin toss. For a non-weighted coin, the chance of heads is 50%. For any single coin toss, however either heads will come up (100%) or tails will come up (0%). As we continue to toss the coin a large number of times, the frequency of heads will get closer and closer to 50%. The same thing would happen if we tossed 1000 identical coins at the same time. We know that the total proportion of heads would be very close to 50%, but probability does not tell us whether any individual coin will be heads or tails.

Does that mean that the idea of individual risk is useless? Not if we understand how risk works. If there are things we can change that put is in a different population that has a lower risk for any bad outcome, then fewer people will have that bad outcome. We still could be one of the few that have the bad outcome, but it is more likely that we will be in the larger group that does not have the bad outcome.

On the other hand, we need to understand that even if the risk of a bad outcome for a population is low, those bad outcomes do happen to some people.

Let us use our example of a coin toss again. What if heads comes up twice in a row? We understand intuitively that two heads in a row could happen by chance, even if the probability is 50%. We can actually calculate the probability of two heads in a row. Since the probability for heads at each coin toss is 50%, the probability of two heads in a row is 0.5 x 0.5 or .25. In other words, if we toss a coin multiple times, there is a one in 4 chance of tossing two heads in a row. What about the probability of tossing 12 heads in a row? That probability is 0.5 x 0.5 repeated 12 times. The answer is 0.00024414062. Even though that is a very small probability, if we toss the coin enough times we will eventually get a sequence of 12 heads in a row.

The point I’m trying to make with this example is that very unlikely things happen all the time. When something unlikely happens to us that is bad, we call that bad luck. If something unlikely happens to us that is good (like winning the lottery), we call that good luck. The idea of luck as a state of being is of course a myth. It is simply the operation of the laws of probability. Unlikely things are bound to happen, but it is impossible to predict to whom they will happen. That is random statistical chance, not luck.

I will give you a real world example from my own experience. My favorite recreation is mountain biking. I have been doing it for 30 years with my only injuries being scrapes and bruises. Last summer I had a major injury that involved a broken shoulder, a concussion, and bleeding in my brain. I have absolutely no memory of the accident (I am fine now, by the way). The risk of injury in mountain biking is 0.6% per year. 10% of those injuries are severe enough to require hospitalization. The risk of injury requiring hospitalization would then be equal to 0.006 x .1 or .0006 per year. That is a very unlikely event. But there is a formula to calculate the cumulative risk of having one accident requiring hospitalization over 30 years of mountain biking. That formula is 1 – chance of not having a serious accident for 30 years: 1 – .998 = .002. Although that risk is 3 times the risk for 1 year, that is still a very small risk. Nonetheless it happened.

With all that as a preamble, I’m going to give you a link to a risk calculator for COVID-19. It lets you put in your zip code, age, gender, and any underlying risk factors you might have. It also asks if you follow CDC guidance for hand washing (reduces your overall risk by 55%) and mask wearing (reduces your risk by 68%). You also enter how many people you contact per week at less than 6 feet distance. Remember that these are just estimates. The usefulness of these risk estimates is to help you understand the risk for the population you are currently a part of and how changing your behavior might make you part of a lower risk population. Here is the link to 19 and Me.

Antibody Tests for SARS-COV-2

There has been much confusing news coverage about antibody tests for SARS-CoV-2. This post will discuss antibody tests in language that hopefully will clarify the issues surrounding antibody testing.

False Positive Tests

The news media have recently reported that antibody tests have a high false positive rate. The implication is that the tests are unreliable. Although many of the initial antibody tests that were sold were in fact unreliable, the FDA has withdrawn approval for almost all of those. The tests that are left perform very well, yet the false positive rate is still high. What is going on?

It turns out that no matter how good a test is, there will still be a high false positive rate in populations that have a low proportion of people who have had COVID-19.

Every medical test of any kind has two characteristics, called sensitivity and specificity. A test is very sensitive if it has a very low chance of missing the condition being tested for (low false negative rate) A test is very specific if it has a very low chance of being falsely positive (low false positive rate). There is no test that has 100% sensitivity and 100% specificity. Usually the higher the sensitivity, the lower the specificity and vice versa.

The sensitivity of the best SARS-COV-2 antibody test is 90% (Out of 100 people who have had COVID-19, 10 will have a false negative test). The specificity of the best test is 99% (Out of 100 positive tests, 1 will be a false positive)

From these numbers, it sounds like the false positive rate would be just 1%, but the chance of a positive test being false is also affected by the proportion of people likely to have the disease in the population being tested. This is called the pre-test probability. If the pre-test probability is low, that is if the condition you are testing for is not very common in the population of people being tested, then the chance of any positive test being false positive is high.

A reasonable assumption in most populations is that the number of people that have had COVID-19 is about 1 person out of 100. This number will be higher in places like New York City, and lower in places like Montana, but this is a reasonable number to use for the pre-test probability for most of the country.

The chance that a positive test is a false positive can actually be calculated using a formula called Baye’s Theorem.

Here is a description of the theorem in words: Probability of positive test being false positive equals the Sensitivity of the test (.90) times the pre-test probability (.01) divided by the total probability of a positive test (.0198). (The total probability of a positive test is equal to the probability of a false positive plus the probability of a true positive.)

If we plug these numbers into the equation we get:

(.9 x .01)/.0198 = .45

This means that even with a very good test there is more than a 50/50 chance that a positive test is a false positive.

The higher the proportion of a population that has had COVID-19 (whether they know it or not) the lower the chance that a positive test is a false positive.

For example, using the same test in a population where 40% of people have had COVID-19, the Bayes theorem gives a much different answer:

(.9 x .4)/.370 = .97

This time, 97% of positive tests are true positives and only 3% false positives

Thus, even with a very good test we must be very careful about interpreting positive results because of the dependence on the pre-test probability. You can see why, even with a very good test, it would not be a good idea to issue back-to-work permits based on positive antibody tests.

False Negatives

The false negative rate also depends upon the pre-test probability (the proportion of people in the population who have had COVID-19). Unlike the false positive rate, the lower the pre-test probability, the lower the false negative rate.

Here is the formula: False negative rate equals 1-Sensitivity times the pre-test probability. Plugging in the numbers we get:

False negative rate = (1-.9) x .01 = .1 x .01 = .001

There is only one in one thousand chance of getting a false negative antibody test if 1% of the population has had COVID-19

Lets see what happens if 40 % of people in a population have had COVID-19.

False negative rate = (1-.9) x .4 = .1 x .4 = .04

In this case the chance of false negative test is four in one hundred. This is forty times higher than if the pre-test probability is 1 in 100.

Bottom Line

The chance of false positive SARS-COV-2 antibody tests is too high for them to be used by indviduals or by employers to figure out who is immune. The only use for them at present is for health departments to figure out the approximate number of people in a population who have had COVID=19.

COVID-19 Deaths – Not Just a Bad Flu

A number of authors have emphasized the fact that 80% of people who contract COVID-19 have mild to moderate illness that does not require hospitalization. Some have described it as no worse than a bad flu season. A recent BBC article favorably compared the risk of getting COVID-19 with other risks people take every day. Here is a link to that article.

The worst influenza pandemic in our lifetimes was in 1957. It was caused by a flu virus called H2N2 but it was also called “Asian Flu.” Interestingly, it began in China. By the end of the pandemic 70,000 people in the U.S. had died. A vaccine was developed within a few months and more than 2 million people world wide received the vaccine. This stopped the pandemic. I was only 9 years old during the H2N2 epidemic, so I remember very little about it.

The next most severe influenza pandemic was in 2009. That was caused by the H1N1 virus. It killed 12,469 people in the U.S. Most of them were young. I do vividly remember the 2009 H1N1 pandemic. My niece was hospitalized and intubated in the ICU and nearly died. Two young people from my home town in Arkansas were also critically ill. Although both survived, they had permanent damage including amputation of fingers and toes. Once again, a vaccine was available quickly and stopped the pandemic.

The current pandemic is caused by SARS-COV-2, which is a coronavirus and not an influenza virus. The death toll both world wide and in the U.S. from COVID-19, the disease caused by SARS-COV-2, is a different order of magnitude from the previous influenza epidemics in modern times, even the 1957 one. By the end of this week, the death toll in the U.S. alone from COVID-19 will exceed 100,000. Mathematical models, which assume that mask wearing and social distancing continue, predict that the total U.S. death toll from COVID-19 will be 140,000 to as many as 230,000. Here is a link to the IHME website current modeling data.

Unlike previous influenza epidemics, we have no “back bone” vaccine that will allow the rapid development and deployment of a vaccine for SARS-COV-2. As I pointed out in the previous post, a vaccine is unlikely to be ready before summer of 2021 at the earliest.

It has also been pointed out that people older than 65 and people with underlying conditions account for the vast majority of hospitalizations and deaths. “Vast majority” means about 70% to 80%. Younger people feel that they are not at significant risk, and are less likely to follow recommendations for mask wearing and social distancing (witness pictures of crowds at the beach and at pool parties this weekend). Although it is true that 80 % of them will have mild to moderate disease if they contract COVID-19, 20% to 30% of them will end up in the ICU and a substantial number of those will die. That is not to mention the fact that even if they don’t get very sick, they are likely to infect others at higher risk.

This is clearly the worst pandemic since the 1918 influenza pandemic, and that was horrible. It is a huge mistake to take the COVID-19 pandemic lightly.