Originally Published on Facebook, March 15, 2020

Covid19 Fatality Rate: More data science, my analysis and actionable insights: I'm not going to populate this post with graphs and pictures...there are plenty of graphs around. I'm just going to give numbers and analysis. I'm also going to say up front that I don't think you need to panic over this (e.g. the supply chain of TP is intact and there is no shortage aside from panic buying--once those people have a year of TP in their closet, you will be able to buy whatever you need). Stress weakens your immune system...including stress worrying about a pandemic. Be sure to read the four positive outlooks below, as well as the cautions. I invite you to shoot holes in my conclusions by making comments--bring on the data.

We finally have enough data to know the mortality rate for Covid19. The reason we have not been able, prior, to really know what the rate was is that the exponential growth curve of the spread masked the mortality rate. Estimates were between 1% and 7%; we still don't know precisely(1), but we can now look at China where social distancing and closures of businesses that deal in nice to haves and conveniences and see that social distancing has made an impact. What we now have that we did not have before is a large number of closed cases, in fact, of 80889 cases, 66931 have recovered and 3199 have died. 10719 are still active and the outcome is not known.(2)

The method I am using is "(number of deaths)/((total cases)-(active cases))" or, as of this writing in China, 3199/(80849-10719) = 4.56%

In the interest of being conservative with the mortality rate I will consider the same rate for active cases and remove them from the equation. The reality is that deaths happen somewhat more quickly than cured cases which will mean the above method results in a slightly lower than actual percentage--but we are pretty close. Of the 10719 cases remaining active in China, 3226 are critical. Some of those people will die and that will change this number somewhat. Hopefully we are learning how to care for critical patients and the correction will be a lower rate.

We can also surmise that the total number of cases in China was higher than the discovered cases. Approximately 80% of people infected with covid19 do not get sick enough to seek medical treatment so it is safe to say that the total number of cases would have been much higher than the known cases.

My early estimates when China only had a few thousand cases reported that 1% of people who contract the virus will die are borne out in the actuals from China. Assuming that the actual cases are 5 times the known cases, the death rate from china is 4.56%/5 or 0.91%, so my early estimate of 1% was pretty close if 80% of cases in China went unidentified. At least two other models using different methods are also arriving at roughly the same conclusions.

The symptoms (a fever, dry cough, shortness of breath, and others) manifest anywhere between 2 and 14 days after a person has contracted the virus. Verified cases in China had a ten or more day lag where the patient was experiencing symptoms, but had not sought medical treatment. During that time, and during the asymptomatic time, these patients had the potential to infect anyone they contacted. The same thing is true everywhere. Where there is a known case, there are hidden cases.

Positive outlook no 1: we humans are inventive and inquisitive, as more time passes we will find more ways to improve outcomes for critical patients. The mortality rate may decline as we figure out how to best treat covid19 patients. The more we slow the growth, the greater chance we have to make a difference in the outcome when you or your loved ones become sick. Chances are, you will eventually get this virus--do everything in your power to do that later rather than sooner

Positive outlook no 2: People who have recovered are not likely to catch it again(3), nor are they likely to be significant vectors for the transmission of the virus(4). This means that those who have already been sick and are now healthy can essentially resume life as normal without risk to themselves or others.

Positive outlook no 3: During the 1918 flu epidemic, cities which took social distancing measures (e.g. St Louis) had much lower rates of infection than cities which did not take drastic measures early (e.g. Philadelphia). We are already seeing social distancing even in cities where there are relatively few known cases. This is encouraging and will help reduce the impact.

Positive outlook no 4: As testing expands we can improve the ratio of unknown cases to known cases. A known case can be kept in total quarantine until cured which seems to take as long as 6 weeks to test negative.

Caution no 1: My early spread model from last Tuesday predicted that by the end of the day yesterday in the city of 1 million people where I live, we would have three known cases (where symptoms have manifested) and somewhere between 272 and 320 unknown cases. We were still reporting 2 when I went to bed last night but hit three this morning. In ten days, the unknown cases will start showing up as official cases. These are the 20% who will seek medical treatment. The math says that in the next 20 days we will see 12-14 people die in Tucson. I hope that I am wrong, but so far my mathematical models and analysis have been more accurate than I would like. The math says that there are between 1360 to 1600 people in the community who are infected and may be actively spreading virus. I could be one of those. So could you. The fact is, we just don't know, and won't for 14 days, who has it and who doesn't. If the 1600 exercise very vigilant hygiene and social distancing we might be able to stop it at that or significantly slow it.

Caution no 2: Since we are a retirement community, particularly in winter, our (Tucson’s) elderly population, which is above the average percentage of the population for the country as a whole, will be particularly hard hit. This could tax our health care system more than average. The positive side of this is that elderly tend to be more conservative about their health and will likely isolate more on average.

Caution no 3: there is some indication that elderberry, which fights the flu, may be contraindicated for covid19 due to the specific way it works on your immune system. Avoid Sambucol to treat covid19, but if you have the flu (caveat, it can be hard to know the difference without testing), that's still an effective treatment. If you want sources on this, make a comment.

As I write, it has been reported(5) that Turin, Italy has made the suggestion to ration care and not treat patients over 80 years old. With medical treatment, 1 in 6 people over the age of 80 with covid19 die; without care that ratio will be much higher. It's really important that we do not follow suit. Please social distance now so that the infection curve flattens and we do not see a scenario where we cannot care for our citizens. Be smart: be safe. Stay home whenever you possibly can.

sources:
1) there will still be cases which were never diagnosed because the symptoms for an individual were not severe
2) current worldwide stats taken on 3-15-2020 from https://www.worldometers.info/coronavirus/
3) This is based on history--nobody knows yet if we will develop an immunity to covid19. We also do not know how it will mutate. Strains could become less lethal as it mutates. It could go the other way too.
4) It is still possible for a person who has healed to pick up the virus from a sick patient and deposit the virus on a surface, but the likelihood of transmission is lower than from someone who is sick and actively shedding virus.
5) Turin, Italy’s crisis-management proposal : https://www.telegraph.co.uk/news/2020/03/14/italians-80-will-left-die-country-overwhelmed-coronavirus/