Hacker News Reader

Comment by 2019-nCoV | Hacker News Reader

2019-nCoV

It's currently 171 deaths, 8,288 confirmed infected, which is 2% dead. However, what needs to be taken into consideration is that infected individuals don't die the day they are confirmed infected, there is a lag of several days.

Assuming the average time from symptom onset and diagnosis to death is 7 days, we're looking at at 20% death rate (830 confirmed infected 1 week ago). Even a 5 day lag between symptom onset to death is a ~9% death rate (1,975 confirmed infected 5 days ago).

This is exactly how the fatality rate of SARS played out.

View on HN

I'm not an expert on viruses by any means but that seems unnecessarily pessimistic. The number of infections is likely much much higher than the confirmed cases (laboratories have only so many test kits available per day) but most people stay asymptomatic. This would reduce the actual mortality rate, not increase it.

If indeed most people stay asymptomatic (but contagious), you're right that the actual mortality rate is lower. But this same outlook would translate to an increase in total fatalities as it greatly complicates containment.

The number of infections is likely much much higher than the confirmed cases (laboratories have only so many test kits available per day) but most people stay asymptomatic.

Do you have evidence of that in this case?

Viruses vary widely on how many are asymptomatic. My understanding is that having people be infectious and asymptomatic only happens with diseases that are well-adapted to their hosts. Which this one, having recently crossed a species barrier, isn't. (Its fatality rate is also high for the same reason.)

However that is a general understanding and I don't have data to base it on on this case. Do you?

There have been quite a few cases covered in the press about patients which carried the virus but didn't show any symptoms. Most recently: https://asia.nikkei.com/Spotlight/Coronavirus-outbreak/Three...

Thanks.

If they in fact can't transmit, then this is OK. If they can or, even worse, have a long latency period and will show symptoms in 2 weeks, that is very, very scary.

Our best tool for diseases like this is contact tracing. And asymptomatic carriers make that very hard.

(Useless trivia, the first documented asymptomatic carrier who was infectious was a short order cook named Mary Mallon. She is better known as Typhoid Mary.)

She wasn't a short order cook, she cooked for a number of families.

She was both.

When she was first identified, she cooked for a number of families. The outbreak that got her put away for life was due to working as a cook at Sloane Hospital for Women.

I cannot immediately find record of her being a short order cook specifically. However that is my recollection from https://www.amazon.com/Betrayal-Trust-Collapse-Global-Public...

There's a good docu on her story. It might be on YouTube.

7 days is too short. In the 2 Lancet studies (the 41 person cohort and the 99 person cohort), the time from diagnosis to the first recorded death is 11 days later, but most happened sometime after that. It's likely safe to say that diagnosis to death is 2-3 weeks.

The number of deaths is likely to be much more accurate than the number of infections, which are likely much more than 8,288, leading to a much lower fatality rate.

Too many unknowns to say. We don’t know:

- how many infected but not confirmed. They could be asymptotic or mild enough not to seek medical treatment. This would increase the denominator.

- time between infection to first symptom to death. Therefore we don’t know which way the remaining confirmed cases will go. Probably increase the numerator over time but don’t know by how much.

- unclear how accurate the reported numbers are.

IMHO assuming a 2% rate this early is a bit disingenuous. This due to the 8,288 confirmed cases not having registered an outcome yet. If you look at people cured vs. people dead, the picture looks much worse, but yeah a lot of variables still come into play (they were prob. the most vulnerable, they were the first ones so treatment may not have been the best, etc.).

Do a sanity check: does your 20% number make any sense based on what is known about the cases outside of China?

1 week ago there were only 6 confirmed cases outside of China.

Reply on news.ycombinator.com