April 23, 2003

SARS Watch

Tim Bishop's sarswatch.org.

Posted by DeLong at April 23, 2003 09:53 PM | TrackBack

Comments

I'm wondering . . . Does it serve any purpose to say that people are REALLY blowing the threat of SARS to the public way out of proportion? Because they are.

Posted by: Bobby on April 24, 2003 02:45 AM

Aside from its roll as a possible sunspot and to inspire irrational caution.

Posted by: Bobby on April 24, 2003 02:49 AM

Bobby, I agree there are so many more important issues in the world right now. To be devoting so much media attention to the "epidemic" is insulting. Modern medicine has come a long way since the influneza epidemic of 1918.

Posted by: aluwevol on April 24, 2003 06:29 AM

Are these posts sarcastic? If not, their authors may wish to go country by country and divide the reported fatalities by the total number of discharged and fatal cases, rather than the total number infected. Doing this for Canada, I get a fatality rate of almost 17 percent.

And no, youth and health is no guarantee either.

Posted by: david on April 24, 2003 07:16 AM

Are these posts sarcastic? If not, their authors may wish to try dividing SARS fatalities by the total number of discharged or fatal cases, rather than the total number of infected cases, as seems the custom in the American press. Doing this for Canada, I get a fatality rate of almost 17 percent.

Posted by: david on April 24, 2003 07:27 AM

David wrote, "Doing this for Canada, I get a fatality rate of almost 17 percent."

But that in itself isn't enough data to rationally assess the threat.

You also need to know (a) how easily the virus is transimitted, (b) what fraction develop symptoms of those exposed.

It sounds like the denominator for your 17% comprises only those who came into close contact with the virus AND who exhibit symptoms.

Cheers,

Posted by: Stephen J Fromm on April 24, 2003 10:41 AM

David wrote, "Doing this for Canada, I get a fatality rate of almost 17 percent."

But that in itself isn't enough data to rationally assess the threat.

You also need to know (a) how easily the virus is transimitted, (b) what fraction develop symptoms of those exposed.

It sounds like the denominator for your 17% comprises only those who came into close contact with the virus AND who exhibit symptoms.

Cheers,

Posted by: Stephen J Fromm on April 24, 2003 10:41 AM

I sure hope double vision isn't a symptom of SARS.

I sure hope double vision isn't a symptom of SARS.

Posted by: Fred Boness on April 24, 2003 12:00 PM

Apologies for the double post earlier. Am actually writing from Beijing and had a write-error with my first submission. Guess it got through after all....

Stephen -- while it's possible that some people are immune, the rapid spread of SARS among unprotected health care workers in the first waves of infection suggests this isn't normally the case - at least with the more virulent varieties. It is possible that these virii are difficult to contract, but given evidence of casual transmission in places like Hong Kong and Toronto, what is the better working assumption?

I agree that we don't have enough data to assess the exact threat SARS poses. But it bears considering that it is exactly those places we have the BEST data and the largest N which invite the most depressing thoughts about what might lie ahead (Hong Kong is at 16.1 percent and Singapore at 13.8 percent mortality respectively).

WHO Data:
http://www.who.int/csr/sarscountry/2003_04_24/en/

Posted by: david on April 24, 2003 02:30 PM

I think some people are missing the message here.
If you take the data from the WHO sars page, you can easily find out that:

1) The death rate worldwide is 11% if calculated by the deaths over deaths plus recoveries method, or "only" 5.9% if you calculate it by dividing deaths by total cases.

2) The rate at which cases are growing is currently about 3.5% PER DAY (divide today's reports by yesterday's). That means the number of cases doubles every 20 days or so. Usually, a scary rate like that doesn't persist for very long with a fatal disease. Well, so far we have been almost "spot on" as the British might say for the last 20 days.

3) I agree that media reports are probably making some people outside of Asia feel more panic than they should (so far), but the situation is beginning to look completely horrible within China and Hong Kong. Right now, many people are leaving Beijing for the countryside, and you can almost hear the epidemiologists screaming "No! Don't do that!" If this is a 10% killer for hospitalized victims in major cities, you don't want to think about what it could do in the Chinese countryside.

4) A very valid point can be raised that we don't know how often transmission of the virus leads to full-blown SARS. The data collected in Canada and Hong Kong now are somewhat confusing. The Hong Kong data suggest they can extract the coronavirus from tissue in 90% or more of the SARS patients they have, and from very few people who
don't have it. The Canadians have a lower success rate with patients (which is weird), and find 20% or so of non-SARS people who travelled in affected regions have the virus. Something is screwy here, and needs to be resolved.

5) The problem with controlling epidemics is that you have to do it early while you still only have a handful of cases. I actually believe that the Canadians have seen the worst of their current outbreak, but every announcement of "we have it in control" in Hong Kong has so far proven to be over-optimistic. And there is every indication that things are *still* even worse in China than the government there allows.

6) The situation I find most worrisome at the moment is actually in Singapore. The government there does not have a reputation for, um, "mildness" or lack of control, but their growth rate in cases this month has been about 3% PER DAY. That is better than 3.5% (especially when compounding), but does not qualify as "control" in the big picture.

7) I said that Canada my have the situation in hand, but if I'm the WHO, I look and see that their growth rate in cases has been the same as the world average, and has doubled in the last 20 days. 140 cases in all, the vast majority started from a single chain of transmission in a single individual. In other words, with aggressive quarantine, modern facilities, a hyper-cooperative population, and every advantage really, that's what it has taken to control this.

8) Another point is that it is not unlikely that the current outbreak will be arrested within several weeks. But then, next fall and winter, there is no reason to believe it could not come roaring back.

9) Exponential growth is just plain deadly. At a 3.5% daily growth rate in cumulative cases, do you know how many cases you *could* get in a year? If you had one case to start, that would be 283,941. If you had 4439 cases to start (like we do today), that's over 1 billion cases.

Now, if aggressive control measures could bring that down to a 1.5% daily growth rate, then you're only talking about a horrific but manageable 100,000 cases in year from our current position. So 10,000 people might die, rather than
100 million.

That is why I don't think the media is really over-playing SARS.

Posted by: Jonathan King on April 24, 2003 03:09 PM

Fred,

The double vision comment made me laugh out loud!

Thanks

Posted by: Dan on April 24, 2003 03:35 PM

Brad,

Hello neighbor (I live in the People's Republic of Berkeley as well) and thanks for the link.

To people who think that I may be making too big a deal of SARS, I refer you to the comments of Dr. Julie Gerberding, Director of the CDC (and UC Berkeley school of public health graduate).

April 22: "We still have no capacity to predict where it's going or how large it's ultimately going to be"

April 14: "only one highly-transmissible patient that can infect a very large number of people. And so even though we would like to be able to take a deep breath and relax a little bit here, this is absolutely the wrong time to do that."

there is a lot more on SARS Watch, including the fact that we don't have any effective treatment for SARS, and the fact that the mortality rate seen so far for SARS is at least 2x and maybe 4x the mortality rate of the Spanish Flu. That killed 1/2 a million Americans in a much smaller and less urban population.

Posted by: Tim on April 24, 2003 10:21 PM

Good discussion of SARS mortality rates at The SARS thread on the agonist.net. They basically make the same points that Jonathan King did.

Doc Bears Sars updates at agonist.net

The most in depth discussion about mortality rate is in the 4/23 thread.

I'm completly baffled that anyone could suggest that the media is over hyping SARS. Do you people remember O.J, Monica, Condit? Besides the medical threat, SARS is already having a huge economic impact in Asia and could be a bigger drag on the world economy than the war.

Posted by: alf on April 25, 2003 05:40 AM

woops, messed uo the link

here it is

http://agonist.got.net/cgi-bin/yabb/YaBB.cgi?board=SARS

Posted by: alf on April 25, 2003 05:51 AM

I don't think the media is overplaying the SARS either.

Jonathan King comment number 9 is scary enough for me, and worth the WHO asking for the travel bans to affected areas. Major cities are already affected economically, before the WHO pronouncements, so would it be in the interest of the economic well being of the world to add more cities to that list?

I also think it is ironic that in the recent Iraq crisis the United Nations Security Council was criticized as an ineffectual organization. When the UN WHO tries to contain SARS they are blasted by the affected communities as being too "pre-emptive"?

I think people are missing the point when SARS risk is presented in terms of your odds of getting the disease at present, which may be very small. With air travel and human population in the billions, it wouldn't take long for those odds to reach 100% if there is no isolation in affected areas.

Posted by: NKirsch on April 25, 2003 09:56 AM

I don't think the media is overplaying the SARS either.

Jonathan King comment number 9 is scary enough for me, and worth the WHO asking for the travel bans to affected areas. Major cities are already affected economically, before the WHO pronouncements, so would it be in the interest of the economic well being of the world to add more cities to that list?

I also think it is ironic that in the recent Iraq crisis the United Nations Security Council was criticized as an ineffectual organization. When the UN WHO tries to contain SARS they are blasted by the affected communities as being too "pre-emptive"?

I think people are missing the point when SARS risk is presented in terms of your odds of getting the disease at present, which may be very small. With air travel and human population in the billions, it wouldn't take long for those odds to reach 100% if there is no isolation in affected areas.

Posted by: NKirsch on April 25, 2003 09:57 AM

Why don't we call it what it is: Asian Flu. That will justify some excess media hype.

yitbos

Posted by: bruinbirdman on April 26, 2003 03:44 PM

I'm trying to model the cumulative number of worldwide SARS infections and deaths with a formula.

t is the number of days since November 14, 2002.
D is the cumulative number of SARS deaths.
I is the cumulative number of SARS infections.
e is the natural log base i.e. 2.71828

D=(e**(t/28))-1 and I=14*D

with t=2, I=1 at November 16, 2002
with t=20, D=1 at December 4, 2002

with t=174, D=499 and I=6984 at May 7, 2003

with t=539, D=229,175,810 at May 7, 2004
with t=539, I=3,208,461,338 at May 7, 2004

The doubling rate for cumulative deaths or infections can be found by solving for r in the following equation:

2*((e**(100/28))-1)=(e**((100+r)/28))-1

So, r is about 19.4 days.

Yogi Buchon

Posted by: Yogi Buchon on May 8, 2003 10:00 AM
Post a comment