« Treacle at Treasury - a moral crisis and more [updated] | eds's Blog | Treacle at Treasury - a moral crisis and more [May 5 update] »
Flu mortality is still zero outside Mexico
What's that about? What does this say about overreaction -- is it an example of overreaction reducing casualties or is it by contrast just an object lesson in political terrorism of the collective consciousness?
WHO map lists 25 deaths for Mexico and 1 for the US, but the US death was a Mexican kid who had recently arrived in the US. The map allocates casualties by reporting government.
So actual mortality in Mexico stands at 26/590 or about 4%, down from early estimates around 9%.
Actual mortality for all other official cases is 0/534, per the map's current data (May 5).
The official mortality numbers for Mexico have apparently been falling steadily from early reports around 149 which may have peaked a bit higher a bit later, to an official stat a few days ago of 84, to this current stat of 26. Is reporting on this "interest rate" headed for a liquidity trap? That is, could it like interest rates possibly go below zero at some point (see comment 25 in his blog)?
(As an off-topic aside, Krugman's analysis is flawed as comment 25 aka Borrowing Trouble hints.)
WHO map lists 25 deaths for Mexico and 1 for the US, but the US death was a Mexican kid who had recently arrived in the US. The map allocates casualties by reporting government.
So actual mortality in Mexico stands at 26/590 or about 4%, down from early estimates around 9%.
Actual mortality for all other official cases is 0/534, per the map's current data (May 5).
The official mortality numbers for Mexico have apparently been falling steadily from early reports around 149 which may have peaked a bit higher a bit later, to an official stat a few days ago of 84, to this current stat of 26. Is reporting on this "interest rate" headed for a liquidity trap? That is, could it like interest rates possibly go below zero at some point (see comment 25 in his blog)?
(As an off-topic aside, Krugman's analysis is flawed as comment 25 aka Borrowing Trouble hints.)
Advertisement
And yet if the everybody had sort of said; "Oh don't worry it's nothing." The blogs would be alive with one post after another about how the government failed again.
May 5, 2009 4:18 PM | Reply | Permalink
True. It's also better to alert early with infectious disease. After the horse is out of the barn, it's usually too late.
May 5, 2009 5:21 PM | Reply | Permalink
speaking of out of the barn, our 5:21 comments are exactly sequential! (comments are numbered)
3458525
3458526
... just numerology here, not statistics! :-)
May 5, 2009 5:36 PM | Reply | Permalink
Sure, assuming that it did turn out to get worse and worse to the point of catastrophe. That follows from my line "is it an example of overreaction reducing casualties..." We can criticize government for doing too much or for doing too little of the right thing, as well as for doing the wrong thing.
But let's be clear about just which "government" is being criticized!
Thanks for the comment.
May 5, 2009 5:21 PM | Reply | Permalink
By this fall, the swine flu will be just another flue strain that's rolled into the vaccine. I think we've all read too many books like The Stand. Watch out folks, here comes Captain Trips!
May 5, 2009 5:19 PM | Reply | Permalink
Read The Great Influenza and you might change your mind.
May 5, 2009 6:39 PM | Reply | Permalink
The question should first be: By what criteria do we judge the reaction to the outbreak.
From there we can get a feel for underreaction and overreaction.
Another issue is: How does the media's treatment of the issue affect people perception? (Wall to wall coverage of any issue makes the issue seem more important). [1]
For my part, when i get out of the car after arriving safely at my destination, i don't usually feel like having put on my seat belt was a huge wasteful overreaction for some over-hyped threat of getting in an accident.
Lastly, there was another death today of someone in the US with an underlying condition.
[1] What would actually demand 24 hour news coverage? Perhaps large scale disaster affecting you (not just disaster porn) which gives you actionable info constantly?
May 5, 2009 6:25 PM | Reply | Permalink
If I get the point of the seat belt remark, you're saying that reminding people to wash their hands is like the seat belt law.
But what I am questioning is not the seat belt law (aside: I always put mine on but I object to the law as a law), the more like the borderline hysteria about such issues.
Do you have any such criteria to offer? I agree that something like criteria is/are necessary to a deeper understanding of the issue.
The 24h news issue has two faces: You're only bombarded by 24h flu stories if you watch TV all the time. Maybe seeing repeat stories is a sign that we're watching too much TV. But so as to not dismiss the point entirely, I think the main problem is not the frequency of stories about the flu, but the sensationalism which seems to be implicit in most modern media (gotta attract viewers to attract funding to make a profit or even break even).
May 5, 2009 6:47 PM | Reply | Permalink
Ok here's a first crack at it:
With regard to judging the government's response:
1. Are the government's actions based on sound epidemiology?
2. Were those actions, *at the time they were undertaken*, supported by the evidence available of the situation? [2]
3. Were those actions commensurate with the severity of probable outcomes.
Finally, we can't really judge the response with the benefit of hindsight and knowledge of the eventual outcome. Although, we can/should factor that hindsight into planning the response to the next outbreak.
The above criteria mainly concern concrete actions such as closing schools, quarantines, etc.
Actions i would have seen as an over-reaction: closing the borders (because of #1,2), dispensing antivirals as prophylaxis (1,2), advocating widescale testing in non-symptomatic people, etc.
The real wildcard in all this is communication. We might all agree on the statistics and models and science [1], but we may disagree with how to inform the public. There is an interplay here between the public's response and government's action.
With respect to communication, the question is how urgent we must make the message to get people to react to the information in a positive way. If on day one, the government makes an announcement, and everyone goes nuts, then the government should probably make an effort to re-assure the public; if on the other hand, people don't react at all, the government should probably make a statement underlining the importance of the information.
The perception of over/under-reaction has to be made with respect to the whole of the population. We could probably ascertain the public's reaction in a number of ways: hits/calls to the cdc, public health officials, etc; school and work attendance before and after announcements; the increase in people going to doctors worried about having the disease; and so on.
People respond differently and have different thresholds. Someone who watches too much tv will probably be oversaturated and alarmed by coverage, whereas, folks who don't might not realize anything is amiss. So it is likely that there may be both over reaction and under reaction to the information at the same time within the population.
I think, the best we can do at this time (before extensive post-analysis by public health scientists which will eventually happen) is to look at it this way:
1. Did the government gauge public reaction and modify their message appropriately? Or did they keep pounding on the issue after the populace was freaked out?
2. When new information became available did the government modify its message?
I think the government did all right in those regards, but i am not representative, since i tend to look for my information from more direct sources (also i sort of know about statistics). Many of my co-workers seemed as bit more concerned than i was, but also seemed to be seeking out more information to figure out how they should react.
I'll withhold judging too much until next flu season is over with.
[1] If one *doesn't* agree with the science and models, then we have to step back and deal with that first.
[2] It seem thats the statistics from Mexican public health officials were revised extensively later on.
May 6, 2009 1:21 AM | Reply | Permalink
Nice.
"Finally, we can't really judge the response with the benefit of hindsight and knowledge of the eventual outcome."
I think you meant "without". If so, I would say that factually you would be correct. But the point of theory is to be able to make decisions before "the fat lady sings". So we should be able to make interim judgments based on the models representing whatever theories are applicable.
I agree that this is mostly about communication. I think you cover that pretty well. One problem is that information has to percolate into the society as a whole. So sometimes feed-forward is used to push towards final equilibrium faster, based on lag models of the population. This also applies in political campaigns where exaggeration is used to drive a less extreme point home to more people via the grapevine etc.
Terrorism is violence done to achieve political ends. The mental and emotional violence of sensationalism is at the heart of my point in this blog. I'm also a critic of "wheel spinning", that is, wasteful responses such as closing down schools, businesses, or borders without good reason and in a way which wastes resources (has deleterious impact on an economy for instance). Of course we don't know for a fact just how much, even AFTER the "opera" is over because we cannt run the "experiment" over again. But in the sense of examining the models and how close actual government and media participation is to ideal models, we can judge locally if not finally.
Thanks!
May 6, 2009 2:17 PM | Reply | Permalink
With regard to the "Finally...", i didn't phrase it very clearly, but what i was meaning to say was this:
There are several components to the government's (concrete) response that need judging
1. The government's playbook
2. The government's situational awareness
3. The application of the evidence to the playbook.
Item (1) represents the best epidemiological practices at the time of the outbreak. It prescribes what action to take under what circumstances.
Item (2) concerns how well we can gather disparate information to get a picture of the situation, and how accurate that picture is.
Item (3) is about the execution of the playbook based on the facts at the time.
Judging item (1) is about looking at the outcome of the actions the government takes and their effects. How effective actions were, how necessary they were, how they affected people, etc.
To do this (well) we need knowledge of the outcome. We can use this to create a new playbook that is more effective, less disruptive, and closer to ideal than the current one. (I think this is what you thought i had meant).
Item (3) though is what i wanted to focus on: the sole question of this item is did they apply what they knew about the situation at the time as prescribed by the playbook. The ideal is nearly mechanical: information in is mapped to external actions. This does not require any knowledge of the outcome, only an evaluation of whether the government did what they said they were going to do (what was prescribed by the playbook).
Item (2) affects what happens during execution and we can't assume the government to be clairvoyant.
After hearing some more information about the situation, i have some more thoughts:
Concerning item (1): school closure policy should have been layed out by the CDC. leaving it to the local school system caused confusion and apprehension when closure was inconsistent, especially between geographically adjacent districts. If there had been a worse outbreak, it might have made overall efforts ineffective.
Item (2): This seemed to be the major failure. Inaccurate info coming from Mexico gave us a skewed picture of the outbreak which led us to believe it would be more severe than it was. It may have even been more than one virus. One thing that might prove useful is closer cooperation (and perhaps an active involvement) with other countries health institutions.
Item (3): i felt for the most part it did alright with the exception of Biden's comment. the admin seemed to let the experts drive rather than taking feel-good-but-ineffective actions.
Moving to the communication aspects, i found the announcements by the media of the WHO pandemic alert levels to be annoying. Hitting level 5 of 5 sounds bad, but really it only meant that the outbreak was propagating and self-sustaining in several countries. I'm not sure people knew what to make of it though...except that it was the highest it could be.
The options the government has with regard to information is transparency vs. control.
Controlling information, although it might make people feel better, seems counter to my values, and is in the end impossible with the internet.
Rather than dictating what the news reports, it might be better to provide copious information so that people can look at it first hand, rather than paying attention to the most newsworthy (explosive or sensational) item on the news. Giving good, accurate information will establish trust and give people confidence in the actions being taken.
The question of politics raises the question of whether public health is a political end. It certainly is by definition (ie it is a concern of people at large). Or do you see there being an ulterior motive (perhaps using fear to make people welcome government intervention or to drive news channel ratings)?
I am not keen to adopt your terms just because they are so emotionally loaded, but i think i understand what you mean.
The weakest links in my opinion are the media who just likes juicy stories and the US population who doesn't like to educate themselves.
However, making them do the right thing (by coercion or not) leads to some very bad places quickly (or at least the appearance of those things). I think both issues can be mitigated by providing understandable, accurate, and trustworthy information to people.
May 7, 2009 6:29 PM | Reply | Permalink
Texan dies of swine flu; Mexico ready for business.
May 5, 2009 10:16 PM | Reply | Permalink
That is the second article I've seen with next to zero info about the death.
May 5, 2009 11:02 PM | Reply | Permalink
"Health officials stopped short of saying that swine flu caused the woman's death. "
http://www.cbsnews.com/stories/2009/05/05/health/main4992647.shtml
Many more details here, she had just given birth after being admitted to the hospital with flu symptoms!
May 5, 2009 11:06 PM | Reply | Permalink
The 1918 flu mutated and made a potent return during the regular flu season. The media shouldn't create panic over it, but researchers should definitely keep an eye on it.
May 5, 2009 10:43 PM | Reply | Permalink
That sorta goes without saying over here. But I'm interested in what has happened so far and how to understand it.
Thanks for the comment.
May 5, 2009 10:55 PM | Reply | Permalink