Comments on: H1N1 Deaths and the Severity Calculus http://72.10.34.174/vss/2009/12/h1n1-deaths-and-the-severity-calculus/ An ARC Collaboration Thu, 20 May 2010 03:21:29 +0000 hourly 1 http://wordpress.org/?v=3.0.1 By: Carlo Caduff http://72.10.34.174/vss/2009/12/h1n1-deaths-and-the-severity-calculus/comment-page-1/#comment-50567 Carlo Caduff Sun, 27 Dec 2009 18:48:03 +0000 http://anthropos-lab.net/vss/?p=476#comment-50567 If you look at the "pandemic" of 1977, when H1N1 re-emerged after a 20 year absence, there is no shift in age-related mortality pattern. The 1977 "pandemic" is, of course, not considered a true pandemic by experts today, for reasons that are not entierely consistent. It certainly was an antigenic shift and not an antigenic drift. So you then get into an argument about what is and what is not considered a pandemic and what kinds of events you are comparing. If you compare the 1918 pandemic with the pandemics of 1957 and 1968, the W-shaped age-specific morality curve is unlike those of all other official pandemics. The 1918 pandemic is associated with a massive shift in age specific mortality which is very different from all that followed. As far as I have been able to follow the current events, the most significant factor seems to have been that most people, who were severely affected, were people with other medical conditions. If you look at the “pandemic” of 1977, when H1N1 re-emerged after a 20 year absence, there is no shift in age-related mortality pattern. The 1977 “pandemic” is, of course, not considered a true pandemic by experts today, for reasons that are not entierely consistent. It certainly was an antigenic shift and not an antigenic drift. So you then get into an argument about what is and what is not considered a pandemic and what kinds of events you are comparing.

If you compare the 1918 pandemic with the pandemics of 1957 and 1968, the W-shaped age-specific morality curve is unlike those of all other official pandemics. The 1918 pandemic is associated with a massive shift in age specific mortality which is very different from all that followed.

As far as I have been able to follow the current events, the most significant factor seems to have been that most people, who were severely affected, were people with other medical conditions.

]]> By: Lyle Fearnley http://72.10.34.174/vss/2009/12/h1n1-deaths-and-the-severity-calculus/comment-page-1/#comment-50563 Lyle Fearnley Sun, 27 Dec 2009 17:31:38 +0000 http://anthropos-lab.net/vss/?p=476#comment-50563 I have always heard that age-shift was associated with all three twentieth-century pandemics. From an article co-authored by Don Olson of New York, "A characteristic feature of influenza epi- demiology has historically been that the burden of excess mortality in interpandemic seasons occurs primarily in older age groups, whereas the burden in pandemic seasons shifts disproportionately to younger ages (19). During and following each of the three major 20th-century pandemics in which this pattern occurred, in subsequent epidemic seasons, the burden of mortality shifted proportionally back to older age groups." This is why New York City was (back a couple of years ago) developing syndromic surveillance systems that monitored for age-shifts as a 'sign' of a possible pandemic event. Of course, as it turned out, viral surveillance detected pandemic potential (by identifying 'novelty') long before any syndromic surveillance system produced any information at all. I have always heard that age-shift was associated with all three twentieth-century pandemics. From an article co-authored by Don Olson of New York, “A characteristic feature of influenza epi-
demiology has historically been that the burden of excess mortality in interpandemic seasons occurs primarily in older age groups, whereas the burden in pandemic seasons shifts disproportionately to younger ages (19). During and following each of the three major 20th-century pandemics in which this pattern occurred, in subsequent epidemic seasons, the burden of mortality shifted proportionally back to older age groups.” This is why New York City was (back a couple of years ago) developing syndromic surveillance systems that monitored for age-shifts as a ‘sign’ of a possible pandemic event. Of course, as it turned out, viral surveillance detected pandemic potential (by identifying ‘novelty’) long before any syndromic surveillance system produced any information at all.

]]> By: Carlo Caduff http://72.10.34.174/vss/2009/12/h1n1-deaths-and-the-severity-calculus/comment-page-1/#comment-50498 Carlo Caduff Sat, 26 Dec 2009 08:57:20 +0000 http://anthropos-lab.net/vss/?p=476#comment-50498 Many thanks for this interesting post, Lyle! I have a couple of comments. "Such an age-shift is a classic sign of pandemic strains and was observed in many previous pandemics." As far as I know, an age-shift was primarily observed during the 1918 pandemic. The famous W curve. I think this is not true for the 1957 and the 1968 pandemics. The other point is that during the current pandemic, the important thing may not be the age of these people, but the fact that most of them had underlying conditions. In 1918, however, the 18-64 age bracket referred to a healthy population. Experts were stunned that so many healthy people in their best age died. Hence the comparison doesn't really work, because you may be comparing two different kinds of populations. What is really surprising with this pandemic is that senior people have not really been affected a lot. That's really interesting, but seems not to be very much discussed these days. The obsession with the 18-64 bracket, by contrast, is due to the on-going controversy about the "mildness" of the current pandemic. In terms of vaccination, a better sense of severity doesn't necessarily mean a better vaccination plan (see my paper). At the very least, this is an open question. More and better information doesn't always and necessarily mean that the intervention will be better, too. As far as Laurie Garrett is concerned, her argument seems a bit strange, especially if you re-read her articles. She has certainly played the game, made a lof cash, increased her authority as an "expert," etc.. And if she would like to accuse someone for confusing pandemic influenza with severity I think it would be best if she would start with herself. Stephen: "On the other hand, it seems that a crucial feature of a novel strain is that the likelihood of mutations that move things in a much more serious direction is great." Different viruses have different mutations rates, but this is not related to epidemic vs. pandemic viruses and mutations can go both ways (i.e. more or less virulence, transmissibility, pathogenicity, etc.). Finally, the virus is just one side of the equation, the other side is the host (and the biomedical means of intervention, etc.). What is also interesting, is the question of what happens when there is a shift in medium, i.e. when numbers become public. What is interesting with this pandemic is its incredible publicity. That has a certain impact on numbers and how they circulate. Here is a recent pro-med post: "Not so, says the head of the CDC, Dr Thomas Frieden. "I think we've been completely transparent with what we think is happening. I think we have a difference of opinion on whether that is mild or severe," he says. He points out that the CDC has counted more than 250 deaths among children. "Any flu season that kills at least 3 times more children than a usual flu season I think it would be very misleading to describe that as mild," he says. But Frieden agrees that perception is what matters. The more that people think the pandemic threat is over, the fewer who will get vaccinated." So, to go back to Lyle, there are different methods of calculation, which all seem to measure different things. And different people have different stakes in these numbers. It's therefore very hard to say what this intensive debate about numbers is really about. Many thanks for this interesting post, Lyle! I have a couple of comments.

“Such an age-shift is a classic sign of pandemic strains and was observed in many previous pandemics.” As far as I know, an age-shift was primarily observed during the 1918 pandemic. The famous W curve. I think this is not true for the 1957 and the 1968 pandemics. The other point is that during the current pandemic, the important thing may not be the age of these people, but the fact that most of them had underlying conditions. In 1918, however, the 18-64 age bracket referred to a healthy population. Experts were stunned that so many healthy people in their best age died. Hence the comparison doesn’t really work, because you may be comparing two different kinds of populations.

What is really surprising with this pandemic is that senior people have not really been affected a lot. That’s really interesting, but seems not to be very much discussed these days. The obsession with the 18-64 bracket, by contrast, is due to the on-going controversy about the “mildness” of the current pandemic.

In terms of vaccination, a better sense of severity doesn’t necessarily mean a better vaccination plan (see my paper). At the very least, this is an open question. More and better information doesn’t always and necessarily mean that the intervention will be better, too.

As far as Laurie Garrett is concerned, her argument seems a bit strange, especially if you re-read her articles. She has certainly played the game, made a lof cash, increased her authority as an “expert,” etc.. And if she would like to accuse someone for confusing pandemic influenza with severity I think it would be best if she would start with herself.

Stephen: “On the other hand, it seems that a crucial feature of a novel strain is that the likelihood of mutations that move things in a much more serious direction is great.” Different viruses have different mutations rates, but this is not related to epidemic vs. pandemic viruses and mutations can go both ways (i.e. more or less virulence, transmissibility, pathogenicity, etc.). Finally, the virus is just one side of the equation, the other side is the host (and the biomedical means of intervention, etc.).

What is also interesting, is the question of what happens when there is a shift in medium, i.e. when numbers become public. What is interesting with this pandemic is its incredible publicity. That has a certain impact on numbers and how they circulate. Here is a recent pro-med post:

“Not so, says the head of the CDC, Dr Thomas Frieden. “I think we’ve
been completely transparent with what we think is happening. I think
we have a difference of opinion on whether that is mild or severe,”
he says. He points out that the CDC has counted more than 250 deaths
among children. “Any flu season that kills at least 3 times more
children than a usual flu season I think it would be very misleading
to describe that as mild,” he says. But Frieden agrees that
perception is what matters. The more that people think the pandemic
threat is over, the fewer who will get vaccinated.”

So, to go back to Lyle, there are different methods of calculation, which all seem to measure different things. And different people have different stakes in these numbers. It’s therefore very hard to say what this intensive debate about numbers is really about.

]]> By: Stephen Collier http://72.10.34.174/vss/2009/12/h1n1-deaths-and-the-severity-calculus/comment-page-1/#comment-50374 Stephen Collier Thu, 24 Dec 2009 13:56:17 +0000 http://anthropos-lab.net/vss/?p=476#comment-50374 Lyle -- Thanks for this post. It is really quite helpful. It seems that in addition to uncertainty about the number of people killed, and the demographics of mortality from H1N1, there is something to be said about the way that public health experts have to deal with a current unfolding event in relation to an uncertain future. On the one hand, it seems to me that I have read numerous references to an anxiety among public health experts and officials not to cry wolf, and thus lower their credibility when the next event comes. On the other hand, it seems that a crucial feature of a novel strain is that the likelihood of mutations that move things in a much more serious direction is great, such that as a matter of preparedness rather than as response to current severity it makes sense to do things like undertake vaccination programs that may (if I understand things correctly) have some chance of giving some herd immunity to more virulent future mutations. I guess the point in this case would be that "severity" is not the only relevant metric for determining what kind of response is appropriate. Lyle — Thanks for this post. It is really quite helpful. It seems that in addition to uncertainty about the number of people killed, and the demographics of mortality from H1N1, there is something to be said about the way that public health experts have to deal with a current unfolding event in relation to an uncertain future. On the one hand, it seems to me that I have read numerous references to an anxiety among public health experts and officials not to cry wolf, and thus lower their credibility when the next event comes. On the other hand, it seems that a crucial feature of a novel strain is that the likelihood of mutations that move things in a much more serious direction is great, such that as a matter of preparedness rather than as response to current severity it makes sense to do things like undertake vaccination programs that may (if I understand things correctly) have some chance of giving some herd immunity to more virulent future mutations. I guess the point in this case would be that “severity” is not the only relevant metric for determining what kind of response is appropriate.

]]> By: Paul Rabinow http://72.10.34.174/vss/2009/12/h1n1-deaths-and-the-severity-calculus/comment-page-1/#comment-50329 Paul Rabinow Wed, 23 Dec 2009 16:39:44 +0000 http://anthropos-lab.net/vss/?p=476#comment-50329 Lyle, Many thanks for this informative post. It is analytically clear. It leaves us where we belong conceptually-- clear that we don't have a metric or even a single venue to decide what is going on. I suppose it is also possible that the label H1N1 covers over a range of diversity as well? Paul Lyle,
Many thanks for this informative post. It is analytically clear. It leaves us where we belong conceptually– clear that we don’t have a metric or even a single venue to decide what is going on.
I suppose it is also possible that the label H1N1 covers over a range of diversity as well?

Paul

]]>