Tweet
I just wanted to share the way I organize my thinking regarding key pandemic indicators. I have read a lot of commentary by folks suggesting that just the size of the illness clusters is what they are looking at. This is similar to the way WHO has organized its Pandemic Phases 4 & 5. But cluster size may just be indicative of the size of the gathering where the first or second common source H2H transmission event took place.
So if you are trying to detect the onset of the pandemic I suggest you look for qualitative (five stages) and then quantitative indicators. Initially the qualitative indicators are most important, since they point to the virus?s H2H progression. Here the amount of viral load required to generate illness indicates transmission efficiency.
First stage requires large viral loads ? with frequent or long exposures generating illness. This would be typical of common source individual/cluster, animal to human outbreaks.
Second stage requires large viral loads ? with less frequent or shorter exposures generating illness. Typically, small clusters of people with different onset dates of illness. Since the Fall of 2004 most H5N1 illness has been associated with this kind of clustering, mostly within families.
Third stage is where the virus has become significantly more transmissible with lower viral loads required to cause illness ? frequent or long exposures, large clusters of people who spend time with each other or share a common environment. This would include transmission between patients and health care workers and the larger periodic outbreaks we have been seeing (e.g.<st1 =""> N. Vietnam</st1> doctor/nurses, Jakarta Zoo incident, Turkey outbreak, etc.)
Fourth stage also requires small viral loads ? with infrequent or short exposures, very large single human source clusters of illness. These clusters would be sustained over longer periods of time and over larger geographical areas with limited outbreaks of new subclusters.
Fifth stage casual contact H2H illness - pandemic
<!--[if !supportLineBreakNewLine]-->
<!--[endif]-->
By providing this classification format, I am suggesting that the character of the clusters is what is initially telling - considering duration, localization.
Looking at the various stages statistically, in terms of the duration and frequency of the potential transmission contact indicates how genetically the virus is changing.
Since when we are infected it is not by just a single virus with a single genetic signature but instead by a mutant swarm of virus (cf. ?The Great Influenza? page 105). It is a population of virus that has a lot of genetic variation but is centered upon a particular genetic sequence. The larger the viral load (duration and frequency) the more likely one will be infected by a virus with a genetic sequence that can cause illness, even if the center of the variation would not likely cause illness.
As the center of the viral population moves closer toward what it required to cause illness, finding a particular illness causing viral sequence becomes more likely ? thus it is more infectious.
<!--[if !supportLineBreakNewLine]-->
<!--[endif]-->
We do not need to know anything about the viral load to qualitatively follow the progress of this virus. What we need to pay attention to is the frequency of stage 3 and then 4 events.
The primary difference between these two is the number of continuous H2H2H?2H transmission events. My best guesstimate is that stage 3 would be up to 3-4 clinically interrupted H2H events and that stage 4 would be 5 or more. If this guesstimate is accurate then we are looking for a stage 4 illness cluster whose duration exceeds 3.5 weeks (first illness onset date to latest/last onset date).
Longer chains of H2H?.2H are significant because each ?passage? that the virus makes through a human host allows it to greatly adapt itself for the next human transmission event (cf. ?The Great Influenza? page 177). So without medical/circumstantial intervention, very long unbroken H2H chains signals a well-adapted virus that is ready to pandemically breakout.
<!--[if !supportLineBreakNewLine]-->
Additionally, like the branches of a tree radiating out from the trunk, when we see sub-clusters being generated from an initial illness cluster this is a sign of a stage 4 event.
The difference between my stage 4 and 5 is the virus getting closer to large scale, self-sustainment. When scientist model pandemic flu they speak of the reproductive number (Ro), which is the average number of secondary cases of disease generated by a typical primary case. As this number moves toward the most commonly used values of ~1.8 to 2.0 we will see sustained unstoppable pandemic - with the number of ill people roughly doubling every 4.5 days.
Once we start seeing stage 4 events we will know that we are very close to the onset of a pandemic - then we should look for its quantitative signature. Specifically we are looking for the likely loss of H2H transmission containment.
Stage 4 sub-clusters breaking out radially from the main cluster are problematic because they rapidly increase the number of points at which H2H transmission can take place. This puts enormous time pressure on epidemiologic tracing and systematic containment.
Besides the number of transmission points the other quantitative signal is the rate at which secondary cases are created from each H2H transmission stage (the reproductive number). To be largely self-sustaining we should look at the number of new cases being created from each successive generation of H2H transmission to be about 1.5 or more.
Containment is not about dealing with what is in front of us; it is about dealing with what is coming. Remember, given its incubation period, when one generation of sick shows up at a hospital the next larger generation has already been created.
<!--[if !supportLineBreakNewLine]-->
So while the qualitative indicators are subject to obfuscation and cover-up, the quantitative indicators are inescapable signs of a coming pandemic and cannot be covered up for longer than 3-4 weeks ? since the sheer number of ill will be very obvious and locally overwhelming.
To date I think, we have seen (though not officially acknowledged) many clusters whose H2H transmission chain has extended to 3-4 before being broken. So, I believe we are starting to coalesce on stage 3 and that we have not yet seen any stage 4 events. Moving forward we will continue to see stage 1 and 2 events with an increasing frequency of stage 3 and then some initial stage 4 events.
So the critical question is ?. Roughly every 4-5 days, are we seeing noticeably increasing numbers of H5N1 victims and are these new victims being created from 5 -10 different transmission points?
If so then I think we are at the threshold of containment loss - a pandemic.
When people are able to answer yes to this question I recommend that they quickly complete all their pandemic preparations.
nonlocality
So if you are trying to detect the onset of the pandemic I suggest you look for qualitative (five stages) and then quantitative indicators. Initially the qualitative indicators are most important, since they point to the virus?s H2H progression. Here the amount of viral load required to generate illness indicates transmission efficiency.
First stage requires large viral loads ? with frequent or long exposures generating illness. This would be typical of common source individual/cluster, animal to human outbreaks.
Second stage requires large viral loads ? with less frequent or shorter exposures generating illness. Typically, small clusters of people with different onset dates of illness. Since the Fall of 2004 most H5N1 illness has been associated with this kind of clustering, mostly within families.
Third stage is where the virus has become significantly more transmissible with lower viral loads required to cause illness ? frequent or long exposures, large clusters of people who spend time with each other or share a common environment. This would include transmission between patients and health care workers and the larger periodic outbreaks we have been seeing (e.g.<st1 =""> N. Vietnam</st1> doctor/nurses, Jakarta Zoo incident, Turkey outbreak, etc.)
Fourth stage also requires small viral loads ? with infrequent or short exposures, very large single human source clusters of illness. These clusters would be sustained over longer periods of time and over larger geographical areas with limited outbreaks of new subclusters.
Fifth stage casual contact H2H illness - pandemic
<!--[if !supportLineBreakNewLine]-->
<!--[endif]-->
By providing this classification format, I am suggesting that the character of the clusters is what is initially telling - considering duration, localization.
Looking at the various stages statistically, in terms of the duration and frequency of the potential transmission contact indicates how genetically the virus is changing.
Since when we are infected it is not by just a single virus with a single genetic signature but instead by a mutant swarm of virus (cf. ?The Great Influenza? page 105). It is a population of virus that has a lot of genetic variation but is centered upon a particular genetic sequence. The larger the viral load (duration and frequency) the more likely one will be infected by a virus with a genetic sequence that can cause illness, even if the center of the variation would not likely cause illness.
As the center of the viral population moves closer toward what it required to cause illness, finding a particular illness causing viral sequence becomes more likely ? thus it is more infectious.
<!--[if !supportLineBreakNewLine]-->
<!--[endif]-->
We do not need to know anything about the viral load to qualitatively follow the progress of this virus. What we need to pay attention to is the frequency of stage 3 and then 4 events.
The primary difference between these two is the number of continuous H2H2H?2H transmission events. My best guesstimate is that stage 3 would be up to 3-4 clinically interrupted H2H events and that stage 4 would be 5 or more. If this guesstimate is accurate then we are looking for a stage 4 illness cluster whose duration exceeds 3.5 weeks (first illness onset date to latest/last onset date).
Longer chains of H2H?.2H are significant because each ?passage? that the virus makes through a human host allows it to greatly adapt itself for the next human transmission event (cf. ?The Great Influenza? page 177). So without medical/circumstantial intervention, very long unbroken H2H chains signals a well-adapted virus that is ready to pandemically breakout.
<!--[if !supportLineBreakNewLine]-->
Additionally, like the branches of a tree radiating out from the trunk, when we see sub-clusters being generated from an initial illness cluster this is a sign of a stage 4 event.
The difference between my stage 4 and 5 is the virus getting closer to large scale, self-sustainment. When scientist model pandemic flu they speak of the reproductive number (Ro), which is the average number of secondary cases of disease generated by a typical primary case. As this number moves toward the most commonly used values of ~1.8 to 2.0 we will see sustained unstoppable pandemic - with the number of ill people roughly doubling every 4.5 days.
Once we start seeing stage 4 events we will know that we are very close to the onset of a pandemic - then we should look for its quantitative signature. Specifically we are looking for the likely loss of H2H transmission containment.
Stage 4 sub-clusters breaking out radially from the main cluster are problematic because they rapidly increase the number of points at which H2H transmission can take place. This puts enormous time pressure on epidemiologic tracing and systematic containment.
Besides the number of transmission points the other quantitative signal is the rate at which secondary cases are created from each H2H transmission stage (the reproductive number). To be largely self-sustaining we should look at the number of new cases being created from each successive generation of H2H transmission to be about 1.5 or more.
Containment is not about dealing with what is in front of us; it is about dealing with what is coming. Remember, given its incubation period, when one generation of sick shows up at a hospital the next larger generation has already been created.
<!--[if !supportLineBreakNewLine]-->
So while the qualitative indicators are subject to obfuscation and cover-up, the quantitative indicators are inescapable signs of a coming pandemic and cannot be covered up for longer than 3-4 weeks ? since the sheer number of ill will be very obvious and locally overwhelming.
To date I think, we have seen (though not officially acknowledged) many clusters whose H2H transmission chain has extended to 3-4 before being broken. So, I believe we are starting to coalesce on stage 3 and that we have not yet seen any stage 4 events. Moving forward we will continue to see stage 1 and 2 events with an increasing frequency of stage 3 and then some initial stage 4 events.
So the critical question is ?. Roughly every 4-5 days, are we seeing noticeably increasing numbers of H5N1 victims and are these new victims being created from 5 -10 different transmission points?
If so then I think we are at the threshold of containment loss - a pandemic.
When people are able to answer yes to this question I recommend that they quickly complete all their pandemic preparations.
nonlocality
Comment