H5N1 Cluster Concerns
Recombinomics Commentary 03:53
December 29, 2007
In Pakistan, for example, the World Health Organization is investigating what could be the largest cluster of human-to-human transmission so far. A veterinarian and five members of his family are suspected cases, along with three others involved in culling infected chickens. Such clusters are rare, but they have occurred in Thailand, Vietnam and Indonesia.
The comments above are among the many describing the longest and most sustained H5N1 human to human (H2H) transmission chain reported to date, but these comments create confusion because the simple concept of H2H has been very carefully defined to avoid calling short transmission chains H2H.
Therefore, it is worth review the history, since the number of H2H clusters is much greater than the general public (and journalists writing for the general public) realizes, and clusters are likely to increase in number and length in the near term
All H2H is a cluster, but not all clusters are H2H. For H5N1, virtually all clusters really are H2H, but they aren’t PROVEN H2H, and WHO makes a distinction, creating the illusion that most clusters are not H2H, when in fact the exact opposite is true.
H5N1 is transmitted from birds to humans very inefficiently. Consequently a cluster, defined as two or more people linked by time and space, is cause for concern because it signals a more efficient transmission to a human.
Clusters can have two different origins. One is a common source such as poultry, while the second is H2H. These two possibilities are distinguished by the time gap between disease onset dates, which is usually 2-4 days from exposure. Consequently, two people infected by a common source will develop symptoms at the same time (within 1-2 days of each other), while H2H will create a gap of 5-10 days because of the incubation time between the initial infection, and the infection rising to a level that supports H2H.
The vast majority of clusters have this time gap between disease onset dates, because most clusters involve family members in close contact. That is because an infected family member creates many opportunities for transmission during care. Other family members don’t have training in universal precautions, don’t wear gloves, and are under the misconception that almost all H5N1 come from birds (which is perpetuated by the media)..
WHO avoids acknowledging that these small clusters are H2H by saying they are not PROVEN H2H, and requiring as proof an absence of a common source. H5N1 is not governed by WHO definitions, and the H2H happens, regardless of how many birds died within a 50 mile radius of the cluster.
Consequently the number of H2H examples that fit the WHO definition is few because most of the time the index case is infected by a poultry source and family members live with the index case. Although individual clusters are difficult to prove using WHO’s definition, an overview clearly demonstrates that most of these clusters have a 5-10 day gap between disease inset dates and are H2H.
However, since even common source clusters can be due to an ease of transmission, clusters alone signal problems. When the transmission chains grow from H2H to H2H2H or H2H2H2H, the transmission become obvious, as is the case for the cluster from Pakistan (false negatives not withstanding).
In summary, short chains (H2H) are common and most short clusters do have the time gap. The number of PROVEN H2H clusters is not very relevant, because the transmission is the issue. The WHO definition is just for deniability of H2H for use in panic control.
However, when the transmission chain length makes H2H obvious, then a new definition comes into play, which currently has evolved into sustained COMMUNITY transmission. H2H is not an all or none event, and as the transmission chains become longer and more common, the approaching pandemic becomes more obvious.
The H2H trend is clear, WHO press releases and evolving definitions not withstanding.
Media Links
Recombinomics Presentations
Recombinomics Publications
Recombinomics Paper at Nature Precedings