Friday, May 26, 2006

Considering Bandung

It appears the West Java provincial capital of Bandung may be the site of another (so far) small familial cluster, two young siblings. West Java has a population of almost 40 million, the most populous province of Indonesia and the second most densely populated. Xinhua is reporting the deaths of a brother and a sister earlier this week. Local tests indicated H5N1 infection. The younger sister was reported to be 10 years old, her brother 18 years old (Adnkron). Dead chickens were seen in the environment.

If this is indeed another familial cluster, does this signal a new phase in the human bird flu story? Possibly, but probably not. The official WHO and CDC line is that there have been other familial clusters with evidence of human to human transmission and the big one in Sumatra is just another example, although unusually large. In that context, Bandung, too, would be an example. I've also heard rumors of a couple of more small clusters. Is it plausible these are just more in an ongoing story of sporadic clusters?

Plausible, yes. The problem is not that WHO and CDC are concocting an implausible explanation for the Sumatra cluster but that they had previously downplayed the existence of even limited human to human spread in clusters, preferring instead to emphasize the poultry origin of virtually all cases, a position we find misleading, probably deliberately so. WHO has a mixed record on transparency, but habit of "happy talk" hs severely dmagaed their credibility. WHO will have to work hard to recoup and for some it won't be possible. But they should try. Straightforward reporting of facts would be a good start.

Back to the clusters. If we accept the virus has not changed genetically (and we are no longer willing to do so just because they say it, unfortunately), then we must also infer that there have likely been many other clusters, probably undetected. To return to Bandung, West Java had 186,000 cases of childhood pneumonia in 2005, about 6000 of them severe. This is the haystack a bird flu needle would hide in. Plausibly clusters and sporadic cases would be missed. Now that the index of suspicion is increased with the publicity from the Sumatran cluster, we may start seeing other reports of clusters here and there.

But why such a big cluster in Sumatra? Several possibilities. The index case could have been a super spreader (as was found in some cases of SARS) or it might just have been the upper tail of a distribution of cluster sizes. We'll probably never know, as investigators got there too late. Getting into the actual village at all proved difficult because of hostility toward the central government. The Indonesian government has now asked 33 villagers to self-quarantine at home. They are being followed for any signs of illness. So far, either publicly or via several of my own sources in-country, none has appeared outside this extended family.

If there continues to be no spread beyond the extended family, this is probably what WHO and CDC are saying it is, "another cluster," larger but not different in kind. The fact that there is third generation spread, however, sets it apart from other known clusters, so while my guess is there is nothing more to it, I wouldn't bet the farm on it. WHO is trying hard to do contact follow-up.

Too bad we are losing confidence they are telling us everything.