Yesterday I read a new article in
Chinese Science Bulletin: "Epidemiological and risk analysis of the H7N9 subtype influenza outbreak in China at its early stage." You can obtain it here:
Download CSB Risk Analysis.
The paper scared me, and I promptly wrote a post about it H7N9: Are we in for the duration? That, in turn, scared other people, and if you're one of them, I apologize. One of my purposes in blogging about scary diseases is to learn about them while retaining (and propagating) my stark insensibility about their consequences. Calm seems to me a much better foundation for an effective response than high anxiety.
Rattled though I was, I hoped for cooler heads to prevail, and some have: Marion Koopmans tweeted some common sense about the severity scores in the CSB paper, and wondered if H7N9 was really so severe if it kills weak old men rather than healthy young people (as H1N1 did).
I'm extremely grateful for those like Marion who've helped me gather my wits after I lost them. But I think the CSB paper still deserves careful scrutiny and analysis. On re-examination, it seems weak in some respects but pretty strong in others; and politically, it's explosive.
First, what about the authors? Most are associated with the China National Avian Influenza Professional Laboratory, part of the China Animal Health and Epidemiology Center in Qingdao, Shandong province. One, Chen JiWang, is with the Institute for Personalized Respiratory Medicine at the University of Illinois in Chicago.
So it sounds as if they're chiefly veterinarians, and perhaps their analysis of H7N9 in humans isn't as knowledgeable as it might seem. On the other hand, they do work for an avian-flu research lab, and they express considerable confidence that H7N9 can infect pigs and even horses as well as birds. They cite the "iceberg phenomenon principle" to explain the relative absence of the virus in tests so far.
Can we trust their expertise? Maybe. They were quick off the mark, submitting their risk analysis only 12 days after the first cases were announced on March 31. That means they were working with a limited sample of 91 cases (some possibly added to the study before it was accepted on March 23). As Marion Koopmans points out, the case fatality rate they use may be unreliable, given our ignorance of the real extent of H7N9 in both human and animal populations.
Second, what about the Chinese Science Bulletin itself? Someone's described it as "obscure," and perhaps it is; I'd never heard of it in years of surfing Chinese websites in search of H5N1 news. The article is written and edited in far from perfect English, though I confess I'm charmed by the idea of "horse flocks" infected by H7N9.
But it was published in English, in China, and that's important. The authors evidently wanted to alert their colleagues both in China and around the world, and to do so quickly.
I read it yesterday at the same time that I was re-reading Karl Taro Greenfeld's The China Syndrome, in which he describes the disastrous epidemic-intelligence system in effect in China in the winter of 2002-03, when news of outbreaks went up the chain to the party and was considered neibu, top secret, until the highest authorities released it back down the chain. So SARS spread to many hospitals all over Guangdong, where staff had no idea what they were fighting (and themselves falling ill with). The only real news the public had was a report or two about "rumours" of an outbreak that were inspiring runs on local supplies of vinegar and various herbs. You had to read between the lines to interpret what was really going on.
I am old enough to recall the lost science of Kremlinology, when western experts had to figure out the Soviet state of politics by who was standing next to whom to watch the big May Day military parade in Red Square. That's how little we knew about the real state of affairs.
Maybe we need to consider something similar here. Maybe the Chinese really are being open and transparent, but in an oblique and translucent way: They permitted a team of veterinary researchers to go public with a report that may be hinting at much more knowledge of H7N9 than they've divulged through other means. God knows H7N9 has "teleported" itself across China from one province to another, and the only plausible means of doing so would be by rapid transmission through animals—especially symptom-free poultry being shipped hither and yon to serve booming markets. (As for the patients reporting no poultry contact, I suspect contact with poultry feces is inescapable in China.)
Given the recent policy of treating outbreaks as neibu, it's amazing that this report came out so quickly, and with such detailed forecasts. So either the highest authorities gave the go-ahead, or the authors went public without permission. I would bet the first, because a very old Chinese policy is that the nail that sticks up is hammered down. The team must therefore have persuaded the leadership, very quickly and persuasively, that they had a big problem on their hands.
The weakest part of the report is the method by which the authors predict the future course of the outbreak. Marion Koopmans thinks the severity score is "quite crude" and the prediction is "speculation." I certainly hope she's right. The paper accurately foretold the severity score of the outbreak as of the end of April, but that may have been too easy a prediction. Predictions for the end of May, June, and July would be helpful—and if they were accurate, the authors' arguments for the extent of H7N9 and the need to ditch the whole wet-market food supply system would have far greater weight.
So on balance I'm dialling back my own response from "Get your affairs in order" to that wise New England policy, "This bears watching." And I intend to do just that, encouraged by the thought that other Flublogians will kindly administer the firehose of good sense if I set my hair on fire again.
Recent Comments