Reading this week’s report by the President’s Council of Advisors on Science and Technology (PCAST) on swine flu preparations…
The PCAST’s 2009-H1N1 Working Group has some illustrious names, and some great scientists. So did the Advisory Committee on Immunization Practices which met in early March 1976, resolving to recommend mass immunization against swine flu. And the parallels don’t end there.
This month’s PCAST report has some strengths. One is its emphatic assertion that we are not looking at a reprise of the 1918 flu. Another is its reminder that America must occupy a generous place in the world — offering advice or help to countries whose structures or resources don’t allow them to purchase vaccine or otherwise organize themselves for a bad flu outbreak.
But some of the report’s pieces just don’t quite connect up.
For one, the third chapter “Anticipating the Return of H1N1,” makes clear that the PCAST’s flu working group aimed to develop scenarios for a second wave of H1N1 cases in the U.S. It set out to look at possibilities, not to make predictions. “We emphasize again that the baseline scenario and the alternatives above are given as examples for planning purposes; they are not predictions of what will happen,” reads a caveat on p. 18.
Fair enough — but that begs two questions.
First, what’s the distinction between a scenario and a prediction? Surely, when a Washington Post article is published within hours of the report’s release, with the lede that “Swine flu could infect half the U.S. population this fall and winter, hospitalizing up to 1.8 million people and causing as many as 90,000 deaths,” the PCAST is understood to have made a prediction — not just projected possibilities in an academic way.
Second, what predictions the PCAST makes! By the day after the report was released CDC was expressing doubts about the estimate (sorry, “scenario”) of 90,000 deaths. As VaccineEthics reports, CDC officials distanced themselves quickly — one telling Don McNeil, Jr. of the NY Times that “if the virus keeps behaving the way it is now, I don’t think anyone here [at CDC] expects anything like 90,000 deaths.” And the estimate of 50% of Americans being infected by H1N1 would require much greater infectivity than we’ve seen so far.
The report doesn’t address the caution about the timing of H1N1 “waves” offered by Morens and Taubenberger in their recent JAMA article “Understanding Influenza Backward” (JAMA.2009; 302: 679-680) — PCAST’s scenarios simply assume that H1N1 will be back in the fall. With WHO now explicit about a “second wave,” there will be even less impetus to (as Morens and Taubenberger suggest), look back.
The PCAST report also features a disconnect between the infectivity estimate and the mortality estimate.
It’s hard to explain how, if flu transmissibility really were to become high enough that a third to a half of all Americans were infected with H1N1 flu, virulence would remain so low that only 0.03% of the population would die of it. If PCAST’s scenario of 150 million infections came to pass, then surely PCAST would want to caution authorities to watch for the development of high-virulence viral variants, either arising spontaneously within the genome of the current strain or through recombination with other circulating human or animal flu viruses.
Why bother to get people worked up over a horror scenario of 150 million infections if you aren’t going to remind flu watchers that your darkly viewed future would allow for even further horrors in the form of new strains?
Narrative seems relevant here. The PCAST report, its weak disclaimers about scenarios-not-predictions aside, sometimes seems to aim at crafting the leading narrative more than at practical planning.
The narrative, as told by PCAST, involves inevitable return of swine flu, America unprepared, special needs that can only be met by vaccine manufacturers and pharmaceutical companies, and vulnerable groups who need special administrative attention.
Here, too, the PCAST report is reminiscent of the 1976 swine flu episode. The main effect of the meetings held by officials in the Department of Health, Education, and Welfare (the predecessor of today’s Health and Human Services) in March of ’76 was to create a narrative of inevitable return of a dreadful flu strain, America unprepared, and special needs that can only be met by immediate production of vaccine.
One lesson we learned from 1976 was the danger of allowing the narrative to precede the facts.
This entry was posted on Saturday, August 29th, 2009 at 7:00 pm and is filed under Disease, epidemics, Health Professions, Narratives, News, Outbreaks, public health, Uncategorized. You can follow any responses to this entry through the RSS 2.0 feed. Both comments and pings are currently closed.