I was stimulated to look at the numbers more by a rather unusually foolish question on the Yahoo Answers website the other day. The young person involved was making the point that “the government” was constantly over-hyping dangers that turned out to be unimportant (think of the swine flu fear of Gerald Ford’s administration, for example). He particularly referred to the recent Bird Flu (H5N1/N3) and Swine Flu (H1N1) scares as examples of this. Having just nursed a kid through a rather worrying case of what was presumably swine flu, I felt that this young fellow, so proudly cynical and arrogantly ignorant, deserved some rebuttal. I couldn’t do it then, but I can try now to assess the situation from a layman’s point of view and try to get a little perspective. My search is for a set of data which will let me take a stab at answering the question “Am I right that these flu variations are significantly worse than normal, and therefore that the danger is not over-hyped, or is he right that the dangers are over-hyped and not significantly worse than normal versions of flu”? As it happens, I found most of the information I wanted on the CDC website.
So – to start with Avian Flu (specifically the H5N1 variant, as the H5N3 variant seems to have been a short-term worry). H5N1 is now found throughout the world in both wild and domestic birds, and is therefore going to be a continuing threat of infection to both humans and other mammals. Since humans have little or no natural resistance to H5N1, and since circulating strains are resistant to at least 2 of the 4 common anti-viral medications (with occasional oseltamivir-resistant human infections suggesting that resistance to that third medication is spreading), creation of a vaccine that can be produced and distributed quickly would seem to be a high priority. Unfortunately, as has been demonstrated just in the past month with the H1N1 vaccine, this is easier desired than accomplished. With a mortality rate that as of a year ago stood at 60%, and with several hundred human cases over the past decade (with numbers increasing as the pool of infectious birds has increased), the worry about a jump to a re-assorted mammal-avian genotype that would allow easier human-human spread is significant. Factor in that over the past few years, H5N1 has been showing increased ability to infect mammals, including especially pigs and cats, and I think there is good reason to continue to be concerned.
This is a toss-up; the risk has not disappeared, even though the H5N1 virus has not completely made the jump to human-human transmission. It may never break out, but it could very well do so, and if it does there will be big trouble.
Now – as to the swine flu (H1N1) situation. Seasonal flu, based on the last decade’s worth of records from the CDC, begins appearing in the records of doctors visits at somewhere about 1.5% ILI (Influenza-like Illness doctor visits) around week 42, or late October. It rises at an increasing rate to a peak of 3-4% in a mild year, about 6% in a moderate year, and 8-9% or more in a severe year, generally about mid-February, and declines to baseline around the end of March. These numbers are the percentage of visits rated as ‘flu-like’, and can be any of the 2 or 3 primary strains circulating that year – many cases are not tested to determine the specific strain. The 2008-2009 curve looks absolutely normal down to week 15 (middle of April, 2009). The first US cases of H1N1 were noted in early 2009 and in April the swine flu began spreading quickly, but then died down through August. At the end of August 2008, in the US and its dominions and territories, there had been almost 600 confirmed deaths from H1N1 in the previous 17 weeks (an average of about 35 deaths per week – a somewhat misleading number, but useful as a comparison), with around 9,000 hospitalizations for an average rate of about 530 per week. The ILI visit rate during this period jumped to about 2.5% at the beginning of May, then declined to about 1.5% at the end of August. In a normal year, the rate would have been at or below 1% by May, and fairly flat at less than 1% for the rest of the summer.
Starting in week 33 (the beginning of September), the national ILI visit rate takes an extraordinary jump, rising from about 1% to about 3.5% in 4 weeks, and to over 6% in week 39 (early October). In a normal year, the rate would have been puttering along at about 1% at this time, with only a smattering of cases and the usual increase still a month or so away. A jump that much, that fast, in a regular flu year would have marked it as moderately severe or worse. The death rate from flu/pneumonia had jumped from [average] 35/week (see above) to [average] 345/week, based on about 2400 deaths in 7 weeks. Hospitalizations for flu/pneumonia had jumped from [average] 530/week (see above) to [average] 3,120/week based on about 21,800 in 7 weeks. Since the seasonal flu has barely shown up in tests, this has been almost entirely due to the H1N1 flu. The P&I (Pneumonia and Influenza) death rate for week 40 was over 6.5% of all deaths in the 122-Cities reporting system, putting it into the epidemic range for this time of year. It’s not clear whether the H1N1 caseload has peaked yet, and the hoped-for vaccine deliveries have been delayed and reduced, suggesting that the usual seasonal flu load will arrive while the H1N1 strain is still clogging hospitals and using up anti-viral medications. The death-rate as far as flus go in general, has been like a nasty Type A (which is, after all, what it is), and has been disproportionately shifted to kids and young adults, which is what was expected. Unlike normal influenzas, this one has been killing people who would normally not be considered at serious risk (eg otherwise healthy teenagers). In this way, it’s like the 1918 Spanish flu and the Avian flu discussed above. The overall mortality rate is not 30-50%, thank goodness, but it’s high enough – we’re getting a double flu season, is what it amounts to, and we haven’t seen the peak of the H1N1 yet. I suppose you could be disappointed that it hasn’t been apocalyptic, but this is not far off what most of the people I know were thinking was likely to happen. Only in the media, I think, was there a craving for, and prediction of, widespread death and destruction. “Plan for the worst and hope for the best” is a time-honored manner of dealing with oncoming problems, at least if you’re a reasonably responsible adult. It’s only among a political subset of Americans that “Plan for the best and disregard the worst” has become a mantra for life. That’s not a very good recipe for success in life, even if your leaders provide you with scapegoats to blame for your inevitable failures and losses.