In Seattle recently, I met a pulmonologist who said that the H1N1 virus has him busier than he’s ever been, his hospital beds full of flu patients. The uptick hit particularly hard about 10 days ago, he said.
How has the flu been playing out across the country?
Craig Feied, the physician and technologist we write about in SuperFreakonomics (yes, we’ll run a virtual book club session with him soon; he’s in Chapter 2), sent along the following data picture. “Some doctor,” he writes, “made in a few seconds using Amalga.” That’s the hospital software system that Feied and Mark Smith developed at Washington (D.C.) Hospital Center, and which was later acquired by Microsoft.
Here’s Feied’s commentary on the picture:
For anybody who wonders how much “hype” there is in the H1N1 story, here is an Amalga-created graph of flu seasons in Washington, D.C. from 2002 to 2009.
ILI stands for “Influenza-Like-Illness,” meaning that this includes everybody who shows up at the hospital primarily for symptoms of flu (there’s no attempt to confirm the diagnosis of influenza with specific tests, so this likely includes some people with the “common cold” and people with pneumonia or other respiratory infections).
For each year, January 1 is at the left side of the screen and December 31 is at the right. Historically, the “flu season” has a small bump in the fall, with the serious spike usually beginning early in January and extending into the beginning of March.
It’s instantly apparent that this year is unprecedented in recent history. Long before the typical spike season we are already seeing more cases than we’ve ever seen — and presumably the worst is yet to come.
It’s possible, of course, that some of the surge was caused by patients who wouldn’t have gone to an E.R. but for the media “hype,” as Feied puts it, about H1N1.
To that end, Feied offers another data tool: the the Emory screening algorithm — a self-assessed flu survey “to help people decide when it’s time to go to the E.D.,” he says, “versus when they can safely stay home. As this problem reaches disaster proportions, emergency departments and physician offices will need all the help they can get.”
and 
. You can also watch it on 
According to the Emory screening (in my test at least) you should call the doctor (and maybe go to the hospital) if you’ve a 30 year old male who’s been sick for 1-2 days, and you have a fever, coughing, a sore throat, and are dizzy when you stand up. I’ve had the flu before, and that describes my symptoms pretty well, but I never thought about going to a doctor. Am I crazy? Should you really go to the doctor for symptoms like that? Shouldn’t you just stay at home and wait to recover? What’s the doctor going to do for you? Of course if it gets much worse or lasts a long time I can understand… but those symptoms didn’t sound unusual to me.
Our pediatrician said “come in if you have three days of continuous fever and lethargy, or a fever over 104.” Then he noted that most patients he sees have neither of these symptoms, and he is seeing a lot of people with low grade colds who panic. I think the assumption that all flu is H1N1 is ridiculous without at least substantial testing of the non-admitted population.
While on the surface it certainly appears that the death rate at this point may be slightly higher than the normal death rate in any given flu season, there is one questions that is not being answered, or asked. What is the death rate of individuals who have contracted a seconday infection due to having contracted the H1N1 virus? As was the case in 1918-1919 most victims did not die from the virus itself, but from the secondary infection. The secondary infection presents itself in the form of bacterial, viral, or hemorrahgic pneumonia. Currently in the Ukraine, of the 189 patients that have died of viral pneumonia, 90 have died of hemorrahgic pneumonia. These cases are very similar to how things went in 1918-1919, and again, it was the secondary infections which did more of the killing. At this point the medical community does not consider the secondary infection an H1N1 death unles the victim actually has the virus upon death, even though the virus percipitated the secondary infection. So again, in order to get a more acurate death toll it would be helpful to know how many deaths were caused by a secondary infection, as it is directly related to the H1N1 virus itself.
My bad, of the 189 pneumonia deaths 90 were hemorrahgic pneumonia.
You guys would probably love Hans Rosling’s approach with his News:Death Ratio:
http://www.youtube.com/watch?v=V8bUtbODV-Q
H1N1 has been a bit of a dud, even by normal flu season standards in Australia. I wouldn’t get to excited there on the other side of the pond unless the virus mutates into something a heck of a lot nastier than it is now…
There are bigger problems in the world, no?
I’d be interested in a comparison between the Amalga data and absentee data from school districts in areas around the Amalga-reported hospitals.
Locally , the school district has seen a huge rise in absenteeism.
Anecdotally, my own family’s bout with the flu has been pretty similar to that graph. Most years we see one of the six of us really sick, and it’s usually in January or February. For the last two weeks, though, two-thirds of us have fallen ill with several days of fevers, aches, chills, and the rest of the Type A Influenza symptoms (H1N1 is a subtype of Type A).
Fortunately (and again, anecdotally) the symptoms respond well to fluids, sleep, and acetaminophen.
I live in Edmonton Alberta, where we normally experience ‘flu season’ earlier than many other regions (possibly due to the diminishing sun exposure in our northern clime). What we’re seeing over the last few weeks is an incredible surge in H1N1 cases. Both myself and 2 of my young boys (ages 4 and 7) came down with what was likely H1N1 with symptoms of fever, chest and sinus congestion, body aches, headaches and fatigue. Truthfully, the symptoms weren’t that bad, though it was concerning that my 4 year old’s fever lingered for quite some time above 105 F and was unresponsive to anti-pyretics (Tylenol). Though this situation would normally warrant a trip to the local hospital, we were informed by the local healthcare phone line that unless my son’s lips started turning blue or we couldn’t wake him, we should keep him at home as the hospitals were so over-run that he would essentially be neglected at the ER. Many parents that I talk to described similar situations – some worse, some better. Also, for my first time ever, I know of persons who are not the elderly who have died, or nearly died of flu infection.
I assume that most people who catch H1N1 will experience something similar to my family, in that they may feel mild to moderately ill fo a few days, and then go along their merry way. What I see as the scary aspect to H1N1 is echoed by the graph printed above – the speed at which the virus spreads. The effect of this isn’t neccessarily that H1N1 is any more severe, but that the sheer numbers of people that are infected over a short period of time and the effect that will have on health care delivery. It may possibly turn out to be a tidal wave of cases that overwhelms the health care system over a period of several weeks.
If you think of how many people get hospitalized in a typical flu season stretching over 5 months and squish all of these folks into a 6 to 8 week period, you end up with shortages of hospital beds, staff and equipment. In our city, health administrators are extrapolating a significant shortage of ventilators, respirators and other equipment in the coming weeks. That’s the scary part – that if you are one of the unfortunate souls who does get quite ill, the medical care that would save you in a normal flu season may not be available to you.
So… either get the shot early, or get the flu early.
Isn’t there evidence of previous “epidemics” that have been spurred by mass panic akin to that caused by the H1N1 media coverage? Has anyone ruled out mass hysteria? Is H1N1 truly that much more virulent than any other strain of influenza? It doesn’t make sense that this year’s spike in cases would be so out of whack with the data from previous years, does it?