The CDC in the US have come out with some interesting new numbers on H1N1.

They now estimate that 22,000,000 US residents have contracted H1N1. Of those, 98,000 have been hospitalized and 3,900 have died.

Several weeks ago I very conservatively estimated an 11 in 1,000,000 chance of dying from H1N1 and a 176 in 1,000,000 chance of being hospitalized because of it, vs. a 10 in 1,000,000 chance of serious vaccine complications. I advised that the shot was therefore a good bet.

The chance of serious vaccine complications has not changed, so we’re still looking at 10 in a million there. But those US numbers now indicate that we’re comparing that to a 177 in 1,000,000 chance of death and a 4,454 in 1,000,000 chance of hospitalization from H1N1 itself (that’s about 1 in 224).

It seems you are in fact 445 times more likely to end up in hospital because of H1N1 than because of the vaccine for it. And of course those numbers also indicate that you’re 17 times more likely to die of H1N1 than you are to have a serious complication from the vaccine.

The numbers will keep changing, of course, as H1N1 progresses and as we keep figuring out new ways to count things, and those US numbers may not reflect the situation in Canada, but at this point there’s no way it’ll ever look better than my conservative estimates.

So yes. Get the shot, when you can.

(Also, I admit I am amused that people are calling H1N1 “hiney”. Heh.)

It’s making me a bit crazy that people are avoiding the H1N1 vaccination. You’d rather be flat on your back for a week with fever, cough, body aches and a bad headache than have one small needle-stick and a vaguely sore arm for a day? Well, to each his own. But let’s look at the relative risks.

First, let’s do some math, just based on the known numbers.

As of yesterday, there have been 95 deaths due to H1N1 in Canada, and roughly 1600 hospitalizations. How can we estimate what our personal risk might be?

Well, that’s tricky, because they stopped reporting the number of actual cases of flu some time ago. So for the moment, let’s assume everyone — all thirty-four million of us in Canada — were exposed and got sick. That’s the most conservative assumption we can make when we’re calculating odds about how many people have died, right? That so far we’ve all been exposed and only ninety-five people have died?

A one in a million chance of dying would then have caused 34 deaths across the country, but there have in fact been 95 deaths, so in this case your odds of dying would be roughly three in one million. Your odds of being hospitalized are about sixteen times that (1600 vs. 95), if we continue along the same trajectory of illness, so about 48 in one million.

Of course, we know the actual rate must be higher, because NOT everyone has been exposed or has become sick, and yet we still have those 95 deaths: the 95 deaths have come from a smaller number of people than thirty-four million. So let’s use a less-conservative estimate, one that’s currently being proposed as reasonably likely, and say a quarter of the population — 25% of 34 million, so 8.5 million people — get sick, and we still have those 95 deaths. Then what are your odds? Basic math puts it at about one in 90,000, or about eleven in one million. And your odds of being hospitalized are still about sixteen times that, 176 in a million, or 1 in about 5,700.

We know the true odds are higher for two reasons. One, not even a quarter of the population of Canada has become sick yet, and we still have those 95 deaths and 1600 hospitalizations, so the true odds of death or hospitalization even now are higher than my calculations above. Two, we know more people will be hospitalized and more people will die, again raising the odds, since people are still getting sick.

But for argument’s sake, let’s just use the eleven-in-a-million number, even though we know it’s too low.

OK, now on to the vaccine.

Serious reactions to flu vaccines are normally about a 1 in 100,000 occurrence. We can reasonably assume this vaccine will be similar, since it’s been made in exactly the same way as previous vaccines have been made: you pick the strain, you grow it in eggs, etc.

Sometimes really strange reactions happen, like dystonia (which can also be caused, as I understand it, by the flu itself, and is a one-in-a-million kind of thing) or Guillain-Barré Syndrome (although getting the flu shot does not increase your odds of contracting Guillain-Barré generally, and is also about a 1-in-a-million risk).

Sometimes the flu vaccine is how people discover they’re allergic to eggs (since it’s grown in eggs) and they go into anaphylactic shock. This is apparently one of the most common bad reactions, and it is why responsible clinics will make you sit there for ten or fifteen minutes after you get the shot — they’re making sure you’re not going to go anaphylactic on them. But it is extremely, extremely rare for people to die from getting a flu shot.

The complication numbers here are pretty solid, known numbers; we’re not pulling them out of the air as we are in calculating death & hospitalization rates from actual flu. I won’t get into the adjuvanted vs. non-adjuvanted issue here, but the complication rates out of Europe for adjuvanted vaccines generally seem to be comparable to the 1 in 100,000 figure I’m using here.

So getting the shot is a 10 in a million (=1 in 100,000) kind of risk. Even if you’re one of the unlucky ten, though, you’re most likely still alive. And here I will repeat that because the shot is made with dead virus, you cannot contract the flu from the vaccine.

Compare that to the chance of hospitalization if you do contract flu, which we conservatively calculated above to be 176 in a million. You’re about eighteen (remember these are low estimates) times more likely to end up in the hospital if you get the flu than if you get the shot.

Also, compare that 10 in a million risk of complication from the vaccine to the 11 in a million (again, estimating conservatively) risk of death from actually contracting the flu.

They’re comparable numbers. But which risk would you rather take? The one that leaves you alive, or the one that leaves you dead?

Get the shot.

Ah, how I love Lenore Skenazy. And STATS, who interviewed her.

Perhaps the problem needed to be approached from a different angle, she thought. What if you actually wanted your child to be kidnapped by a stranger and held overnight? How long would you have to leave him outside, and unattended for that to be likely to happen? When she asked people to take a guess, the most she ever heard was three months. Some people ventured a day, an hour, and even – implausibly – ten minutes.

…

The answer to Skenazy’s question was… 750,000 years. By reframing the way the risk was framed, she took the focus away from one, and placed it on what the chance was in real time – and 750,000 years is a far more arresting and reassuring number than one in 1.5 million.

…

“I haven’t seen horrible diseases sweeping the country as a result of any child rearing technique that we’ve been using, whether it’s drinking baby formula or using a sippy cup,“ she says. “So, rather than worry about these, I worry about cars. They are the number one way children are killed.”

There are lots of interesting statistics down the side of the article (because it is STATS, after all). I would’ve like to see similar “one in” and “x years” numbers for other forms of child mortality, particularly car crashes and injuries from toys. They do give either numbers or rates-per-million, but without numbers you can compare directly it’s hard to grasp how many orders of magnitude there are between the various risks. Some sort of graph or image, even, might help, since our brains are notoriously bad at relative risk analysis.

Anyway, great interview with Ms. Skenazy. Her blog Free Range Kids has much more.

I’ve been doing my best to restrain my usual cynicism about the hysterical media coverage of health issues in the case of the swine flu, as until quite recently there have been too many unknowns to do much other than shrug and say “we’ll see”.

But it seems now that we do know some things:

1. It’s not terribly lethal, although (like the 1918 flu) it does seem to have a tendency to kill youngish, otherwise healthy people. I’ve seen death rate estimates up to 6% – which makes it several times more lethal than the 1918 flu, which had about a 2.5% mortality rate. That means that if you’re an excitable news type you get to say “200%+++ more lethal than the 1918 pandemic!!!!” because that makes it sound very dramatic indeed. But it also means, if you think about it, that even if you do catch it you stand 94 chances in 100 of recovering. Or to put it another way, a bit better than 15 chances in 16.

We should thank our collective lucky stars that if this genie is as out of the bottle as it seems to be, at least it’s a 6% — or less? It’s early days still, and the rural Mexican experience may not reflect the global experience, particularly that part of the global experience that has ready access to modern medicine — lethal genie with a 1-week incubation and not a 50-80% lethal genie with a disastrously long 3-week incubation like, say, Ebola.

Also, let’s remember that plain-Jane annual influenza strains are responsible for at least 700 deaths a year in Canada (possibly up to 2500). Not to sound callous, as every single preventable death is a horrible tragedy, but this flu may not even push us over normal numbers. (The potential difference which might not be reflected in the numbers, of course, is that we’re more used to influenza killing the very young and very old, not — as in this case — youngish adults.)(Like me.)

BoingBoing reposted this analysis of the 1918 numbers, which is useful in that it takes a sensible and calming approach. However, it also doesn’t tell us much that’s really applicable. Yes, we have good public health measures and near-instant communications now so perhaps less than 1918’s 28% of people would catch the thing. But on the other hand, we now have air travel which totally changes the game. It allows people to cross the globe well within the incubation period of the virus, and this sort of transmission has in fact already happened. No more do we have the leisurely plagues of old, which you could see coming across the world towards you for weeks or months before they arrived. I’d like to see some solid epidemiological modelling of air travel’s potential effects on this influenza virus’ transmission before I’ll be ready to agree we’re comparing apples to apples when we look at 1918. Anyway, it may be a reassuring piece to read.

2. Eating pork can’t give you the flu. Throwing out your pork products is entirely pointless. Enjoy your bacon, folks. It’s fine and it’s actually counterproductive to do otherwise and leave dead pigs moldering about. Better to clean up the pigs’ waste, which attracts flies which are (or seem to be) the actual swine-to-human disease vector in this case. The flies seem to be more of an issue than the pigs, and they are certainly more of an issue than processed pig flesh from the butcher. (If you eat or otherwise consort with flies that regularly feast on pig waste, you may wish to temporarily or permanently cease that practice.)(And look into therapy.)

3. It can be transmitted from one person to another. So the surgical masks are not necessarily pointless, depending on where you are in the world. I’m certainly comfortable taking the Toronto subway without one. In Mexico City, on the other hand, a mask is maybe not the worst idea, at least this week.

It ought to be pointed out that there are many kinds of “surgical masks”. I have not yet seen recommendations about which types are most appropriate for this purpose, so many of the people you see wearing those masks in pictures in the newspaper may not be doing as much as they think they are. Still, even the worst mask keeps you from touching your potentially icky fingers to your mouth and nose, so that’s something.

At this point, though, in Toronto I’d say just wash your hands after you get off the subway (which you probably do anyway). Or ride your bike.

4. It’s responsive to the more common antivirals in our current arsenal.

My conclusion: there’s no need to freak out just yet.

Yes, it’s making its way around the globe, and we can’t stop it.

Yes, it could easily mutate and become more deadly, although this is not generally a successful tactic for viruses; usually viruses become less lethal over time so its victims are better able to spread it. You can’t do much to spread your virus if you’re completely flattened by it, lying alone in your room.

Yes, the flu is a whole lot of no fun — someone online today used the expression “she felt she’d have to get better in order to die,” which well reflects my own experience with it.

Indications at this point seem to be that if you do catch it, the odds are heavily on your side, even more so if you are able to receive modern medical assistance in the form of antivirals and such.

So. You know. Keep calm and carry on, and never mind the excitable media.

But listen to the MOH and wash the heck out of your hands*. Think of it as an excuse to buy yourself some nice hand lotion.

–

* I’ll insert a corollary plea here: kindly do NOT use hand sanitizers unless you’re somewhere that totally lacks running water. Those things a) stink and b) are helping breed the superbug that WILL kill us all.

File under “not enough information to draw the conclusion they’ve drawn”:

Why did 90 children die?
Ontario’s child advocate was appalled to learn how many in the province’s welfare system die each year and is equally shocked at how difficult it is to get answers

First, I think we can all agree that the province’s unwillingness to cough up any useful information about the relevant cases to the child advocate’s office is inexcusable whatever the number of deaths or other issues. It’s hard to advocate effectively when you’re being stonewalled by those ostensibly working toward the same ends and I don’t blame the child advocate’s office for one second for using whatever numbers will get them the attention and cooperation they may need.

I think — hope — we can all also agree that any greater than zero number of deaths of children is very sad and horrible and such deaths are most urgently to be avoided.

What is not in this article or — just so it doesn’t look like I’m picking on this one piece, which I’m not — in any of the coverage I’ve seen, is any information that puts 90 child deaths in context for proper comparison and evaluation. How many children are there in Ontario? How many die each year, in what age groups, for what reasons? How do those population-level numbers and rates relate to the numbers and rates of deaths of children in care? Is it disproportionately high (or low, although that seems wildly unlikely), or are the rates not significantly different from rates in the population as a whole? Do the rates vary between groups — are, say, babies in care more (or less) likely to die than babies in the population in general? Small children? Teenagers? Disabled children?

I haven’t read the whole report yet so this may merely be a complaint about its media coverage. Still, if I were the child advocate’s office, I would be speaking loudly in my initial press releases about both the raw numbers and, if it’s relevant and useful, the rates. Ninety instinctively seems like a big number (awful thought, to think of ninety children dying) but it needs context to have real meaning. Perhaps something like this: “90 children in care or within a year of being in care died in 2007. This is n times the rate of death for all children in Ontario. This is inexcusable; children in our care deserve better. Wouldn’t it be nice if the government shared more information about at-risk children with the Child Advocate’s office so we could help bring down this rate?” etc.

I don’t mean to disparage the great work the advocates are doing in this case. I do regret that there are numbers being thrown around for shock value with no way to assess their real meaning.

55: Approximate weight of a sheet of 1/2″ x 4′ x 8′ drywall, in pounds.

9: Height in feet of our kitchen ceiling

639: Approximate number of muscles in the human body

600: Approximate number of those muscles that will hurt the next day if you spend a lot of time lifting 1/2″ x 4′ x 8′ sheets of drywall up to the ceiling, balancing with varying degrees of precariousness on ladders and the corners of tables, and holding the drywall sheets up there while they’re screwed down. This includes the muscles between your ribs (the ones that make it hurt when you breath), the muscles in your instep, the full pectoral suite, the full gluteal suite, and more. Many more.

39: Approximate number of muscles in your face. I am happy to report that you can drywall without pulling these.*

–

*Assuming your marriage is good, that is. I suppose if you spend the day frowning and hollering at your spouse while also holding sheets of drywall over your head, you may be risking your facial muscles too. And then how will you call your RMT, hmm?

Under the Employment Standards Act, 2000, if you are an employee who

• installs or maintains swimming pools
• grows mushrooms
• breeds and boards horses

you are not entitled to statutory holiday pay.

Study: Blue Skies for Blueberries

In terms of area under production, blueberries rank as the number one fruit crop in the country. Blueberries officially surpassed apples in top spot in 1996.

I don’t think I would have even guessed that blueberries were in the top 5, let alone bigger than apples.

Egg production in Canada totalled 436.1 million dozen during the first nine months of 2005.

That’s about 14.5 dozen eggs per Canadian. And the year’s not over yet!

Statscan’s gems for today:

Fresh vegetable prices were down 10.8% in January. This situation is unusual for the month of January and represents only the second such reduction in the past 20 years. Lower prices for tomatoes (-33.1%) and “other fresh vegetables” (-2.6%) accounted for most of this decrease. Supply has picked up after the squeeze caused by the hurricanes in the east and the wet weather in the west of the United States.

And also:

Combined ridership on 10 large urban transit systems in Canada was 2.6% higher in 2004 than it was in 2003. Approximately 1.3 billion passenger trips were taken on these transit systems, which account for about 80% of total urban transit in Canada.

Now you know!