Given that I have a 4-year-old with asthma, I’ve been doing a lot of reading and researching on this year’s H1N1 outbreak. I’ve written various things to various folks in private emails, and I thought I’d post them here as they may be of benefit to those looking for answers in a media environment characterized by fear-mongering on both sides of the debate. Or patronization (“Vaccines are safe!” end of discussion). Or conspiracy (“Never trust government / corporations / media / scientists!”) For those wanting to try to weigh risks and benefits in a scientific / mathematical way, read on. For those who’s minds cannot be changed no matter what new evidence may come to light, on either side, power to you—but know that your view is not based on science. Science is always open to new evidence and changed conclusions because of it.
I’ve gathered information from three primary sources. In the US, the CDC. In Canada, FluWatch, from the Public Health Agency of Canada. And from Australia, who has now been through their usual winter flu season, the government’s Influenza Surveillance page. At the time of writing, the most recent data comes from the week ending October 9/10th. I have relied on wikipedia for information about Guillain-Barre syndrome and the 1976 flu season.
Here’s an article I thought was one of the more even-handed discussions of H1N1 and the vaccine, Andre Picard’s Reader Questions on H1N1 Answered from Canada’s Globe and Mail newspaper.
Here’s an article arguing against the vaccine, that was sent to me by a friend. I found this article raised a number of valid issues of concern, and was far better than most on the anti-vaccination side, as she/he gave numerous references: Seven Inconvenient Truths about the 2009 H1N1 Pandemic. It’ll probably be easiest to open this article in another window to follow along with my comments about it, which are as follows:
1 - Yes, correct. ‘Pandemic’ refers to how contagious something is, and how far/easily it has spread, not to how deadly or severe the illness is. And yes, many folks misunderstand the word.
2 - She/he is using old data (August), but is basically correct that H1N1 has overall been less deadly than a typical seasonal flu in the US. However, the deadliness of this illness varies dramatically by age group, underlying condition, etc. From my read of the evidence, this flu has been far less deadly overall because those who flu normally kills are not coming down with it, perhaps due to natural immunity built up during previous H1N1 epidemics (such as in '57). Among children though, H1N1 has killed as many (I believe as of today slightly more) than any of the three previous flu seasons. (86 confirmed H1N1 deaths as of last Friday, compared with 46-88 pediatric deaths in the US in the previous 3 seasons.) Note too that 43 of these deaths have come about in the past 6 weeks (Sept and first half of October), and we are by no means thru what is typically the peak flu season as yet. BTW, Australia, which is now thru its typical flu season, reported a death rate of 8.8 per million (185 deaths, pop. 21 million). If the US has a similar death rate, it should have 2675 deaths--far less than in a typical flu season, but certain groups will have far higher death rates than in a typical year.
3 - I know the US stopped counting H1N1 cases after lab results were showing that 97% of all suspected flu turned out to be H1N1 rather than seasonal. My understanding was that deaths and hospitalizations are still being sent for confirmation re: H1N1; in fact, one child's death suspected of being H1N1 turned out not to be, which is why the US went from 76 child deaths two weeks ago, had 11 more last week, and only totalled to 86. I also wonder if this is any different than with seasonal flu, where what ends up killing the elderly is secondary infections like pneumonia and not the flu itself, but it is recorded as death by flu.
4 - (which is actually labelled 3 again). This I think is a very serious issue, as there does not seem to be a lot of studies looking at the effectiveness of vaccines, especially for flu. But I have a qualm with the data presented. For children over two, she/he says:
"The same meta analysis found influenza vaccines effective 33% of the time in children over the age of two. (16) Followed to it's logical conclusion, this means the flu vaccines are ineffective 67% of the time in children over the age of two."
I went and found the referenced study online, and this is what I found:
"Fifty-one studies with 294,159 observations were included. Sixteen RCTs and 18 cohort studies were included in the analysis of vaccine efficacy and effectiveness. From RCTs, live vaccines showed an efficacy of 82% (95% confidence interval (CI) 71% to 89%) and an effectiveness of 33% (95% CI 28% to 38%) in children older than two compared with placebo or no intervention. Inactivated vaccines had a lower efficacy of 59% (95% CI 41% to 71%) than live vaccines but similar effectiveness: 36% (95% CI 24% to 46%)."
Clearly we need more research here, but we also need to understand the difference between efficacy and effectiveness: from her logic, the vaccines are "ineffective" 67% of the time, but the study being referenced says they "have an efficacy rate of 59-82%". From what I could gather doing more investigation, efficacy is the rate by which incidence (not severity) of a disease decreases between vaccinated and unvaccinated populations, under ideal conditions and carefully controlled clinical trials. "Effectiveness" is whether or not the vaccine benefited those getting it in the real world, not under idealized clinical conditions (looking at data and rate of disease after the fact). Of course, there are many reasons why real world conditions show worse results, like vaccines being administered incorrectly (wrong dosage, wrong timing/spacing of vaccines), folks already being ill with disease at time of vaccination, folks consuming other substances that limits the vaccines effectiveness (like perhaps Tylenol, a recent study suggests), etc. But she/he does highlight some serious issues here--we need more study of the very young, and we need more investigation into why there is such a discrepancy between efficacy and effectiveness rates.
5 - One the one hand, I agree, but see little way around this when a disease is spreading quickly. On the other, I find the fact that pretty much ALL first world nations are rolling out H1N1 mass immunization campaigns, even those who have differences from the US when it comes to routine childhood immunizations and schedules, to be of some comfort. I cannot find any other country using the live FluMist version, BTW, other than the US, and that alone is enough to make me suspicious of it (that version will not be available in Canada anyways).
6 - Yes, vaccine risk may be worse than disease risks, and that is why you need to have hard data (scarce!) to compare risks to benefits. Guillain-Barre is a serious illness, to be sure, but we need to know how deadly or otherwise life-altering it is for those who get immunized. The data I have found suggest rates of G-B varying from 2 per 100,000 immunizations thru 1 per 1 million. It's also important to note that of those diagnosed with G-B, the vast majority make a full recovery after treatment; about 5-10% continue to have severe disability, and 2-4% die. If I'm doing my math right, we get (at most) 4% of 2 in 100,000 shots leading to death by G-B, which means 2 in 2,500,000, or 0.8 per million. On the low side, the rate would be 2% of 1 in 1 million dying of G-B, which means 2 in 50 million, or 0.04 per million. Right now, Canada's death rate from H1N1 is 2.4 per million, and again, Australia's was 8.8 over the flu season. Of course, there are other possible side effects to shots, and these need to be factored in. I am open to the idea that various other illnesses are linked (e.g. autism), but have yet to find solid scientific proof (replicable, peer-reviewed).
It's also interesting to note that the 1976 H1N1 flu did not behave like the current outbreak. The majority of those who had flu that year had seasonal flu, not H1N1, and deaths from flu affected the typical groups (elderly, etc.). Very sad as the H1N1 vaccinations were damaging with G-B, and probably unnecessary given this information.
7 - I honestly don't know enough about "herd immunity" arguments, and have never been persuaded by them anyways. I think each individual needs to protect themselves as best they can given their particular health issues, age, etc.
So to sum up looking at her final 6 points:
1 - I agree in some ways.
2 - Overall true, but the severity is much higher for some groups compared to seasonal flu.
3 - Not sure about risk of contracting (I'd say it's higher than average given that it is a novel strain). And while her risks of death may be less than with seasonal flu, they are not for all--e.g., my 4-year-old asthmatic child
4 - The evidence here is conflicting, and we need to understand the difference between efficacy in ideal setting and effectiveness in the real world. I am not sure if any of these studies address the severity of illness--e.g., a vaccine that did not prevent people from getting flu, but did reduce hospitalization and death rates from it, would that be considered effective?
5 - Don't know enough to comment.
6 - Agreed, but these need to be weighed against known risks from H1N1 (or whatever vaccine in question). We also need to ask if the outcomes for those with H1N1 are only two: death, or full recovery. Do any folks have lifelong complications / disability from it?
And as for her conclusion? I agree there are many other proactive things one can do to lessen the chance of getting H1N1, and increase the chance of full and easy recovery if you do--and we're doing them. But "make sure" I (or my asthmatic child) suffers only a mild case? Sadly, I doubt it.
Here’s another link I came across from a friend, Dr. Jay Gordon’s Swine Flu, Other Viruses and High Anxiety, and my thoughts on it:
He is correct that for the vast majority of folks this is no different--the illness is minor, and they will not feel any worse than regular seasonal flu. He is also correct that the reason we have such a high infection rate is because it is a new strain, and in new strain years infection rates skyrocket as virtually no one has built any immunity to similar strains (this is one reason the elderly may not be being infected as much this year; it is hypothesized the '57 flu was similar enough that older folks built immunity to this years' strain then.)
However,
1 - many studies are showing that rates of death for those with no underlying health conditions to be far higher than his 20-25% (though he quotes only children, and there I am not sure. The studies I can find relate to adults, a common one mentioning that 45% had no underlying health issues.) This result—45% without underlying health issues—has been presented in two very different ways in the media, and both headlines are technically correct:
"Almost half of all H1N1 victims had no underlying health problems" and
"The majority of H1N1 victims had underlying health problems" Reader beware!
2 - Child deaths from H1N1 in the US are already at levels seen in a bad regular flu season (86, whereas in the past 3 years child flu deaths have ranged from 46-88). 43 of these have come in September and the first half of October alone; in other words, in the past 6-ish weeks, the pediatric death rate is equal to that of an entire regular (low) flu season. Peak flu season usually occurs in the Xmas-January time frame.
3 - In various studies, pregnant women comprise 6 percent of those who have died from H1N1. Pregnant women comprise 1 percent of the general population. In Australia, pregnant women aged 20-39 comprised a whopping 32% of all hospitalizations for H1N1.
4 - His mathematical illiteracy drives me crazy. Here's a quote: "Some of us will get high fevers and have to miss school and work for a few days and 99.9999% of us will remain completely unaffected after the flu season except that those who contract Swine Flu this year will be protected if it gets meaner and more virulent in coming years as expected."
Ok, bear with me, those of you who are math-phobic. Let's remove the decimals and start as though he said 99% of us remain unaffected. That would mean 1 in 100 of us are affected. Ok so far? Let's continue this line of reasoning.
99% of us unaffected means 1 in 100 are affected by H1N1
99.9% means 1 in 1000 are affected
99.99% means 1 in 10,000
99.999% means 1 in 100,000
and, his original quote:
99.9999% means 1 in 1,000,000 of us are affected. In plain English, only 1 in 1 million of us are or will be affected by H1N1.
Now, let's just look at the most recent stats from Canada.
"The national hospitalization rate was 4.6 per 100,000 population with the highest rates in children under 15 years of age (10.9 per 100,000). The national mortality rate was 0.24 per 100,000 population; those 45 years and older had the highest mortality rate (0.35 per 100,000). ICU admission rate and ventilation rate were also elevated in children under five years of age (1.6 and 1.1 per 100,000, respectively)."
Right off the bat, a hospitalization rate of 4.6 per 100,000 means 46 people per 1 million hospitalized, and I don't know about you, but I'd count being hospitalized as being "affected" by the flu. That becomes 109 out of 1 million for the under-15 crowd. And the national mortality (DEATH!) rate is .24 per 100,000, which is 2.4 deaths per million...again, something you'd have to call being "affected by swine flu". (Sorry for the sarcasm, and I mean no disrespect to anyone who has lost friends or family to this disease.) Australia’s death rate was 8.8 per million, and hospitalization rate 227 per million. It was 679 per million for males under age 5---or 679 times Dr. Gordon’s 99.9999% estimate of who would be affected.
So, while he is correct that more than 99% of folks will sail thru this season just fine, his innumeracy re: 99.9999% of us makes me question his reasoning, and hence judgment, on the whole issue. And yes, you read that right, more than 99% of the population will have no problems with this illness. Do keep that in mind when making your vaccinate or not decisions. And he does point out a legitimate issue as to future strains of swine flu, and whether those who have not built anti-bodies naturally will be worse affected in future. But I really don't think it is asking too much of him (or anyone else) to get the numbers right.
Ok, I’m done. Hope that was helpful to someone out there. I had one friend ask me to do a mathematical analysis and come up with a clear answer of whether to vaccinate or not. Unfortunately, this is not possible to do, as first, certain data is unknown, and second, how much you weight various information depends on your personal values, not numbers. I wish everyone out there health, and peace in coming to a decision on this issue that you are comfortable with.
I welcome respectful comments. Feel free to disagree, ask questions, etc. But anyone being nasty will have their comments simply deleted, in the interest of maintaining a place where discussion is welcome.