Let’s run the vaccine risk/benefit numbers!
March 12th, 2010
While reading another ‘blog, I found - in the comments - one of the worst examples of “bad epidemiology” I have ever seen. I won’t embarass the commenter by name or by quoting, but the gist of their “argument” was that (in the US) the risk of vaccination exceeds the risk of the vaccine-preventable disease.
I think we’ve all heard that before.
There were many problems with their math, not the least of which was being too lazy to look up the actual numbers, but I realised that this was the tip of a much larger iceberg of innumeracy, especially as it pertains to understanding prevalence and risk.
So, to begin at the beginning, let us start with fractions.
Most of the time, risk or prevalence is expressed as a fraction, although it may not always look like a fraction (e.g. 1 in 100 is the same as the fraction 1/100). And - harkening back to our elementary school days - the two components of a fraction are the numerator (the top number) and the denominator (bottom number).
I bring up these apparently irrelevant mathematical issues because, in the world of risk and prevalence, there are two major types of errors: numerator errors and denominator errors (although, sometimes, there are errors of both).
In the example I mentioned above (the ‘blog comment), the risk of contracting a certain vaccine-preventable disease was calculated (wrongly, I must add) by dividing the number of people in the US who contracted that disease in a year by the population of the US. This was then stated - indirectly - to be the risk of an unvaccinated person contracting the disease.
Perhaps you’ve already noticed the error - it’s a denominator error. Since most people in the US are already immune to this disease (mostly by vaccination), the proper denominator would have been the number of unvaccinated people in the US. Let’s see how this changes the numbers:
If we use measles as an example, there were 140 cases of measles reported in 2008 (still compiling and verifying 2009 reports). At the end of 2008, according to the US Census Bureau, there were 300,459,786 people in the US. If we use the incorrect method from the example, that would give a “risk” of contracting measles of 0.47 per million per year (1 in 2,146,141 per year).
That seems a pretty low risk, doesn’t it? It’s a bit higher than your “risk” of winning the Powerball lottery, but still quite low.
But is it accurate? [Hint: No]
The correct way to calculate your risk of contracting measles is to divide the number of reported cases in the US by the number of vulnerable people in the US. By “vulnerable”, I mean those people who haven’t been vaccinated and haven’t had measles. That is a bit harder number to find.
The CDC’s NIS shows that, in 2008, 92.1% of children ages 19-35 months had received at least one MMR vaccination. Going back as far as 1994, that number seems fairly steady - about 90 - 92%. By the age of school entry, that percentage (in the 2007 - 2008 school year) was up to 94.9%. Even if we assume that this percentage doesn’t change, it would mean that - at most - 5% of the population is vulnerable to measles.
But even that isn’t an accurate number, because people born before the measles vaccine was available (1963) - and even the years immediately after the vaccine was introduced - would have gotten the disease if they weren’t vaccinated (it is highly contagious).
By 1968, the incidence of measles had dropped low enough to assume that anyone born after 1968 who was not vaccinated is not immune. So, that means that 5% of the US population age 41 or less is vulnerable to measles. This estimate compares with the value found by Hutchins et al (2004) for measles immunity in 1999, which supports the estimate.
According to the US Census Bureau, there are about 172 million people in the US age 41 years or less, so that gives us - at most - 8.6 million vulnerable people. Now the risk of contracting measles is 140 divided by 8.6 million or 16.3 per million per year (1 in 61,428).
After calculating the risk of contracting measles, we need to calculate the risk of death or serious complications. Measles has a case-fatality rate of 2 per thousand, so the risk of contracting measles and dying of it is about 0.03 per million per year - in the current situation, where 95% of the population is immune.
Other serious complications of measles include pneumonia (about 6% of cases) and encephalitis (1 per 1000 cases). Adding these to the risk of dying brings the total risk of serious complications to 0.8 per million per year. If we exclude pneumonia as a “serious” complication, the combined risk of contracting measles and having a permanent, life-altering (or life-ending) complication is 0.05 per million per year.
The risk of serious complications (i.e. death or permanent disability) from the MMR vaccine (discounting the as-yet-undemonstrated “autism connection”) is less than 1 per ten million doses (1 per million allergic reaction, less than 10% of which are “life threatening” = less than 1 per ten million), which (because the recommendation is two doses) works out to less than 0.2 per million per lifetime. With an average lifespan of 75 years, that works out to less than 0.003 per million per year, so the risk from the disease is over ten times greater than the risk of the vaccine even with 95% of the population immune.
Oh, and by the way - the MMR vaccine protects against three diseases, not just measles. We’ll just ignore that for right now.
And even this approximation doesn’t show the true risk of forgoing just the measles vaccine (let alone the MMR) because we haven’t considered how having a large immune population prevents spreading and how that has limited the number of measles cases reported.
Measles is transmitted from person-to-person, a single infection provides life-long immunity and it has no non-human reservoir and no known long-term carrier or dormant state. In this respect, it is similar to smallpox, polio, mumps, rubella, and many other vaccine-preventable diseases. If it is not transmitted, the measles virus “dies out”. It doesn’t “hang out” in the environment. That is why measles could be eradicated, just as smallpox was.
Currently (since 2000), measles is not endemic in the US, largely because there aren’t enough susceptible (non-immune) people in close enough contact to keep the virus going. Measles in the US is an imported disease that, until 2008, was rarely transmitted beyond the person importing it and any under-age (i.e. less than 2 years old) or immune-compromised people they came in contact with.
Starting in about 2008, the percentage of immune people in the US had slipped far enough that imported cases were able to spread locally in pockets of non-immune people. The August 22, 2008 edition of Morbidity and Mortality Weekly Report (MMWR) details two outbreaks of measles that occured in the US that year. In both cases, the outbreaks occured within groups that did not vaccinate for religious or philosophical reasons and were home-schooled.
This latter point is worth noting - even though these children did not attend a public or private school, they still contracted measles from one another.
Here is a telling statement from the MMWR report:
The number of measles cases reported during January 1–July 31, 2008, is the highest year-to-date since 1996. This increase was not the result of a greater number of imported cases, but was the result of greater viral transmission after importation into the United States, leading to a greater number of importation-associated cases. These importation-associated cases have occurred largely among school-aged children who were eligible for vaccination but whose parents chose not to have them vaccinated. [emphasis added]
As the percentage of non-immune people in the country rises, imported measles cases will spread to more people, further raising the risk of infection to non-immune people and increasing the already large benefit to risk ratio of vaccines.
What this shows is that those people who choose to not vaccinate should - at the least - take precautions against associating with other people who don’t vaccinate. This would help reduce their risk of infection to the levels I calculated above.
Perhaps they should wear some sort of lapel pin, similar to what many fraternal organisations (e.g. Masons, Rotarians, etc.) have. Except, of course, that instead of stepping forward and embracing when they see a fellow member (with or without secret handshake), they should immediately turn about and walk briskly in opposite directions, to avoid transmitting vaccine-preventable diseases.
As the events of 2008 showed us, there will not be a gradual increase in measles spread as vaccine coverage declines - there will most likely be an abrupt increase as the percentage of non-immune people (and their proximity to one another) crosses a critical threshold.
And it is important to note that non-immune people are not just the children of parents who choose not to have them vaccinated. They include children too young to be vaccinated and people who are immune-suppressed due to disease, cancer or genetic disorders. They include the elderly, whose immune systems are weaker, and those people who - for one reason or another - did not develop an adequate immune response to vaccination.
Those who choose to not vaccinate and think they are letting others take the risks for them are fooling themselves; they are taking the greater risk - even now.
Prometheus
Non-hyperlinked References:
Hutchins SS, Bellini WJ, Coronado V, et al. Population immunity to measles in the United States, 1999. J. Infec. Dis.. 2004; 189(Suppl 1):S91–7
Filed under: Critical Thinking, Help for the bewildered
March 13th, 2010 at 3:36 am
You know, I think some people just want to believe that health risks–of various kinds–aren’t out there at all. There’s no reason to vaccinate or otherwise protect against a disease, the disease isn’t a problem! It’s baffling.
There’s a very nasty disease in horses, called Equine Infectious Anemia. It’s spread by mosquitoes (doesn’t require direct contact) and mortality is high. This was very hard to initially isolate and diagnose because some horses carry it without symptoms–but are still contagious. These carriers may travel around and spread EIA, causing horrible outbreaks with high levels of mortality. There is still no vaccine for EIA. There’s no treatment, either.
The disease is controlled by requiring that all horses have a current negative EIA test, which is done once a year. People can’t even argue that there’s a risk associated with vaccination; there’s no vaccine. It’s *just* a blood draw and a test. This test is the only way we have to make sure we don’t have terrible outbreaks any more.
Making sure that EIA carriers don’t interact with other horses has brought the incidence of EIA down until it’s near negligible. A lot of people may never know of a positive EIA horse. So now there are people claiming that EIA is (and always has been) a gov’t conspiracy: EIA was never a problem, EIA is under control, and so EIA testing is entirely a way for the state and the veterinarian to make money together. (The blood test runs about $30 total–sounds like a real moneymaker, huh?)
Just bizarre.
March 13th, 2010 at 11:37 pm
Thanks for this Prometheus. I will get a lot of use out of it in my internet travels.
March 14th, 2010 at 12:19 pm
Mike Adams made a similar error in his recent newsletter where he crowed that, in a recent outbreak, 77% of children who got the mumps, were vaccinated. He thinks that means vaccines increase your risk of getting the disease. As I explained, the reason more vaccinated children got the mumps because there were more vaccinated children to start with.
March 16th, 2010 at 8:59 pm
Skeptico,
Since about 95% of children are vaccinated, the fact that only 77% of the mumps victims were vaccinated tells us that the vaccine is protective.
Lets run the numbers:
If we pick 100 people at random, we would expect 95 who have had their MMR vaccine and 5 who haven’t.
If the vaccine were completely ineffective, we would expect that, out of every 100 victims, 95 would have been vaccinated and 5 unvaccinated. The numbers from the latest mumps outbreak would result in 77 vaccinated and 23 unvaccinated.
The odds ratio for contracting mumps in this outbreak shows that the unvaccinated are 5.7 times more likely to contract mumps than those who were vaccinated.
The Chi Square for those numbers gives a p-value of 0.0005; the Fisher exact test gives a p-value of 0.0003.
In other words, there is a highly signficant difference in mumps incidence between the two groups. And that difference is in the direction of the vaccine being effective.
Now, the mumps component of the MMR is notorious for being the least effective one of the three. Even so, the mumps vaccine reduced the risk of contracting mumps in this outbreak by over 82%.
Prometheus
March 22nd, 2010 at 12:24 am
Do you know whether the measles vaccine requires a booster shot to retain its effectiveness over time? That seems to be the case with the rubella shot–often a woman who received adequate immunization as a child will find out during routine pregnancy screening that her rubella titers are undetectable, and therefore it is recommended that she get a booster shot after delivery. I have heard some physicians speculate that when overall vaccination levels were lower, there was enough passive exposure to the disease in the general community to act as a natural booster shot, but now that is no longer the case.
March 23rd, 2010 at 12:27 pm
mHbot
Sorry, this has nothing to do with your post but I wanted your opinion on this:
http://www.youtube.com/watch?v=lvuax9yGLLA
Are mHbot chambers dangerous or not?
March 27th, 2010 at 10:47 pm
The YouTube video - for those who haven’t seen it - shows two people in a “soft” chamber pressurised to 1.29 atm and with a (corrected) oxygen concentration of 27% when measured at the site of the “experiment”.
It’s not surprising that they didn’t get much increase in flammability with such a modest increase in oxygen concentration. However, if they had let the oxygen run a bit longer and build the concentration up a bit more, the results might have been more….dramatic.
The “soft” chambers are pretty safe….if they are used as directed. The FDA has specifically directed that they are not to be used with supplemental oxygen, since the material of the chamber will support combustion.
Again, this is a risk:benefit issue.
The risk of a “soft” hyperbaric chamber - at or below 1.3 atm - is minimal. Even adding a modest amount of supplemental oxygen probably only increases the risk (per use) slightly (although that one time that a fire does occur will be very catastrophic).
The benefit - as a treatment for autism - however, has been shown to be indistinguishable from zero. Thus, the risk:benefit ratio is highly skewed toward risk and away from benefit, even though the individual risk is very small.
If the “O2 MythBusters” want to redo their “experiment” to simulate the way some practitioners really use these chambers “treat” autism, they should wait until the oxygen concentration is higher. On second thought, we already have YouTube videos of the Hindenberg, so that might be redundant.
Prometheus
March 27th, 2010 at 10:55 pm
Jane,
The MMR requires a “booster” vaccination at 4 - 6 years in order to maintain adequate immunity in adults. As you surmised, this is probably due to the drop in wild-type measles and rubella exposure.
Before vaccine coverage was widespread, vaccinated people were periodically exposed to the wild-type viruses, which acted as a “natural booster”. Now that measles, mumps and rubella are rare, we need a “booster” vaccination.
Of course, if the anti-vaccination movement gains sufficient popularity, the wild-type measles, mumps and rubella will come back to levels sufficient to “boost” the immunity of the vaccinated population and “booster” vaccinations may no longer be needed.
Prometheus
April 2nd, 2010 at 12:26 am
Prometheus,
The 2nd MMR shot is actually not a “booster”, it was implemented when it was discovered that a small minority (2-5%) of children do not build antibody response to the first shot.
From the CDC-
http://cdc.gov/vaccines/vpd-vac/combo-vaccines/mmr/faqs-mmr-hcp.htm
Why is a second dose of MMR necessary?
About 2%-5% of persons do not develop measles immunity after the first dose of vaccine. This occurs for a variety of reasons. The second dose is to provide another chance to develop measles immunity for persons who did not respond to the first dose.
April 2nd, 2010 at 12:28 am
sorry I left off part of the quote from the CDC page:
If you can give the second dose of MMR as early as 28 days after the first dose, why do we routinely wait until kindergarten entry to give the second dose?
The second dose of MMR may be given as early as a month after the first dose, and be counted as a valid dose if both doses were given after the first birthday. It is convenient to give the second dose at school entry, since the child will have an immunization visit for other school entry vaccines. The risk of measles is higher in school-age children than those of preschool age, so it is important to receive the second dose by school entry. The second dose is not a “booster”; it is intended to produce immunity in the small number of persons who fail to respond to the first dose.
April 2nd, 2010 at 1:17 pm
Me got moi’s shots, fur sure.
April 2nd, 2010 at 2:39 pm
Ms. Sears,
That’s why I put “booster” in inverted commas. The second MMR has two purposes: first, to “pick up” those who didn’t develop an adequate response to the first MMR and, second, to “boost” the immunity of those whose antibody levels had “drifted” lower.
If measles, mumps and rubella were still endemic, there would be no need for a second MMR because everybody would be exposed to the wild-type viruses, which would provide either a “boost” or a primary infection.
Much has been said about how the immunity provided by the MMR vaccine strain viruses is not as good or as long-lasting as that provided by the wild-type viruses. This is true.
However, there would be little point in having the vaccine strain as virulent as the wild-type strain, so we have to put up with a small (for measles and rubella) failure rate in return for way less (over ten thousand times less) morbidity and mortality.
Prometheus
April 2nd, 2010 at 6:41 pm
Are those numbers (10 thousand times less) for the US or globally? Why does the current schedule only have 2 shots of MMR, if this is an established issue, should there not be at least a couple more boosters? Is the mumps outbreak in NJ/NY, at least in part, because of this waning immunity? How long exactly does the MMR last? As a 28 year old am I vulnerable to infection with any of these if I travel to a place where they are endemic? Also I am curious whether your argument applies to all of the vaccines on the schedule. Thank you for your time.
April 4th, 2010 at 2:14 pm
Ms. Sears,
The morbidity and mortality from measles is lower in the developed world because of better nutrition, better sanitation and better medical care. Let me explain:
Like all illnesses, measles causes more morbidity and mortality in the undernourished. Some studies have suggested that vitamin A may be especially important in measles.
Wild-type measles causes a significant drop in immune function, which - in areas with poor sanitation - generally leads to secondary diarrheal illnesses that can be fatal when the measles is not.
The most common complication of measles is pneumonia. In areas where modern medical care is readily available, pneumonia is a serious but treatable complication. In places with minimal medical care, pneumonia is a death sentence.
In short, the measles vaccine is 10,000 times safer than the disease in developed countries. In undeveloped countries, the vaccine has a much better safety margin because the disease is so much more dangerous in those areas.
As for the number of “boosters” - data so far suggests that two jabs is sufficient. Not “100% effective” (nothing in medicine is 100% effective) or “100% safe” (ditto), but “sufficient” to prevent widespread outbreaks with the least number of “jabs”, given the levels of vaccine uptake at the time the recommendations were made.
The mumps outbreak in NJ/NY is due to two factors: inadequate uptake of the MMR and the inherent poor immunogenicity of the mumps vaccine.
While the measles and rubella component of the MMR provide effective immunity in about 90+% of the people who receive a single injection, the mumps component is only about 60 - 70% effective. There are mumps vaccine strains that are more effective, but their complication rate (generally viral meningitis) is much higher (although still over a thousand times less than the wild-type mumps).
So, the current mumps vaccine is a compromise between efficacy and safety, with the US (and most developed countries) opting for less efficacy and more safety. This “worked” for many years because mumps was no longer endemic in the US and other countries. However, with MMR uptake falling (esp. in the UK), there is a risk that mumps will become endemic again, requiring a re-thinking of vaccine policy.
If you are planning a trip to a country where measles, mumps or rubella are endemic, it would be safest (and easiest) to get another “booster”. The armed forces rountinely re-vaccinate adults with the MMR for this reason.
As for my “argument” - if you are referring to the risk of the disease exceeding the risk of the vaccination even when the disease is not endemic, I would have to “run the numbers” to be sure.
For many of the vaccines (tetanus, pertussis, diptheria, HiB, HepB, meningitis, chicken pox), this would be a moot point, because the diseases are still endemic and the vaccines are definitely safer than the disease. Mumps appears to be “trying” to become endemic again, as has measles, at least in the UK.
The polio vaccine was changed from oral to injected (going back to the “older” vaccine) because the risk of polio had declined to the point where even the miniscule risk of the oral polio vaccine was excessive. There have been no native cases of polio reported in the US, but vaccination continues because of the risk of reintroduction.
Prometheus
April 5th, 2010 at 1:25 pm
How many cases of imported polio have been reported? The IPV does not prevent transmission does it? Just the OPV, right? So how is that helping the vulnerable populations that are not vaccinated (babies, medically ineligible, by parental choice)? Which vaccines have been shown in studies to prevent transmission? Has pertussis? I have not seen a study showing that pertussis transmission is prevented, even though there is a campaign saying just that. If you have seen one I would be interested in a link or reference. I read somewhere several vaccines do not prevent transmission, it seems that would be important to the herd immunity theory, wouldn’t you think? Especially since we are a global society with the risk of importation. I am pretty sure that the meningitis vaccine also falls into the category of vaccines that does not prevent transmission, likewise the obvious one, tetanus, which is not infectious. Diphtheria is still transmittable, despite vaccination, right? So when we are discussing herd immunity shouldn’t this also be discussed? Has this risk been addressed by the ACIP? I also have read that the uptake of Tdap by the adult population is around 2%. It is my understanding (possibly flawed, feel free to correct me as I am no expert) that the reason we need boosters for that vaccine is that it wanes in effectiveness, specifically the pertussis and tetanus components (not sure about diphtheria). So that is a large portion of the “herd” that is possibly vulnerable as well (and potentially disease spreaders). Do you believe this pattern of non compliance will be a factor with other vaccines that may prove to need adult boosters? Like chicken pox, for instance? Isn’t it more dangerous for an adult to be unprotected against chicken pox than say a 5 year old? Again, thank you for your time.
April 5th, 2010 at 1:36 pm
I was also curious, when you say you would need to run the numbers, what is your source for those numbers? Specifically the vaccine attributed injuries? That is where I run into an issue when I am trying to do an analysis, due to the known issue of underreporting (I have even seen an estimate that only 10% of vaccine reactions are reported to VAERS).
April 6th, 2010 at 6:51 pm
Ms. Sears,
The last case of polio reported in the US - to the best of my knowledge - was in an unvaccinated young man in 2005. He had just returned from a polio endemic area. The last year for which the CDC has released complete data is 2007 - there were no cases of polio reported that year, imported or domestic.
The IPV cannot prevent the (generally transient) intestinal carrier state for polio, which is why OPV is needed to completely eradicate polio. However, in the US, the risk of contracting polio is essentially nil, so long as polio immunisation rates remain high. Unvaccinated individuals are protected by the “firewall” of vaccinated individuals around them.
Pertussis vaccination has been shown to reduce the rate of nasal carriage of Bordetella pertussis, but the bacteria is found in healthy vaccinated individuals, which is why vaccination is critical. Vaccination prevents the development of the disease pertussis, which, with its explosive cough, spreads large amounts of the bacteria into the environment.
Diphtheria vaccine is a toxoid (like the tetanus vaccine). It is directed at the disease-causing toxin and not at preventing carriage of the causative organism (Corynebacterium diphtheriae). The same is true about tetanus - it is directed at the tetanospasmin toxin, not Clostridium tetanii.
So, yes, some vaccines don’t prevent spread of disease, per se. For tetanus, the causative organism is a soil bacteria, so we would have to “vaccinate” the soil to eradicate it or to provide “herd immunity”.
For other diseases, it is difficult to prevent spread, since the organism (usually a bacteria) is able to establish chronic carrier states or is transferrable from other species. In those situations, the vaccine protects only those who are vaccinated. The meningococcal vaccine is a good example of these vaccines. Up to 15% of adults carry Neisseria meningitidis in their nose and mouth; the vaccine prevents meningitis and other invasive disease caused my this organism but does not end or prevent the carriage of the organism.
“Chicken pox” (herpes zoster) is definitely more dangerous as an adult - if that is your first (primary) infection. Even if the immunity is allowed to wane (i.e. no “boosters”), the wild-type infection would be seriously attenuated. Also, getting “boosters” as an adult has been shown to reduce the risk of shingles, a recurrence of herpes zoster that carries of high risk of causing chronic neuropathic pain.
Finally, I don’t use the VAERS database for anything. It is hopelessly contaminated with coincidental, unsubstantiated and lawyer-generated reports. My data on vaccine-attributed injuries comes from published studies of the vaccines’ efficacy and safety.
Prometheus
April 7th, 2010 at 7:01 pm
Thank you for your answer. So what is the idea behind the campaign to vaccinate parents for pertussis to protect their baby? I have seen Jennifer Lopez (actress) in an advertisement telling parents to vaccinate themselves to protect infants. So is there any scientific evidence (a study published in a reputable peer reviewed publication) that says this is effective?
April 8th, 2010 at 12:18 am
Ms. Sears,
I would suggest that you read:
Ward JI, et al. Bordetella Pertussis infections in vaccinated and unvaccinated adolescents and adults, as assessed in a national prospective randomized Acellular Pertussis Vaccine Trial (APERT). Clin. Infect. Dis. 2006 Jul 15;43(2):151-7.
That should answer all of your questions about why there is a campaign to vaccinate adults in order to prevent pertussis in infants.
The short answer is that carrying pertussis in your nose presents a low risk of transmission. Coughing pertussis bacteria all over the place while your upper respiratory tract is loaded with Bordetella pertussis presents a high risk of transmission. Preventing symptomatic disease greatly reduces the risk of transmission.
Another problem with the pertussis vaccine - one it shares with many bacterial vaccines - is that the immunity doesn’t last for life (BTW, even having symptomatic pertussis doesn’t give you life-long immunity).
One of the many concerns with pertussis is that it is so dangerous to very young infants, infants too young to vaccinate (the smaller the airway, the greater the danger). Adults, especially adults who have been vaccinated as children, can have a milder form of pertussis that looks like just a bad “cold”. However, each cough and sneeze sprays Bordetella pertussis all over the place.
In short, Jennifer Lopez is correct. There’s a first time for everything.
Prometheus
April 10th, 2010 at 7:54 pm
Thanks for your answer.
April 14th, 2010 at 7:43 am
Great blog. I hope you don’t mind if I repost this over at http://factsnotfantasy.com Since you have been over to Bad Astronomy, I will assume you’ve seen Todd W.’s and my efforts to debunk many of the outright LIES told by the antivax movement.
April 14th, 2010 at 8:13 am
Great post, Prometheus! Mind if I work it (or at least a link) into antiantivax.flurf.net?
April 14th, 2010 at 8:54 pm
If you want to re-post, feel free to, as long as you give credit and make no significant changes.
I’ve dropped by both factnotfantasy.com and antiantivax.flurf.net and they’re both great sources of reliable information for anyone who wants it.
Prometheus
April 14th, 2010 at 9:54 pm
@Prometheus
Added a bit of text under Vaccines in General with a link to this article. I figured you did such a good job explaining it, that I would just let your words speak for themselves, rather than me trying to interpret and summarize them.
April 15th, 2010 at 9:32 pm
Prometheus,
You have made an error very similar to the one you critique. In your case, the problem is the numerator. After correctly noting that different people don’t have the same probability of contracting measles, you split the population into “vulnerable” and “protected”. You place all vaccinated individuals in the “protected” group. At this point you erroneously claim that all 140 measles cases should be used as a numerator for the “vulnerable” group. You thus assume that the “protected” group is 100% protected and all measles cases were in the “vulnerable” group. This implies there wasn’t a single measles case in vaccinated individuals.
We know this is not true. Looking at the wikipedia entry for measles, out of 131 cases reported by July 30, 2008, there were 11 cases in patients who received at least one doze of the measles vaccine. While the vaccine is highly effective, it is not 100% effective. You need to make the appropriate adjustments to the numerator. You also need to factor in the risk of getting vaccinated and still getting measles when you evaluate the overall benefit of vaccination.
April 16th, 2010 at 8:23 pm
Good point, BisserR.
If I remove the 11 cases of measles in people vaccinated at least once, that reduces the risk of contracting measles if unvaccinated (in 2008) to 15 per million, which, in turn, reduces the risk of death or serious complication from measles to 0.045 per million per year in unvaccinated individuals (from 0.05).
Since the risk of death or serious complication from the MMR vaccines (which, you will recall, also protects against two other diseases) is 0.003 per million per year, the vaccine is still over ten times safer than going unvaccinated.
Thanks for your correction. It doesn’t change the result, but I do like to be as accurate as possible.
As for the “risk of getting vaccinated and still getting measles”, complications from measles are significantly lower if you’ve been vaccinated, so even if you get the rash (and are diagnosed with measles), your risk of getting pneumonia (which I excluded as a “serious complication”, if you recall) is almost zero and the risk of encephalitis or SSPE is essentially zero.
If we are to use “getting the measles” as the benchmark for risk, I think that the numbers speak pretty convincingly there, too. In 2008, there were 140 cases of measles and - as you pointed out - 11 of them were in people who had received “at least one dose of the MMR”.
So, 92% of the cases were in the 5% of the population that hadn’t been vaccinated and 8% were in the 95% of the population that had been vaccinated. That’s a highly significant difference. (p less than 0.0001).
To be fair, we don’t know for sure how many of the unvaccinated group were too young to have been vaccinated, so it’s not that easy to calculate the expected percentage of vaccinated individuals. However, MMWR reports from 2008 indicate that most of the unvaccinated cases occured in people who were old enough to have received at least one dose of MMR vaccine.
Again, thank you for pointing out an error in my calculations.
Prometheus
April 19th, 2010 at 8:41 pm
You are welcome, Prometeus,
What surprised me a bit was how many people commended and thanked you for a great post without noticing a rather obvious omission. So much about the anti-vaxers being non-scientific
Seriously though, many people seem unwilling to accept that there are many reasonable people with solid scientific background who question our current vaccination policy. One can easily find silly examples like the Mike Adams post quoted by Skeptico above, and then say that “anti-vaxers are unscientific”. Conversely, one can focus on the fine print of any real-life epi study and quesiton its honesty and conclusions.
It is sad this debate has become so political and irrational at the expense of our children. Even though I may sometimes disagree with your conclusions, I always enjoy reading your posts. So thank you for another great post, and let’s hope for a future when people are not labeled as “anti-vaxers” or “vaccine pushers”.
April 19th, 2010 at 11:38 pm
BisserR,
Before you get too gleeful about “obvious omissions” and “anti-vaxers”, I’d like to point out two things:
[1] The omission was not necessarily obvious to the readers of my post. I’ll grant that I should have investigated the MMWR numbers closer to see how many of the measles victims had been vaccinated, but that is my error and not that of the readers. Put your smug away for now.
[2] The “obvious omission”, you’ll note, did not change the outcome significantly. Getting vaccinated is still over ten times safer than “going bare” - even in 2008.
I will admit that I, too, have questions about “our current vaccination policy”, but I don’t make the unsupported claims and outrageous assertions that the “anti-vaxers” do. It is, indeed, possible for “reasonable people with solid scientific background” to question and discuss. However, I haven’t heard any of those “reasonable people…” called “anti-vaxers”.
If someone is opposed to vaccines (or a particular vaccine) for irrational or unsupportable reasons, I think it is perfectly reasonable to call them “anti-vaccine”.
The ongoing effort to find some way to blame vaccines for causing autism is a perfect example. There is no data supporting the assertion that vaccines are associated with autism, let alone cause autism. The “vaccines-cause-autism” goalposts have shifted from mercury (thimerosal) to the measles vaccine strain to “toxins” (always a sign of non-scientific objection) to “too many, too soon”. Who knows where the goalposts will be tomorrow.
I have no objection to labels as long as they are accurate. Not everyone who questions vaccines is “anti-vax”, just as not everyone who questions the reasons people give for opposing vaccination is a “vaccine pusher”. However, some are.
My policy is - as far as possible - to “call a spade a spade”. If the shoe fits, I’ll probably ask you to wear it.
Prometheus
April 20th, 2010 at 2:34 pm
Prometheus,
I would suggest you reconsider your claim that there is no data showing positive correlation between autism and immunizations. Prevalence data from the ADDM network showed 57% increase in autism for children born in 1996 compared to children born in 1992. Hepatitis B immunization for infants was introduced in 1993 and reached coverage of 82% by 1996, accoring to WHO data (see http://www.who.int/immunization_monitoring/data/usa.pdf). This is a positive correlation.
This correlation is actually a good reason to spend time and money for a vaccinated vs. unvaccinated populations study. You have opposed such study in your blog in the past. Would you reconsider your position in light of this data?
Regarding your claim of “shifting goalposts” in the search of a possible vaccine-autism link, this is the way science works. You start with a plausible hypothesis and test it with the available data. If you are unable to fully resolve the issues under investigation you refomulate your hypothesis or state a new one and proceed with more testing.
Are there “shifting goalposts” in genetic autism research? Many genes have been investigated and only a few have shown positive results, and they can only explain a small portion of autism cases. While it would be great if the first targeted gene could explain everything, this is rarely the case and “goalposts” need to shift and new genes need to be tested.
New theories of autsim ethiology will continue to spring until we have a satisfactory explanation. Some of these theories may involve immunizations, some may not. It is possible that in the end immunizations will turn out be responsible for a small portion of autism cases, e.g. comparable with the number of fragile X cases. Or maybe not. We do not know yet, as we don’t have good methods to identify small subgroups of people who may be vulnerable to vaccine reactions.
April 20th, 2010 at 7:22 pm
BisserR comments:
Well, I am disappointed.
Let’s look at those ADDM network (created in 2000) 2000, 2002, 2004 and 2006 numbers.
ADDM Autism Prevalence:
2000: 6.7 per 1000 8 year old children
2002: 6.6 per 1000 8 year old children
2004: 8.0 per 1000 8 year old children
2006: 9.0 per 1000 8 year old children
Funny how there was so little change between the cohort born the year before infants started to receive the Hepatitis B vaccine and the cohort born the year after. Also note that there was a rise in autism prevalence between the group born in ~1996 (when the coverage had reached 82%) and the group born in ~1998 (when the coverage was about 90%) While it is possible that the jump between 2002 and 2004 was a result of the Hepatitis B vaccine, the jump between 2004 and 2006 seems to contradict that. Either way, this isn’t exactly a “smoking gun”. There were, after all, other things that changed during that time.
The “shifting goalposts”, by the way, have to do with the way that the vaccines or vaccine components that allegedly cause autism have shifted, although the “root cause” (i.e. vaccines) has not changed. A real scientist would, at some point, ask whether anything in vaccines was associated with autism, given the number of vaccines and vaccine components that have been “ruled out” to date.
Rather than accept that the data don’t support an association between vaccines and autism, BisserR (and many other people) persist in assuming that we need to “rule out” vaccines as a possible cause of autism. This makes no logical sense, as there is no indication, apart from their persistent assertions, that vaccines do cause autism.
Meanwhile, if you’re looking into possible “environmental” (in the loosest sense) causes of autism, there are literally thousands of chemical, biological and societal “exposures” that have increased at the same time as autism prevalence. Having had no success finding an association between vaccines and autism, why not look at some of the other changes in the environment.
Perhaps that is too radical an idea for some people, but a lot of scientists I know (about 99%) think that way: if one line of hypotheses keeps failing, it’s time to look for a new approach.
I don’t reject the idea that vaccines might cause a small percentage of autism, I simply don’t see that as a tenable hypothesis, given the present data. I’ve even (repeatedly) offered my not-inconsiderable services to research the “vaccine-autism connection”, if some person or group will provide the necessary funding.
No offers so far.
The hard, cold fact is that no research funding agency or institution is going to put up the money to do a “vaccinated vs unvaccinated” study because it is almost certain to show no association and because there is lots of other research, with a higher liklihood of advancing our knowledge, that needs the funding.
In fact, BisserR hits the nail on the head when he/she says “…we don’t have good methods to identify small subgroups of people who may be vulnerable to vaccine reactions.” Absolutely! So doing a “vaccinated vs unvaccinated” study would be worthless. Until we know what autism is (which we don’t), we have little hope of finding “small subgroups”.
And before BisserR (or someone else) accuses me of “not caring if children are made autistic by vaccines”, let me put it this way:
Are we going to stop vaccinating children against infectious diseases simply because they might cause a small number of cases of autism? We already know that many of the vaccine-preventable diseases cause a significant number of cases of permanent mental and physical disability. Should we accept a known number of cases of post-measles, post-mumps, post-rubella, post-etc. disabilities (not to mention the deaths) so that we might be free of the fear of a possible small number of vaccine-caused autism cases?
Does this make sense in anybody’s universe?
Here’s what I suggest. BisserR and/or other like-minded individuals can either raise the money or petition Congress for funding in order to do a “vaccinated vs unvaccinated” study. They will find either no association or (as I’ve said many times before) an association between autism and being unvaccinated (since subsequent siblings of an autistic child are more likely to be autistic and more likely to be unvaccinated).
It’s predicatable. And that’s why nobody wants to do that kind of study.
Prometheus
April 20th, 2010 at 9:47 pm
Prometheus,
The data shows positive correlation. This is a mathematical fact. The size of the correlation changes over time, but the trend is stable and sizable over the period 2002-2006 (birth years 1994-1998).
Be careful with splitting hairs on the 2000-2002 vs. 2002-2004 vs. 2004-2006 jump in rates. The recommended schedule of HepB vaccine is 3 doses over a 6-month period. Frequently people will take longer than recommended or will not get all dozes. I believe the WHO numbers show “at least one doze”. It is likely the number of dozes received was also increasing over these years. So here are the numbers.
ADDM autism prevalence
2000 (born 1992): 6.7 per 1000 8 year old children
2002 (born 1994): 6.6 per 1000 8 year old children
2004 (born 1996): 8.0 per 1000 8 year old children
2006 (born 1998): 9.0 per 1000 8 year old children
HepB vaccine coverage
1993: 16%
1994: 42%
1995: 68%
1996: 82%
1997: 84%
1998: 87%
1999: 88%
Did I ever claim there is causation? I did not. There are indeed other things that changed in this time period that need to be considered. But it is worth noting that there was something very significant about the HepB vaccine recommendation. This was the first time any vaccine was given in large numbers to babies on the day they were born. They were innoculated within hours after their birth when their immune systems are completely immature and they are still under significant stress from the birth.
You mentioned that younger siblings of autistic children are frequently unvaccinated. A comparison of the vaccinated vs. unvaccinated younger siblings of autistics would require a significantly smaller sample size than a comparison in the general popualtion. And clearly, having family members with autism will need to be controlled for. I am under no illusion that such study would be easy or cheap or will yield clear and undeniable conclusions. And yes, there are people who are lobbying for such study.
April 20th, 2010 at 11:06 pm
BisserR,
I did a quick run with the numbers you provided and the correlation doesn’t reach statistical significance (p=0.15). Perhaps the ADDM 2008 numbers will clear it up.
You seem to attach a great deal of significance to the supposed “immature” immune system of the newborn as well as the “significant stress” of birth. It is as though newborns prior to 1993 had never been exposed to antigens.
Here’s something to ponder. Since the evolution of the hepatitis B virus to infect humans, a number of newborns have been exposed to not just the hepatitis B surface antigen (the “active” component of the vaccine) but to the entire, replicating virus.
We have quite a lot of data about newborns who have been infected with hepatitis B at (and before) birth and there hasn’t been any mention of autism. No, autism prevalence wasn’t any higher in children infected with hepatitis B at birth than in the rest of the population.
Funny, that.
Of course, it is possible that being exposed to the hepatitis B core and e antigens in addition was protective against the autismogenic properties of the surface antigen, but that seems a bit of a stretch.
The hepatitis B vaccine isn’t given to infants on the first day of life simply because it is convenient or expedient; it is given then because there is a very narrow window of time in which mother-to-child transmission of the virus can be prevented by vaccination.
A hepatitis B infection can be transmitted while the mother is still in the early undetectable (even by blood tests) stage. And although the mother may know whether she has engaged in any “risky” behaviors, she may not know about the “risky” behaviors her sexual partner has been engaging in.
If an adult contracts hepatitis B, they have a 5% chance of developing a chronic hepatitis B infection (which often leads to cirrhosis or liver cancer). A newborn, on the other hand, has a 95% chance of developing chronic hepatitis.
These are the reasons for vaccinating infants against hepatitis B on the first day of life, and they are very compelling - at least, they are to me.
I’m not trying to make sport of those people who have sincere concerns about vaccines and autism, but I am trying to make it clear why I find their arguments unconvincing. If it makes people feel better to imagine that I am mired in “fixed dogma” or that I am a “Pharma shill”, well, whatever gets them through the day.
I am very willing to be convinced, but I need data to do that.
Prometheus
April 21st, 2010 at 12:57 pm
“I did a quick run with the numbers you provided and the correlation doesn’t reach statistical significance (p=0.15). Perhaps the ADDM 2008 numbers will clear it up.”
Can you elaborate here? What data did you use? What hypothesis did you test to get p=0.15? What does this level of significance mean in terms of evidence for or against the hypothsis that you tested?
“We have quite a lot of data about newborns who have been infected with hepatitis B at (and before) birth and there hasn’t been any mention of autism. No, autism prevalence wasn’t any higher in children infected with hepatitis B at birth than in the rest of the population.”
Can you give a referrence to the data that showed autism prevalence wasn’t any higher in children infected with hepatitis B at birth than in the rest of the population?
April 21st, 2010 at 5:42 pm
BisserR,
I did a basic linear correlation, using the assumption that, since both “presence of autism” and “received hepatitis b vaccine” are binary variables, there is little reason to consider more complex correlations.
Using the ADDM numbers for autism prevalence and the numbers you so kindly provided for the hepatitis B vaccination coverage, the results indicated that there was a 0.15 probability that the relationship found could be due to random chance. Given that the generally accepted standard for medical/biological studies is p less than 0.05, this isn’t statistically significant.
There are more complex analyses that I could do, but I don’t want to turn this into another full-time job: I already have one of those. And frankly, the quality of the data don’t warrant much more precision or complexity. Maybe when somebody finally funds the “vaccinated vs unvaccinated” study, we’ll get better data.
You are correct that I shouldn’t have said “…autism prevalence wasn’t any higher…” - I should have said “…there isn’t any indication that the autism prevelance was any higher…” However, in the entire PubMed and Web of Science literature on neonatal hepatitis B, there isn’t a single case report of autism. This may be an oversight, but I’d have thought someone would have reported if kids with neonatal hepatitis B had a significantly higher prevalence of autism. Fortunately, newborn vaccination is reducing the incidence of neonatal hepatitis B infection, so that data might be harder to collect right now.
We might as well get it out in the open that the ADDM data are probably not the “gold standard”, either, since they rely heavily on children being identified by the educational and medical system as being autistic, thus making them vulnerable to selection biases.
One example of how this may have impacted the ADDM data is the percentage of autistic 8 year olds who had intellectual impairment. If the 8 year olds identified as autistic in 2004 had the same disorder as those identified in 2006, you would expect - with such a large sample - that the percentage with intellectual impairment would remain fairly constant. However, it dropped from 43.8% in 2004 to 41.0% in 2006. Not a big change, perhaps, but possibly indicative of a “drift” in the population identified as “autistic”.
Given all the problems identified with gathering data on autism prevalence, I think it is still too early to see the ADDM as definitive, since the network is still expanding. We may never be able to get “clean” historical data on autism prevalence.
Prometheus
April 21st, 2010 at 7:13 pm
Prometheus,
It looks like you based your claim about hepatitis B and autism rates on the lack of data. You reasoning of “someone would have reported” defies everything you preach about data and science based evidence.
You also show lack of understanding of basic statistics in your analysis of hepatitis B coverage and autism rates. You are looking for statistical significance with 4 data points. The correlation is 0.85, yet you dismiss it as “not statistically significant”.
April 21st, 2010 at 8:35 pm
BisserR,
While “absence of evidence is not evidence of absence”, most people would accept that the lack of case reports of autism or developmental delay, given the vast literature on neonatal hepatitis B - while not data - is suggestive. It also begs the question of why hepatitis B vaccine could cause autism while it hasn’t been reported in children with neonatal hepatitis B.
If I remember correctly, I wasn’t the one drawing the conclusion that the hepatitis B vaccine causes autism based on two data points. The four data points give a correlation coefficient (r) of 0.87 which, with just four data points, yields a two-tailed p-value of 0.13. Yes, the correlation is good, but with only four data points, it’s not quite up to statistical significance.
I’m not arguing that four data points will yield useful information - I’m arguing quite the opposite. Four data points - especially with the acknowledged limitations of the ADDM data - are too few to draw a conclusion or generate meaningful statistical significance, even though you can “run the numbers”.
If I may recap, my assertion is that there is no reason to suspect that any vaccine causes autism. Your assertion was that the ADDM data implicates hepatitis B vaccine. I’ve tried to show - within my limitations - why this hypothesis isn’t biologically plausible and why the ADDM data it relies upon is not reliable.
Maybe we need a “time out”.
Prometheus
April 22nd, 2010 at 12:22 pm
“If the 8 year olds identified as autistic in 2004 had the same disorder as those identified in 2006″
There is a brief discussion about this on The Mystery of autism.
Time out
April 28th, 2010 at 7:26 pm
If measles makes a comeback in developed countries, there will be many victims who are no longer children and are likely to have more severe complications.
May 24th, 2010 at 8:04 am
[...] consent, with a science-based discussion of the benefits versus the risks of vaccines, which is overwhelmingly in favor of vaccinating being safer than not vaccinating. What such groups mean is an “informed consent” where parents are [...]
June 4th, 2010 at 8:47 am
[...] today when the diseases have become so much less common, it is still safer to have the vaccine than risk getting measles, for example. This is demonstrated brilliantly in that example for the US, I’m working on getting the [...]
August 7th, 2010 at 10:17 pm
i thought this quote was really interesting and goes against what a lot of pepole i know are saying around the net that dtap cannot prevent transmission. are there studies or something out there that have found this to be true? something to show that proves the point?
“The short answer is that carrying pertussis in your nose presents a low risk of transmission. Coughing pertussis bacteria all over the place while your upper respiratory tract is loaded with Bordetella pertussis presents a high risk of transmission. Preventing symptomatic disease greatly reduces the risk of transmission. “
August 9th, 2010 at 6:27 pm
Ann,
The issue of an asymptomatic carrier state for Bordatella pertussis has been debated for years. Linmann, Bass and Smith (1968) found an extremely low carrier state (by bacterial culture), even though they tested during a pertussis outbreak. Long, Welkon and Clark (1990) showed - by ELISA testing - that 83% of family members of a clinically apparent pertussis patient were infected and that 66% of these had no symptoms. More recently, PCR techniques have greatly improved the sensitivity over and above that seen with ELISA techniques, with a corresponding increase in the number of asymtomatic B. pertussis infections [for example: Nelson et al (1997)].
The role of vaccination in reducing nasal carriage of B. pertussis has been pretty well established. Ashworth et al (1982) showed that vaccination (of rabbits) allowed them to rapidly clear the bacteria from their nasal mucosa after inocculation. Unvaccinated animals progressed to symptomatic infection. Most recent research has been on nasal vaccines, which show great promise in eliminating asymptomatic carriage of B. pertussis.
While recent vaccination clearly reduces nasal carriage of B. pertussis, lower levels of immunity - adequate to prevent symtoms but not enough to prevent nasal infection - should also reduce spread of the disease by preventing the cough of pertussis, which spreads oral and nasal droplets to other people. I haven’t seen any studies of this (it would be difficult, at best, to study), but it is pretty much “common sense”.
I hope this helps.
Prometheus