It's so cute when anti-vaccine activists try to use citations to bolster their arguments. Quite often, it seems as if they simply read the title of a study and leap to the assumption that it supports their notion that vaccines are useless, dangerous or cause autism. In those instances where they actually do get past the title, they either misinterpret the study or misrepresent the results, hoping, perhaps, that whomever they are speaking to won't go through the trouble of actually reading the citation. Actually, in most instances, they probably rely on people not reading their citations at all. It's very impressive to throw out a bunch of study titles and author names. The casual observer is likely to just think, "Huh, they must have a point. I mean, look at all of those studies."
And I can't really blame the average person for taking that approach. It takes a lot of effort and time to actually examine the citations critically. Scientific papers are generally geared toward academics, people in the same field who already have a basic background education. They understand the methods and why certain things were done, while others weren't. They know the various jargon used. The language of science is probably the biggest barrier to a layperson understanding (let alone reading) a study. After all, there are a lot of new, unknown words and, let's face it, study papers are boring (unless you have a keen and obsessive interest in the subject, maybe). But those who deny some scientific concept rely on this to overawe their audience. It allows them to use a tactic known as the Gish Gallop: throw out lots of studies that you claim support your position and depend on your audience not making the laborious effort to see if the studies say what you claim they do.
I encountered this on a small scale just recently in the comments of an article in the Independent Online discussing what happens when vaccine refusal has fatal results. When one commenter claimed that modern measles outbreaks occurred predominantly among the vaccinated, I countered with examples from recent outbreaks in which the majority (or all) of the cases were unvaccinated. A second commenter responded to me with a list of six studies purporting to support the assertion that measles outbreaks disproportionately occur among vaccinated, rather than unvaccinated, individuals. But as expected, the reality is rather different than this person claimed.
1. Acute measles encephalitis in partially vaccinated adults
First on the list was Acute measles encephalitis in partially vaccinated adults (PMID 23967232), a 2013 paper published in PLoS One by Fox, et al. This study examined 15 patients in Vietnam who presented with acute measles encephalitis (AME). Eleven of these reported receiving a single dose of measles vaccine, two had not been vaccinated, and two did not know whether they were vaccinated or not. There were also 13 controls who presented with measles rash and fever, but did not have AME. Of these, only five were asked their vaccination status. Two reported receiving a single dose, two said they were not vaccinated, and one did not know. For most of the subject, vaccination status was determined by asking the patients' parents. Vaccination cards were not examined. All 15 AME subjects eventually recovered fully. The sample is not representative of the entire outbreak population, since it only looked at patients with AME. The subject population was also quite small, making it difficult to draw meaningful conclusions from the study, not to mention several confounding factors that the authors list. They conclude that it is important to maintain a high coverage of a two-dose measles vaccination schedule. The study supports that a second dose is important to prevent spread and severity of the disease, something that has been known for quite some time; a single dose is about 95% effective, while a second dose leads to around 98% efficacy in generating lifelong immunity. The study says nothing about the total number of cases in the outbreak, how many total cases were vaccinated vs. unvaccinated, not what percentage of individuals who were exposed were vaccinated vs. unvaccinated. This study does not support the claim that there are more vaccinated than unvaccinated in modern measles outbreaks.
2. Largest measles epidemic in North America in a decade--Quebec, Canada, 2011
Next up we have Largest measles epidemic in North America in a decade--Quebec, Canada, 2011 (PMID 23264672) from the Journal of Infectious Diseases, 2013, by De Serres, et al. This study examined a superspreading measles outbreak in Quebec. In 2011, 21 different people imported measles to the province, resulting in 725 cases. The primary event examined involved 1 case causing sustained transmission of the virus, resulting in a total of 678 other cases. Just over half of the cases (56%, or around 380 individuals) in this event were adolescents between the ages of 12 and 17. Of those, an estimated 28% (or about 16% of the total number of cases) had received two doses of measles vaccine. The authors note that 2-dose recipients had a milder illness and lower risk of hospitalization than unvaccinated or single-dose recipients. While the abstract does not mention how many cases received one dose or were unvaccinated, let alone the total number of cases that were immunized vs. under- or unimmunized, the Ministère de la Santé et des Services sociaux reports that of the total 776 (725 confirmed and 51 clinical) cases in the outbreak, 615 (or 79%) of the cases were either unvaccinated or had unknown vaccination status. Clearly, there were more individuals in this outbreak who were not vaccinated than were. This study does not support the claim that there are more vaccinated than unvaccinated in modern measles outbreaks.
3. Measles Outbreak in a Highly Vaccinated Population, San Diego, 2008; Role of the Intentionally Undervaccinated
Third in the list made me chuckle, not because of what it reports, which is awful, but because my interlocutor clearly did not read even the abstract. Measles Outbreak in a Highly Vaccinated Population, San Diego, 2008; Role of the Intentionally Undervaccinated (PMID 20308208), by Sugerman, et al. (2010), published in Pediatrics, describes a January 2008 measles outbreak in San Diego. The outbreak was started by a 7-year-old child whose parents had intentionally not vaccinated him. He contracted measles while visiting Switzerland. After his return, he transmitted the virus to his older sister and younger brother. The virus continued to spread, eventually infecting a total of 12 children, including the index case. None of the cases had been vaccinated; the parents of 9 of them had refused vaccination, and the other three were too young to receive the vaccine. Five children required urgent outpatient care, and one child, only 10 months old, was hospitalized for 72 hours, requiring IV fluids due to severe diarrhea. Investigating and containing the outbreak ended up costing taxpayers $10,376 per case. In addition, there were 48 children who were exposed and were too young to vaccinate. They were quarantined, costing their families roughly $775 per child. This study illustrates an outbreak in which every case was unvaccinated, and which came at significant cost to the public and families involved. Well done, vaccine refusers! Needless to say, this study does not support the claim that there are more vaccinated than unvaccinated in modern measles outbreaks. Oh, and did I mention that the index case in this study was a patient of Dr. Bob Sears (mentioned on Episode 370 of This American Life), who developed his own delayed vaccination schedule which he admits is non-scientific? Yep, a darling of the anti-vaccine movement, his advice may have contributed to the largest outbreak the region had seen in 17 years.
4. Two Case Studies of Modified Measles in Vaccinated Physicians Exposed to Primary Measles Cases: High Risk of Infection But Low Risk of Transmission
The 2011 Journal of Infectious Diseases article Two Case Studies of Modified Measles in Vaccinated Physicians Exposed to Primary Measles Cases: High Risk of Infection But Low Risk of Transmission (PMID 21666213), by Rota, et al., was number four on the list. Presumably, this study was chosen because of its title. The "high risk of infection" mentioned is not because the physicians whose cases were presented were at increased risk of infection due to having been immunized (both had received 2 or more doses of measles vaccine), but rather because they were in positions that "are often exposed to patients at a very infectious stage of measles disease". The study did not look at all at the makeup of the outbreaks mentioned, so we do not know anything from this paper about the number of vaccinated vs. unvaccinated in either of the outbreaks. What's interesting about the study, though, is the idea presented by the authors. While not conclusive, these two cases, along with other cases of infection in unvaccinated patients mentioned by the authors, suggest that measles immunity, whether from vaccination or infection, is not absolute; it is not completely immune or completely vulnerable, but rather works on a spectrum. They state that those who may have antibody levels indicative of protection against clinical disease after casual exposure may display mild to moderate symptoms after prolonged exposure to the virus in close quarters, as you'd find in an exam room. The two physicians in the study both displayed robust immune responses after exposure. And although they displayed mild symptoms of measles, neither one transmitted the infection to anyone else, even those who were unvaccinated. That struck me as particularly interesting, since the virus is one of the most contagious diseases on the current vaccination schedule. It is an interesting paper, with potential implications for how we identify cases: by clinical presentation or by lab results. Further research would be needed to shine a light on this issue. However, as with the three previous studies, this study does not support the claim that there are more vaccinated than unvaccinated in modern measles outbreaks.
5. Mild Measles and Secondary Vaccine Failure During a Sustained outbreak in a highly vaccinated population
The fifth paper is the oldest of the bunch, published in 1990 in JAMA by Edmonson, et al.: Mild Measles and Secondary Vaccine Failure During a Sustained outbreak in a highly vaccinated population (PMID 2278542). Here, we have a paper that actually does support the anti-vaccinationist's claim, at least superficially. The paper describes a 1986 outbreak in Dane County, WI, resulting in 218 confirmed cases of measles. Of these, 36 were unvaccinated and 182 had received only one dose of measles vaccine. If we stop there, we miss the nuance and subtleties from the context of the study, and nuance and subtlety have never been strong points among the antivaccine camp. Around 58% of the vaccinated received their vaccine before 1980, which is important; poor handling by physicians and a less effective stabilizer used until 1979 meant that a large portion of the vaccine stock was rendered less effective. A number were also immunized before 12 months of age, which we later learned resulted in greater failure of the vaccine. It should also be noted that the vaccinated patients tended to have milder illness than the unvaccinated, showing that even when the vaccine fails to fully protect, it can still render the disease less dangerous. Also of note, those who had secondary vaccine failure did not transmit the virus to others, just as in study #4 above. The authors conclude that the high incidence of measles in vaccinated individuals may be due to the fact that the virus is so incredibly virulent, combined with the fact that a single-dose schedule may not be sufficient to establish high enough levels of immunity to limit the spread of an outbreak. In context, this study is not really applicable to the current state of
measles prevention. Since then, the vaccine has been reformulated to be more stable and have a higher effectiveness, vaccine storage and handling has improved, and the schedule has changed from a one-dose to a two-dose schedule, improving the odds of developing immunity. On a side note, this study, as well as others, note that immunity did not appear to wane with time. In the end, while this study superficially supports the anti-vaccine claim, it does not apply to the current state of affairs, and thus, it does not support the claim that there are more vaccinated than unvaccinated in modern measles outbreaks.
6. Measles in Geneva between 2003 and 2010; persistence of measles outbreaks despite high immunization coverage
Finally, we have Measles in Geneva between 2003 and 2010; persistence of measles outbreaks despite high immunization coverage (PMID 21968424), published by Delaporte, et al. in Eurosurveillance in 2011. This study reports on several different outbreaks in Geneva, Switzerland, between 2003 and 2010. Two outbreaks in 2003 and 2005 were followed by several smaller outbreaks that occurred in fairly rapid succession from 2007 to 2010. By the end of the period, there had been 161 cases in the region. In the 2003 outbreak, 6% of cases (3 out of 47) were vaccinated. In 2005, 27% of cases (3 out of 11) were vaccinated. During the period of 2007-2010, only 2% of cases (2 of 82) were vaccinated. None of the vaccinated patients had received a second dose of measles vaccine. Three of the 52 patients (6%) in the 2003 outbreak, 4 of 16 cases (25%) in the 2005 outbreak, and 18 of 93 cases (19%) in the 2007-2010 outbreaks required hospitalization. Two of those hospitalized patients required treatment in the intensive care unit. Every patient that was hospitalized due to measles-related complications was unvaccinated. Overall, 16% of patients were hospitalized during the study period. The authors show that robust surveillance, early and effective control measures and high levels of immunization reduce the size and frequency of outbreaks. In each of the outbreak periods discussed, the absolute number of vaccinated patients was less than the number of unvaccinated patients, and none of the vaccinated patients had been fully immunized with a second dose. The lack of hospitalizations among the vaccinated patients suggests that they also had a milder course of illness. In the end, this study does not support the claim that there are more vaccinated than unvaccinated in modern measles outbreaks.
The Bottom Line
Despite their intent to bolster the claim that modern measles outbreaks occur predominantly among the vaccinated than among the unvaccinated, implying that vaccination is not only useless but makes the recipient more susceptible to infection, the anti-vaccine activist utterly failed. Instead, these studies illustrated several different things. First, they provided evidence of the benefits of immunization: fewer cases, lower risk of hospitalization, lower risk of transmission and milder illness when breakthrough does occur. Second, they demonstrate quite clearly what I mentioned at the beginning of this post. This particular commenter very likely did not bother to even read the abstracts, let alone the full papers themselves. They likely stopped at the titles, if they even did that. It's also possible that they merely copied the list from some anti-vaccine web site, accepting them as accurately supporting their position without even bothering to think for themselves and critically examine them.
There are several good lessons here. First, don't simply accept a study just because someone you trust or believe promotes it as supporting your beliefs. Think for yourself. And actually read the paper. Frequently, the abstract gives an incomplete or even skewed image of the study, and the title is even less reliable. You need more of the details if you have any hope of coming to a sound conclusion. And finally, perhaps most importantly, measles most commonly occurs in those who are not vaccinated, rather than among the vaccinated. This should not be surprising, since immunization makes one immune to the disease (assuming, of course, that you respond to the vaccine). Moreover, measles is not a mild disease. It is highly contagious and has a fairly high rate of complications, the most common being pneumonia. The real kicker is that measles could potentially be eliminated entirely through high global vaccine uptake, since humans are the only known reservoir for the virus. If we eliminate it from the wild, we would no longer need the vaccine, just like with smallpox. Instead, what these studies show, and what is borne out by other recent outbreaks, is that the unvaccinated drive the spread of the virus.
So here's to you, anti-vaccine activists! Congratulations on being some of the best friends a disease could have. And this year is on track to have the most measles cases in 17 years, with all outbreaks