The parrot is still dead, and MMR still doesn’t cause autism

So busy was I writing about America’s quack Dr. Mehmet Oz and, of course, the FDA hearing on regulating homeopathy that I didn’t take note of a story that came out the other day examining a study looking at the association between MMR vaccination and autism. More correctly, the study examines the lack of association between MMR and autism because that’s what every well-designed study that’s looked for such an association has found, a lack of association, as I’ve blogged about more times than I can remember over the last decade. Heck, there’s already been one study like this so far in 2015.

Of course, the myth that the MMR vaccine causes autism is what I like to refer to as a “zombie myth.” It’s undead. Like a herd of walkers in The Walking Dead, it just keeps relentlessly shambling along until it surrounds and devours reason and science. Or maybe a better simile is to liken this myth to slasher movie killers like Jason Voorhees in the Friday the 13th movie franchise or Michael Myers in the Halloween movies. At the end of each movie, the heroes have vanquished the killer. In many of the movies, the killer appears to have died at the end of the movie. Yet, inevitably there’s another movie and it turns out that the killer didn’t die after all. He’s still alive and slashing away. So it is with the myth that the MMR vaccine causes autism. No matter how much science is thrown at it, no matter how much it appears to be dead after each new study failing to find even a hint of a whisper of an association between MMR and autism, it always comes back.

Yet, scientifically, I prefer a different metaphor for the myth, and that’s to invoke Monty Python’s (in)famous Dead Parrot sketch, with antivaccine loons playing the role of the shopkeeper trying to deny to an unhappy customer that the parrot he had sold him was dead, telling the customer that he’s “not dead,” but rather “pinin’ for the fjords.” My response about the hypothesis that MMR causes autism goes along the lines: “It’s not pinin’! It’s passed on! This hypothesis is no more! It has ceased to be! It’s expired and gone to meet its maker! It’s a stiff! Bereft of life, it rests in peace! If you hadn’t nailed it to the perch it’d be pushing up the daisies! Its metabolic processes are now ‘istory! It’s off the twig! It’s kicked the bucket, It’s shuffled off this mortal coil, run down the curtain and joined the bleedin’ choir invisible!! THIS IS AN EX-HYPOTHESIS!!”

And so it is that this study, published in JAMA as a collaboration between the Lewin Group, Optum, and the J. Drexel Autism Institute, Drexel University, entitled Autism Occurrence by MMR Vaccine Status Among US Children With Older Siblings With and Without Autism, is yet another reason to label the MMR-autism hypothesis an “ex-hypothesis.” In fact, I rarely even bother to refer to it as a hypothesis any more because that grants too much credence to what is now a cranky conspiracy theory. (Is there any other kind?) In fact, I almost wasn’t going to write about this study for the simple reason that it’s just another in a long line of such studies that have all shown the same thing: There is no detectable association between MMR and autism. All of this leads me to wonder (and I’m not the only one) why another such study is even necessary. Certainly there doesn’t seem to be much purpose in studying the same question over and over and over again when the answer has been consistently negative. It’s reinventing the wheel, and in the process wasting resources that might otherwise be devoted to studying questions where there is genuine uncertainty about what the answer is.

Be that as it may, this study is different in that it takes into account families with an older sibling with autism or autism spectrum disorder. The rationale, given by the authors, for doing this is not unreasonable, even though they surely must have known what the result would be:

Two doses of measles-mumps-rubella (MMR) vaccine are currently recommended for children in the United States: the first at age 12 to 15 months and the second at age 4 to 6 years. Although a substantial body of research over the last 15 years has found no link between the MMR vaccine and autism spectrum disorders (ASD), parents and others continue to associate the vaccine with ASD. Parents cite vaccinations, especially MMR, as a cause of ASD6 and have deferred or refused vaccinations for their children as a result. Lower vaccination levels threaten public health by reducing both individual and herd immunity and have been associated with several recent outbreaks of measles, with most cases occurring among unvaccinated individuals.

Families with a child affected by ASD may be particularly concerned about reports linking MMR and ASD, despite the lack of evidence. Surveys of parents who have children with ASD suggest that many believe the MMR vaccine was a contributing cause. This belief, combined with knowing that younger siblings of children with ASD are already at higher genetic risk for ASD compared with the general population, might prompt these parents to avoid vaccinating their younger children. In a recent survey of 486 parents of children with ASD, nearly 20% had declined or delayed MMR immunization in the younger siblings of these children. Furthermore, a Canadian study of 98 younger siblings of children with ASD found that younger siblings were less likely to be fully MMR immunized when compared with their older siblings with ASD. However, there were no statistically significant differences in rates of ASD diagnosis between immunized and nonimmunized children. To our knowledge, this very small study is alone in examining MMR immunization and ASD outcomes among the younger siblings of children with ASD.

Thus, we set out to report on ASD occurrence by MMR vaccine status in a large sample of US children having older siblings with ASD and to compare findings with those among children who have older siblings without ASD.

The study itself is a retrospective cohort study carried out using an administrative claims database, the Optum Research Database, which includes more than 34 million individuals per year and contains both commercially insured individuals and Medicare managed care enrollees. The database contains proprietary deidentified health claims data from a geographically diverse US population whose age and sex distribution is similar to that reported by the US Census Bureau. The previous study I discussed was a case control study, in which individuals with the condition under study (autism) were compared with controls who did not have the condition and risk factors associated with the condition assessed. A cohort study, on the other hand, looks at groups exposed to a putative risk factor and those not. In this case, the risk factor under study was the MMR vaccine. Participants included children continuously enrolled during the period of 2001 to 2012 in the health plan from birth to at least 5 years of age who also had an older sibling continuously enrolled for at least six months between 1997 and 2012. The children in the study were stratified according to how many doses of MMR had been received (0, 1, or 2) between birth and five years of age. There ended up being over 95,727 children in the study group.

So what were the findings? Surprise! Surprise! There was no association detected between MMR and autism:

Of 95,727 children with older siblings, 994 (1.04%) were diagnosed with ASD and 1929 (2.01%) had an older sibling with ASD. Of those with older siblings with ASD, 134 (6.9%) had ASD, vs 860 (0.9%) children with unaffected siblings (P <  .001). MMR vaccination rates (≥1 dose) were 84% (n = 78 564) at age 2 years and 92% (n = 86 063) at age 5 years for children with unaffected older siblings, vs 73% (n = 1409) at age 2 years and 86% (n = 1660) at age 5 years for children with affected siblings. MMR vaccine receipt was not associated with an increased risk of ASD at any age. For children with older siblings with ASD, at age 2, the adjusted relative risk (RR) of ASD for 1 dose of MMR vaccine vs no vaccine was 0.76 (95% CI, 0.49-1.18; P = .22), and at age 5, the RR of ASD for 2 doses compared with no vaccine was 0.56 (95% CI, 0.31-1.01; P = .052). For children whose older siblings did not have ASD, at age 2, the adjusted RR of ASD for 1 dose was 0.91 (95% CI, 0.67-1.20; P = .50) and at age 5, the RR of ASD for 2 doses was 1.12 (95% CI, 0.78-1.59; P = .55).

Astute readers will note that most of the relative risks are below 1.0, which is actually in the protective range. They are not statistically significant, of course, but one of them, for children with an older sibling with an ASD who received two doses of MMR, the effect almost reaches statistical significance. Of course, no one is saying that the MMR vaccine is protective against autism/ASD, but rather that it is not associated with the condition. So what might explain these low adjusted RRs:

Although there were no statistically significant RR estimates indicating increased ASD risk at any age in either group of children (those whose older siblings had or did not have ASD), the statistically significant interactions in the final Cox model suggest differences in RR by both age and older sibling ASD status. The pattern in RRs across these groups was such that lower RR estimates (commonly extending into the protective range, ie, below 1.0) were observed at younger vs older ages and in children with older siblings with vs without ASD. Although protective estimates tended not to reach statistical significance, this pattern is worth further consideration. It is possible, for example, that this pattern is driven by selective parental decision making around MMR immunization, ie, parents who notice social or communication delays in their children decide to forestall vaccination. Because as a group children with recognized delays are likely to be at higher risk of ASD, such selectivity could result in a tendency for some higher-risk children to be unexposed. To be consistent with observed data, this would need to happen more often at younger ages. This seems feasible because by the time the child is older, developmental concerns are more likely to have been confirmed or ruled out and parents may then be less worried about a new exposure, such as a vaccination, influencing a child’s developmental trajectory. Estimates at older ages would thus be less susceptible to bias related to selective parental decision making, which also aligns with the pattern observed here. This explanation would also suggest that the estimate for the 1-dose RR estimate at age 5 years (1.10; 95% CI, 0.76-1.54) is least vulnerable to this bias because age 5 is several years removed from the time parents are typically deciding about the first MMR dose or weighing the importance of early developmental concerns.

In other words, the myth that MMR causes autism likely influenced these results to give the appearance of a protective effect of MMR against autism in some of these groups. We can’t conclude from this study that MMR is protective against autism, but we can conclude that there is no association suggesting that MMR could cause or contribute to autism. In other words, this study provides no support for a common antivaccine claim that, well, yes, MMR doesn’t cause autism in most kids, just in “high risk” and “genetically susceptible” kids, such as ones who have an older sibling with an ASD diagnosis. In other words, the parrot is still dead.

So why do we keep studying this question of whether MMR causes autism when so many studies have already been done and have all come to the same conclusion, namely that the answer is no? As I keep saying, from a scientific standpoint the parrot has been dead for quite some time, but unfortunately the antivaccine movement remains like the pet shop owner, claiming that the hypothesis isn’t really dead. It’s just pining for the fjords. Besides increasing degradation of herd immunity in pockets of low vaccine uptake with resulting outbreaks like the recent Disneyland measles outbreak, this is one of the significant harms of the antivaccine movement, a major waste of resources and funding expended studying the same question again and again. The problem, of course, is that these studies do not reassure the very parents who need to be reassured; all they do is to provide incrementally more confidence among scientists and pediatricians to their already high level of confidence that MMR does not cause autism.

Same as it ever was.