An article has been making the rounds on the internet claiming there is a link between vaccines and epilepsy. This is outrageously false, but worth taking a deeper look. Mainly because this article highlights some key points which can be used to teach others how to skeptically read things on the internet.
The article’s first paragraph nicely highlights all that is wrong in what follows:
The rate of epilepsy among children and the elderly has been skyrocketing, with 1 in 20 children under five now suffering from the seizure condition in the United States. More and more parents say that vaccines triggered their children’s seizure disorders. The government maintains that while vaccines can trigger febrile (fever related) seizures, the many cases of epilepsy that begin immediately following infant vaccination are merely coincidental or were bound to occur eventually.
There are two main claims here, which i will dissect separately. The first claim is that epilepsy is skyrocketing. The second claim, which is a bit more complicated, involves the relationship between Doctors and parents, and their differing standards of evidence.
“Skyrocketing”: I Don’t Think It Means What You Think It Means
A rocket is something that goes from slow to fast at a fairly quick pace. In order for the analogy to work, something that skyrockets must dramatically change between two time points in roughly close proximity. For example, in places that do not vaccinate, measles can skyrocket. Where measles was almost eliminated in the United States, 113 cases were linked to the Disney Land outbreak last year, which spread due to low vaccination rates. Therefore, there was an environmental cause of the skyrocketing disease; a decrease in herd immunity.
To my surprise, with the 1 in 20 figure listed in the articles title, there was no mention of a preceding number. We can’t determine based on one number alone what the trend of diagnosis is. With a simple google search, no news was found indicating any trend in epilepsy cases. In fact, the only thing that came up, was the article itself!
Those two facts alone should be sufficient for an immediate quick skeptical read, but I decided to go a bit deeper into the research to see what I could learn about the prevalence of epilepsy. Epilepsy is highest in children younger than 5, which is what the 1 in 20 figure represents. However, kids younger than 17, have a prevalence between 6-10 in 1000, depending on which study you look at. These studies were mostly conducted in the mid 90’s, and as far as I can tell, none of them claim that this number is indicative of an increase in epilepsy, but simply a more robust accounting of its prevalence. Again, this leaves me wondering, how this author concluded epilepsy was skyrocketing?
Even more telling, in the study cited in the last paragraph, they acknowledge the difficulty in measuring childhood epilepsy. What they say is worth quoting in full:
Almost 40% of children in our study previously diagnosed with epilepsy/seizure disorder were not reported by parents to currently have the condition. Although these data could reflect a true remission of seizure activity, they could result from an initial “misdiagnosis” of epilepsy/seizure disorder, inclusion of some children with single febrile seizures, or variation in interpretations of the terms “current” versus “ever” having seizures.
If epilepsy is frequently misdiagnosed, so that the prevalence is probably lower than the 1 in 20 figure, then it doesn’t seem like the rocket has even been launched.
The Pitfalls of Patternicity: The Plural of Anecdote is NOT Data
Michael Shermer, in his book The Believing Brain, describes what he calls patternicity “…the tendency to find meaningful patterns in both meaningful and meaningless noise”. Patternicity is not always bad, after all, our folk psychology, and folk physics, tend to lead us towards correct predictions some if not most of the time. However, our tendency to find patterns, even where none exist, can lead us into trouble. This is why we do science, and why we should approach anecdotes skeptically.
The main body of the article is basically a list of anecdotes, which are almost always useless in determining the causal relationship between an environmental stimulus and a disease or disorder. There are three main problems with the anecdotes provided; the problem of diagnosis, the differences in time between the supposed cause and its effect, and the problem of biased sampling.
Given the frequency of misdiagnosis as pointed out in the previous section, some cases of febrile seizures will be diagnosed as epilepsy. Febrile seizures are a possible side effect of vaccines, although they are relatively harmless they can be scary to witness as a parent. One of the anecdotes is as follows:
My daughter had her first seizure about 12 hours after her mmr shot at 12 months. When I contacted the administering gp we were told it was a normal reaction to the vaccination febrile seizure, and it will not happen again. from then on she started having myoclonic seizures where her eyes would roll for a split second. when i mentioned this to the GP they said, those will go away with time. Instead of going away they have increased in length and frequency.
Notice how the seizures changed? It is possible that the child had a febrile seizure initially, but then later developed epilepsy. This is portrayed as Doctor incompetence, but diagnosis can be complicated. Given that the symptoms of a febrile seizure and epileptic seizure can manifest themselves similarly, diagnosis is difficult. Notice as well that it seems like the patient self-diagnosed their child with myoclonic seizures.
Diagnosis is not simple, you can be misdiagnosed, you will tend to self-diagnose, and in the future your diagnosis could change. In fact, 70% of children outgrow their epilepsy. None of the anecdotes provided gave any indication of whether or not the epilepsy remained, improved, changed etc.
The times between the vaccine and the seizure differ from anecdote to anecdote. Some range between minutes after, to days after, to weeks after.
My son had his 1st seizure IN THE DOCTOR’S OFFICE 1-2 minutes after receiving two vaccinations. The seizures continued from that day forward.
My son is now 6 years old and has been having seizures since he had his first vaccine at 2 months of age. He had his first seizure about 36 hrs after his needle. He did not have any fever, actually has never had a febrile seizure ever, but has had a multitude of others.
My son has been having neurological problems, seizures, movements that doctors aren’t sure what to call, etc. since last May. He was a healthy 18 mos. old little boy and then 11 days after his checkup (vaccines) he vomited and had a 1+ hour seizure.
Even in the first anecdote, there is a gap in time between the first febrile seizure and the supposed myoclonic seizure. We are not told how long the myoclonic seizure occurred after the initial seizure. It could be that the child had both a febrile seizure that was vaccine related, but was also going to eventually develop epilepsy, the two having nothing to do with each other.
If vaccines do cause epilepsy, or autism, or anything, there should be some consistency between the onset of the problem and the supposed cause. Given that no mechanism has been provided, probably because no mechanism exists, people can just decide that their child’s seizure is vaccine related even when it occurs weeks after they received their shot.
There is also the concern of the ‘post hoc, ergo propter hoc’ fallacy, which is related to the oft quoted phrase, correlation does not equal causation. Just because something occurs after something else, does not mean that that something caused it. This is why scientists try to isolate potential causes in order to increase our confidence in the causal relationship. A simple anecdote is worthless, since it wasn’t isolated from the myriad of potential causes which make determining causation almost impossible. This is especially problematic in the field of health, since a lot of what goes on beneath the skin is hidden from us.
Finally, all these anecdotes were culled from a message board of an epilepsy foundation. In other words, this is a biased sample, from a population of worried people. A proper epidemiological study will look at the population at large, and given that there is no reason to suspect that epilepsy is increasing, such a study would most likely be a waste of time, energy, and resources.
All of which is why the Doctors are not worried of, or suspicious of, a link between vaccines and epilepsy. Notice the title of the article states that parents are questioning; not Doctors, not scientists. The parents see this as Doctors being dismissive, but really it is probably because most people don’t understand what consists of good evidence. Personal anecdote, no matter how compelling, is simply not good enough.
A Narrative Confusion
It is clear that the author of this piece is trying to create a narrative, especially considering the lengths she goes to avoid the obvious pieces of evidence sitting right in front of her. Notice the first sentence of the article:
The rate of epilepsy among children and the elderly has been skyrocketing, with 1 in 20 children under five now suffering from the seizure condition in the United States.
It took me awhile to see it, but when it finally struck me, it was so obvious what was going on. Why did she include the elderly in this first sentence? Think about it; if the article is claiming that childhood vaccines are causing an increase, a skyrocketing increase, in epilepsy, why is epilepsy skyrocketing in the elderly? A population which would really only be exposed to a single vaccine.
Notice how she also avoids the evidence that 70% of children outgrow their epilepsy:
Approximately 1 in 100 adults now has epilepsy in the United States, while 1 in 20 children under five years old are now believed to have the disorder.
How odd, that she would include the elderly as having skyrocketing epilepsy, acknowledge that adults do not, but then use that fact to bolster how frightening it is that more children have it. Odd, or clearly the case of someone trying to fit the evidence to their narrative. Funny, however, that the evidence so obviously does not fit her narrative that she can’t even succeed at contorting the evidence to make it work in her favour.
Vaccines don’t cause epilepsy. They don’t. Seriously, internet. Stop.