Seizure Risk After Infant Vaccination - Lost in Translation?
BudWiederman, MD, MA, Evidence eMended Editor, Grand Rounds
"The incidence rate ratio of febrile seizure after vaccination was 23 ..., and the
attributable risk was 3.92 ... febrile seizure cases per 100,000 persons vaccinated."
How do you translate that so most clinicians can understand the impact of these findings?
(Not to mention explaining it to parents!)
The authors did a good job trying to put these numbers into perspective. The hallmark
of evidence-based medicine is, or should be, to translate scientific findings into
information patients and families can use in making informed decisions about their
healthcare. In this instance, any pediatric healthcare provider should be able to
help parents understand the degree of risk (and benefit) of infant immunization.
This study utilized data from 9 of the sites contributing to the Vaccine Safety Datalink (VSD) of the Centers for Disease Control and Prevention. Cutting to the chase, they
looked at almost 600,000 vaccination visits for 1- to 6-month old infants and then
sorted out episodes of seizures following inactivated (i.e. not live) vaccine receipt.
To do this, they compared the rates of febrile seizures during the risk interval,
defined as the day of vaccination plus the following day, compared to seizures occurring
14 - 20 days after vaccination when an inactive vaccine-related seizure is biologically
implausible. The main results quoted in my lead-in concern all probable febrile seizures
and are statistically significantly different between the 2 groups. Further data analysis
showed significant differences when considering both probable and possible febrile
seizures, febrile seizures among infants who received all the recommended vaccines
at the immunization visit, and febrile seizures excluding other causes of fever such
as otitis media or pyelonephritis, but not in the group with seizures without fever.
Without getting too worried about statistical methods, let's look closer at the data,
and especially consider relative versus absolute risk. Just looking for relative risk,
one could say the risk of febrile seizures in vaccinated infants is about 20 times
that of young infants who haven't received vaccines. That's a scary proposition if
stated that way. However, one needs to consider the overall rarity of this event,
only a handful of children per 100,000 vaccinated. The absolute risk increase, tied
more to the attributable risk of 3.92, means that for every 100,000 infants immunized,
an additional 4 of them would experience febrile seizures, compared to the group not
immunized immediately preceding the seizure event. That's pretty rare, and of course
much less than the multitude of complications that can occur following the actual
infection the vaccines will prevent.
Sometimes I use a shopping example to explain relative versus absolute risk to laypersons.
Suppose you see 2 ads on TV. One offers a 50%-off sale, and the other a $1-off sale.
Which store would you rush out to buy their goods? Of course, it depends. If we're
talking about a new car sale, we'd all take the 50%-off route, However, if the sale
is at your local Dollar Store, that means they are giving away the products for free
at the $1-off store, and you'd want to get there quickly before everything is gone.
The relative risk is equivalent to the 50%-off sale, and absolute risk is the $1-off
sale. When the numbers are very low (either an inexpensive product or a rare medical
event), the relative risk can sound seductively high but actually be misleading. Advertisers
frequently use such wording, among other strategies, to trick us into thinking we're
getting a better deal than we really are. Don't be caught in this trap with medical
studies. The febrile seizure risk for infant immunizations isn't zero, but it's pretty
Other methods of explaining vaccine risks and benefits have been studied to try to
help vaccine-hesitant parents make informed decisions. Maybe I'll have a chance in
a future "Fifth Tuesday" posting to explore this more.