Health Policy$ense

Vaccination Status of Kids With Nonmedical Exemptions From School-entry Mandates

Good News, Bad News

It’s been almost a year since the Disneyland measles outbreak offered a startling wake-up call about widespread vaccine hesitancy and refusal among parents of young children. In the aftermath of that outbreak, the California legislature voted in June to eliminate all nonmedical exemptions from school-entry immunization mandates, citing rising rates of exemptions in the state and their role in undermining herd immunity against vaccine-preventable diseases like measles. When the law goes into effect next summer, California will join Mississippi and West Virginia as the only states in the U.S. with no provision for nonmedical exemptions.

Will California’s new law eliminating nonmedical exemptions decrease outbreaks of vaccine-preventable diseases? In theory, yes. The logic goes like this: if there are no nonmedical exemptions, all parents will have to vaccinate their kids to enroll in school. If all kids are vaccinated, herd immunity will be intact and diseases like measles won’t spread. But this argument rests on a crucial assumption about kids with exemptions that has not been rigorously tested: How vaccinated are they, actually? In other words, how well do exemption rates reflect vaccine coverage rates in a population of school-aged children?

My collaborators and I just answered that question for one state and for one vaccine, and I’ll share our findings below. But to back up a bit, you might be asking right about now: “Why would a kid with an exemption from vaccine mandates be vaccinated?” Well, there are many reasons. First, a parent may want an exemption only from a specific vaccine or vaccines, and will still give the kid other vaccines. Second, a parent may have started the recommended vaccination series and then decided to stop, leaving the child partially vaccinated.  Third, parents may not be able to round up all the kid’s vaccine records when it comes time for kindergarten registration. In states where getting an exemption is as easy as signing a form, the parent may choose to file an exemption rather than chase down the records. Similarly, the school may decide it’s easier to nudge the parent towards an exemption rather than wait for the parent to submit records or use a “conditional acceptance” status that requires substantial administrative follow up.

Under any of these scenarios, kids with exemptions will turn up in schools’ own immunization surveillance data as completely unvaccinated, partially vaccinated, or even fully vaccinated. In fact, about half of exempted in kids in California (prior to a 2014 law making exemptions a little harder to get) had no vaccine doses at all listed on the school immunization record. Using these school-based vaccine records from 2009, we looked for clues from the kid’s own vaccine record and data from exempted kids at the same school to fill in the blanks with imputation and bounding methods.

What did we find? At least for the measles-mumps-rubella (MMR) vaccine in California, kids are indeed a lot more vaccinated than the surveillance data would suggest.  For MMR1 (the first dose), the surveillance data indicated that 47% of kids with exemption had received that dose. Our imputation models suggest that this figure likely ranges from 64-92%, depending on the rules we used to make educated guesses about those blank doses.  These were the findings for “typical” schools in our sample – schools with average nonmedical exemption rates. At schools with high exemption rates, we found that kids with exemptions were less likely to have received their MMR1 dose: 34% in the surveillance data, vs. 49-90% in our simulation models. Results for MMR2 were similar but with overall lower coverage rates: 25-58% at typical schools (vs. 18% in the surveillance data); and 16-63% at school with high exemption rates (vs. 11% in the surveillance data).  In general, we estimate that MMR coverage rates are 10-50 percentage points higher than the surveillance data show. Exempted kids are more vaccinated than they look.  That’s the good news.

But there’s also some bad news. Our results highlight the fact that outbreaks like the Disneyland measles outbreak can occur even when herd immunity is in better shape than our surveillance data indicate. Prior studies have found substantially higher risk  (22-35 times higher) of measles associated with vaccine refusal—but in these studies, exemption was used as a proxy for being unvaccinated.  If we use our low estimate that 50% of exempted children in typical schools have received at least 1 dose of MMR, then those original risk analyses need to be doubled; the disease risk associated with vaccine refusal is actually 44-70 times higher. Using our upper bound estimate of 90% of exempted kids having received 1 dose of MMR, the relative risk increases 10-fold, or 220-350 times higher risk associated with vaccine refusal. 

Stay tuned: We’ll be repeating the analyses for more vaccine doses and for later years, including after California’s first exemption law change in 2014.  If our findings from this study hold true, then the Disneyland wake-up call is even more startling than we thought.