Who Is The Anti-vax Movement? Data Science Explains.

It was all theoretical until Jenny McCarthy gave Sidney Crosby the mumps. Then it got real. Ok, I know that’s a sensationalist — not to mention flagrantly untrue — thing to say, but it’s how I suddenly felt a few weeks ago. The abstract notion of the anti-vaccination movement and the re-emergence of these dangerous all-but-forgotten diseases suddenly landed on my doorstep, and it got me thinking. Who exactly are these people who don’t vaccinate their children?

The great state of California, whose Department of Public Health recently made public all of their data surrounding childhood immunization levels, can help us begin to answer this question. Specifically, they’ve posted a child-by-child count of non-immunized California Kindergarteners who have opted out of vaccinations through the state’s Personal Belief Exemption (PBE) program. To get a lay of the land, we’ll start with a very straightforward plot directly from their 2014-2015 Academic Year Annual Immunization Report.

Percentage of California Kindergartners with Personal Beliefs Exemptions

This plot, from the CA Department of Public Health’s annual Immunization Report, gives the percentage of California kindergartners with Personal Beliefs Exemptions, by school type.

So it’s pretty clear that the rate of PBE’s has gone up since 2010. But as with any simple visual, this one raises more questions than answers. For example, we still can’t quite see where the high rates of PBE’s are coming from. From this information, it’s possible that there’s one gigantic mega-school housing all of the unvaccinated kiddos. And this is an important distinction to make, since what we know of herd immunity says that risks for illness increase when large numbers of unvaccinated kids interact regularly. In case you missed it, Evelyn Lamb gave a great run-down on how diseases spread even in populations that are partially immunized.

So, beyond this basic distribution question, it would also be interesting to see how PBE rates look across types of school. For example, are the rates at small schools and large schools comparable? Can we see consistencies across school types within a district? To help us tease out some of the finer details from the mountain of raw data, Kieran Healy at the blog Crooked Timber turned it into some helpful eye-candy.

The personal belief exemption rates for kindergarteners in California schools by school, based on data from the California Department of Public Health.  Courtesy of Kieran Healy.

The personal belief exemption rates for kindergarteners in California schools by school, based on data from the California Department of Public Health. Courtesy of Kieran Healy.

What we can see here is that larger schools actually tend to have lower PBE rates, that is, they have higher rates of immunization. So the unvaccinated-mega-kindergarten theory is out. This plot also gives a pretty clear answer in the public vs private school question. If you study the data points closely, it also shows some interesting inconsistencies within districts. Healy notes,

“The concentration of PBEs in smaller schools is evident, as is the concentration in private schools. Note that regions with high PBE schools can still show a lot of heterogeneity. For example, consider schools in Berkeley. On the one hand it is home to the school with the second-highest PBE rate in the state. On the other hand, six of its fifteen other schools have PBEs of zero, two more are at three percent or lower, and the remainder range from six to sixteen percent PBEs.”

Finally, we might want to get a more granular sense of what types of schools we’re dealing with. Whether parochial private schools and charter private schools have comparable rates, and what sort of public schools have large numbers of unvaccinated kids.

California Dept. of Public Health data gives the personal belief exemption rates for kindergarteners in California schools.  Courtesy of Kieran Healy.

California Dept. of Public Health data gives the personal belief exemption rates for kindergarteners in California schools. Courtesy of Kieran Healy.

I’ll let you draw your own conclusions from this. But I think this data paints an interesting picture of the vaccination culture — at least in California. Check your own state to see if they make their immunization data available to play with, and if you can’t find your own, Healy put the California data set up on Github, along with the R code for his plots. So with that, I wish you happy data mining. And remember, it’s still flu season, so probably go wash your hands right now.

This entry was posted in Biomath, Mathematics and Computing and tagged , , , , . Bookmark the permalink.