Oh no! SB277 is causing autism (except it isn’t)

1 May

When I first started looking online for autism information I was constantly hit with people using public data from California to try to convince me that vaccines cause autism. David Kirby was particularly effective at raising fear. Too bad he didn’t stick around and apologize, as the data now show he was clearly wrong.

Kirby was claiming that the rate of identified autism in California was going up, correlating with more mercury in vaccines. California removed mercury, and autism did not go down. It’s a great lesson. It’s easy to scare people, sell books, get fame. It takes a lot of guts to admit one was spectacularly wrong. Kirby chose the easy road.

That said, we still see armchair epidemiologists trying to make California autism data fit their pet theories. (Yes, SFASA, I’m thinking of you while I write this. But you aren’t the actual topic of today’s post). So I wasn’t surprised to see that people on Facebook were abusing California autism data to scare people about vaccines. You see California passed SB277 a few years ago, eliminating the personal belief exemption for vaccines. In other words, fewer people could opt out of vaccines for their kids. Kids entering preschool (age 3), grade 1 (age 5) and grade 7 (age 11) had to comply with the new law. The kids had to get up to date on vaccines.

Because of this, people are focusing on 3 year olds to see if the data from the California Department of Education indicates a jump in autism People are claiming that the number of 3 year olds in autism category climbed faster than did 4 year olds. And this, of course, means that SB277 caused more kids to get vaccinated and vaccines cause autism. Because everything means vaccines cause autism. (click to enlarge)

(Before one goes too far into this, SB277 doesn’t apply to special education students. So, those 3 year olds didn’t need catch up shots. But, don’t let important facts get in the way of claiming vaccines cause autism.)

When someone makes a claim like that the first thing I think is, how noisy are the data? The second thing I ask myself is, what are they not showing me. In this case, why did they tell us about 3, 4 and 5 year olds and then skip 6 and 7 year olds and show 8 year olds? What happened with those 6 and 7 year olds that they didn’t want me to see? For that matter, what happens with kids older than 8?

So I pulled the data and looked. And I made a table. Because listing these numbers like they did makes it hard to actually compare results. I don’t think they intentionally made it hard for people to compare. I just think they were sloppy. I strongly suspect they were trying to hide something, but not in failing to make a simple table. That all said, here are the number of students in the autism category by age for California in 2015 (pre SB277) and 2017 (post SB277). (click to enlarge)

In the Facebook post we were presented this question:

Other age groups increased 13-15%. What has happened to these poor 3-year olds?

But, we can clearly see that other age groups increased by numbers well outside that 13-15% range. For example, 6 year olds (remember how they just skipped that age group) saw an 18% increase. Not the 24% increase seen for 3 year olds, but a sizable increase. Which might have led a reader to ask, “hey, are these data just noisy?” And, “why didn’t you show me that data point?” Or, “what are you trying to hide?” or “Are you purposely cherry picking to make your point?” Or, “I’ll give you the benefit of doubt and assume you just don’t know what you are doing. Which, given that you believe vaccines cause autism is a very safe bet.”

Let’s keep digging. What does happen with, oh, 16 year olds? OH MY GOD! SB277 caused a huge 22% increase in autism in 16 year olds! That’s almost as big as for 3 year olds! Vaccines are to blame!

Of course, 16 year olds weren’t affected by SB277. They didn’t have to catch up on shots. In other words, there’s scatter in the data. It’s not “Other age groups increased 13-15%” as claimed. It’s “other groups increased between 6-22%.”

Remember how SB277 required older students to get catch-up shots? Like 5 year olds and 11 year olds? 5 year olds were what the Facebook poster used as their baseline, 13% increase. That’s actually below the average increase (which was 16%). How about those 11 year olds? 9% increase. Well below average. So, SB277 caused more autism in 3 year olds, but less in 5 and 11 year olds? Those kids were protected by catch up shots?

No. Let’s say it again–people are trying to put significance on noise. And I think they know it, that’s why they are not showing you all the data. That’s the polite way to say, I think they are trying to mislead you. Maybe they actually believe their claim and, you know, just don’t want you to be confused with all that data that conflicts with their claim. Maybe they believe vaccines cause autism because they have no real skill analyzing data and studies.

There are more important questions here. Questions that actually matter to the autism community.

Go through these data a few times and you should start asking yourself: why does the autism count increase for older kids? 13 year olds in 2015, there were 5874 counted as autistic. 2 years later (as 17 year olds) there were 6084. If autism is obvious, you can’t miss an autistic no way no how, how exactly did 200 or so more of these kids get counted as autistic?

Simple answer–many of these 200 kids were missed before. They were missed. They didn’t get supports and services based on their disability. And this shouldn’t happen. Autism counts, like these, aren’t an accurate count of the real number of autistics in a population. But the fact that autism counts aren’t accurate doesn’t play into the epidemic playbook. This isn’t just a problem for the anti-vaccine community. Yes, they’ve never cared about actually helping autistics. But consider SFASA (San Francisco chapter of the Autism Society of America). That’s a group whose purpose it is to serve autistics. All autistics. No matter whether they have intellectual disability or not. No matter whether they are identified or not. But SFASA denies that autistics have been missed in the past.

Ah, I digress. Back to double checking the anti-vaccine activist claims. So, 3 year olds across the state saw a big jump in the autism count from 2015 to 2017, right? If this is real, then the jump would have to be seen in various locations across the state too, right? For example, in Los Angeles Unified School District (largest in the state), we’d see the same result, right? Simple test. Let’s do it.

LA Unified counted 737 3 year olds as autistic in 2015. They counted 783 as autistic in 2017. That’s a 6% increase. That’s a lot less than the 24% statewide. So if we take the theory that “SB277 caused a jump in autism across the state” then in LA Unified, SB277 is preventing autism!

The anti-vaccine movement has always taken just the data that supports their theory and ignored the rest. They have also always used fear. This is just another small chapter in that story.


By Matt Carey

2 Responses to “Oh no! SB277 is causing autism (except it isn’t)”

  1. doritmi May 1, 2019 at 05:47 #

    Thank you for going through this.

  2. wzrd1 May 2, 2019 at 05:51 #

    I’m infamous for torturing data, just to see if something out of normal band might show up. I also have a number of friends who are medical professionals and epidemiological professionals.
    I don’t fixate and abuse reality to find a solution that’s defective, I look for trends, additive groups, process that data as well and forward it to said friends to use superior mathematical skills than I have.
    After all, to count to 20, I had to pull my shoes and socks off.
    Number 21 typically resulted in an arrest. 😉

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: