Friday, January 12, 2018

What I Learned from the Autism Research Program

How This Happened, and Getting There

I went to Reston, Virginia, just outside Washington DC recently.  I've covered this in recent entries, but for readability's sake, I'll rehash it.  About a year and a half ago, John Elder Robison, one of the most respected autistic self-advocates in the US, put out a call for autistic people to sign up to be research reviewers for the federal government's research programs.  I looked at the information, said, "Why not?" and sent a message to the contact listed.  You need to be vetted by an organization, which I was, via Autism Support of Kent County.  To my surprise, (and after an interview process) I was invited to participate.  2016 went by, and nothing happened, so I sort of forgot about the matter.

Fast forward to around Thanksgiving 2017 this year, and my phone rang with the opportunity to participate in 2017's Autism Research program.  40 hours of work, reading, researching, and reviewing scientific proposals, right through December.   At the end, a trip to near-DC to do the final reviews.  I winced, admittedly.  My December was already going to be stressful with the holidays.  But it was a hard opportunity to pass up- the chance to advocate not just to parents (though that is definitely important) but to researchers, who rarely leave their labs to begin with.  I couldn't say no.  Unsurprisingly, December was kind of insane.  But I did manage to get all the work done, despite much dilly-dallying, the traveling, and the time demands of the holidays.  The government paid for the airplane tickets (but not the baggage fee, which is reimbursed separately), so off I went. 

There were two different groups for this year's grants: theoretical research, and clinical trials.  The theoretical research was unfortunately the much larger category, and also the one I was assigned to.  This was rather disappointing, because part of what I'd wanted to do in DC was explain to these scientists how incredibly important immediate results were.  I wanted to tell them that time was up on researching "mechanisms" and genes, and specific processes that might increase autism rates, because we need help now, and we're not getting it.  I somehow managed to do all my preliminary reviews and get to DC without realizing that I'd been specifically assigned to the "theoretical and new ideas" section of the grants.  This was unfortunate, because I had some variation of, "this isn't helpful immediately, maybe consider that we need help now?" in 7 of my 10 reviews.  Slightly embarrassing, suffice it to say.  I had to do a lot of rewriting my reviews at the end of the trip...

The specific aim of that group wasn't the only problem I had.  At least a third of the research proposals I reviewed involved experiments on mice.  These are called "mouse models," and they're a staple in a lot of autism research.  Instead of experimenting on autistic people (which is hard to get past an ethics board), you experiment on mutant "autistic" mice.  I found this both annoying and utterly perplexing.  How do you make a mouse autistic?  Is that even possible?  How is that even a reasonable thing to do?  So at the end of the first day (which was more meet-and-greet than it was reviewing research), I asked the room of other reviewers if someone would please explain mouse models to me.

Mice Are Not Autistic

I got one taker to explain mouse models, an older man, who roped a friend of his into helping to educate me.  I later pestered a third scientist for a different perspective and better understanding of the subject.  First: the mice are not autistic.  You literally cannot make an autistic mouse right now, and you may never be able to.  Because we don't know what causes autism, only dozens-to-hundreds of factors that can play into the question, you can't make an exact representation of autism in a mouse.  Some of those factors are genes, too, and mice have different DNA than us.  Some mutations that seem to play into autism are lethal if applied to mice.  Especially if you try to do more than one.

So mouse models are relatively simple mutated mice.  Only a few different genetic modifications.  Too much more than that, and the mice stop breeding, or don't produce offspring even if they do breed, or just outright die.  So it's a tricky balancing act, because you need the modifications to make the mice have autistic-like traits.

There are whole jobs revolving around simply creating mutant mice for studies like this.  It's a hard job, too, because you have to adhere to three kinds of ideal standards:
  1. Face Validity- the mouse behaviors need to look like the autistic behaviors.  So, one example is a particular mutant mouse strain is prone to jumping, repeatedly, for long periods of time.  This roughly corresponds to stimming, and so they do experiments on those mice to test things that might affect stimming.  Needless to say, mouse models tend to handle this type of validity fairly well. 
  2. Construct Validity- this one is a bit harder.  Now our mice can't just seem to be stimming, they need to have genetic abnormalities or specific environmental condition triggers like actual autistic people.  Some of these genetic abnormalities are outright fatal or otherwise unusable in mice.  So you can see this starts to get complicated.  
  3. Predictive Validity- the mice have to respond to treatments, like medicine, in the same way an actual autistic person would.  Otherwise the mice are of minimal use for the experiments, because this is the whole point.  If something works for a mouse, you want there to be a pretty good chance it'll work for a human the same way, too.  
Lining these three types of validity up perfectly in a mouse model of autism is not feasible at present, because again, we don't know what causes autism.  Trying to model it perfectly is just not possible.  So instead these mouse models tend to focus on specific identifiable features or side-traits of autism.  Usually at least two.  Stimming, or sensory sensitivities, or social interactions, or specific genetic quirks.

And the big question I had to ask: why mice?  Why not monkeys?  Why not something larger and more close to actual humanity?  The answer is, perhaps unsurprisingly, money.  Mice are very inexpensive.  Techniques for meddling with their genetic codes are well researched and relatively simple.  Also, they have quick life cycles, which lets you test a lot of generations in short order.  If you used monkeys, or dogs, or even chimpanzees, you'd also risk a lot more outcry from animal activists.

The Reviewers Themselves

So it's pretty much mice.  And some rats.  But almost nothing else.  Micemicemicemicemice.  Listening to the researchers talk about their subjects, I kind of wonder if they aren't more familiar with, and more fond of, their rodent subjects than they are the autistic population they're trying to help. (This isn't to say that these researchers weren't willing to learn from me, or were bad people or something.  All of the people I met on this trip were pretty much lovely, and my comments and observations were verbally appreciated by more than a few people.)

Speaking of the researchers, it didn't occur to me at the time, but in retrospect I should have probably guessed: researchers are almost invariably huge nerds.  This was very comforting to me.  Nerds are easy to talk to.  You just ask them what they do, or what they care about, and sit back and listen while they spout whole paragraphs and pages about it.

I did kind of regret not asking more of the researchers about what got them into studying autism.  There were all sorts of people there, with a dozen accents from all over the world, it would probably have been a lot of interesting stories.  At least half the scientist reviewers were women, which I found unusual given the statistics for STEM careers.

Quite a bit of chatting and socialization ensued, even outside of the reviewing hours.  I tried to make it to most of the formal and informal events, but didn't entirely succeed, given how tired I tended to get by the end of the night. 

Every consumer reviewer was biologically female, also.  That was a little strange, considering the ratio of male to female autistics is 4 to 1.  (I say biologically female, because I'm agender and when I talked about it with the other reviewers, at least one of them commented that she didn't feel particularly female either, now that she thought about it.)  At least two of the other autistic reviewers are in the category that I tend to call "The Lost Generation," except they themselves weren't actually lost.  They got their diagnoses late in life, and didn't get shunted into mental institutions and prisons (these two things are sometimes the same).  A lot of their generation weren't so fortunate, particularly the more heavily affected ones.

In addition to the other autistics, there were also a couple mothers of autistic children serving as consumer reviewers.  The one that served as my mentor was retired military, and has a more heavily affected son in his 20s.  She was quieter than the others, but had lots of good things to say when she did speak up.

On the Whole...

This was a pretty good experience.  I learned a lot of useful things, met a lot of interesting people, and I'll be getting paid for it.  I would definitely do it again, particularly if the bulk of the reviewing didn't have to happen around Christmas.  It was still doable, obviously, since I did it, but if I had to do it again, I'd definitely just do it all before leaving for the holidays, and not have to care about it later.  

The reviewing rooms were set up relatively comfortably, with a special laptop for each reviewer and a microphone between every two reviewers.  The chairs were arranged in a very straight U shape, with the moderators at the bottom of the U and the reviewers in two lines facing each other.  This allowed for some ability to see who was talking, and for everyone to see the screen at the far side of the U.  There were also support personnel off to one side, but they kept quiet unless called upon.  They were there to take notes, provide tech support, keep track of time, etc.  

I can't (ever, for some reason) talk about the research I reviewed specifically, though I will say there is definitely one researcher out there that needs to be smacked upside the head repeatedly with an ethics book.  I pointedly didn't memorize the institution and head researcher attached to that study, so I couldn't be tempted to yell at him later.  

As a side note, this program through the US government is the only grant program in the nation (for all conditions, not just autism) that includes consumers in its reviewer ranks.  It's common enough to have scientists look over the other scientists' work, but asking the opinions of people whose lives are directly affected by your research is apparently a very new thing.  This knowledge was rather dispiriting to me, but it probably means I'm never going to be out of a job as a self-advocate...  It does strike me as kind of hilarious that the government is at the forefront of progress in this area.  Y'know, the same government that trundles slowly, ponderously forward in pretty much every other case. 

My amusement aside, the staff of the program were pretty much unfailingly polite and willing to listen.  The scientific reviewers were a bunch of good natured nerds, for the most part.  The consumer reviewers were all interesting, pleasant people.  It was definitely an interesting experience, and one that gave me a lot to think about.  If nothing else, I should also be better able to read research articles for my Reading the Research posts. 

No comments:

Post a Comment