The importance of clear clinical definitions

… and a recommendation for Secretary Kennedy

There has been an uproar over Secretary Kennedy’s latest remarks about autism, and I’ve made some points about this already. Apparently some of this uproar is because parents of children with milder (not profound/severe autism) so desperately want acceptance and for their children to blend into society (definitely agree there), that any mentions of profound/severe autism are rejected by them (that’s bad)… to the point that they are actually adversarial against the parents of the profound/autistic kids. This is illustrative of a broader problem, which in part is an epidemiologic problem.

Let me start with a couple of examples.

I’ve referred to the obesity epidemic before. Let’s think about the definition of obesity. Overweight is defined as BMI ≥ 25, obese is defined as BMI ≥30, and morbidly obese is defined as BMI ≥40. Let’s make that concrete. For a 5’10” man, 180 lb is a BMI of 25.8, 210 lb is a BMI of 30.1, and 280 lb is a BMI of 40.2. If you were referring to just overweight men that were 5’10”, you’d be lumping 180 lb men in with 300 lb men. Further consider that muscle weighs more than fat, and you could be including a muscular guy in with a morbidly obese guy if you are just referring to “overweight”. Clearly, it makes sense to differentiate between these different categories of overweight and obesity.

Let’s look at the impact on research – two examples of tuberculosis clinical trials. A recent, interesting clinical trial of a nutritional supplement to prevent TB disease measured the outcome of disease two ways – one way looking at clinical TB, and a second, stricter definition that required microbiologic confirmation. The protective effect of the nutritional supplement was statistically significant (and stronger effect size) in the microbiologically confirmed group, as opposed to the looser definition of clinical TB. So a stricter definition yielded a stronger, clearer effect. Another TB clinical trial looked at conversion to acute infection as the outcome. The protective effect was only seen in sustained new infections, not in subjects that showed an initial positive test that then reverted. Again, clear clinical definitions are important.

Epidemiology is full of the study of complex disorders and diseases that vary in severity. Read that again – human disorders and disease vary in severity.

Now, let’s apply this to autism. There is profound/severe autism, which in itself is highly variable, but generally characterized by individuals that are disabled to the point that they cannot take care of themselves, and may never be able to care for themselves. There are milder cases of autism, individuals who possibly through therapy or maybe through other means have adapted, and although they still are not neurotypical, they can take care of themselves. And there are variations in that spectrum. As most parents of autistic kids will tell you, if you’ve met one autistic kid, you’ve met one autistic kid. This is especially true when two siblings are both on the spectrum – they are not the same.

So the initial moral of this story is: Be considerate. For the parents of autistic kids who have fought their whole lives for their kids to fit in – that isn’t diminished by those autistic kids who will never be able to fit in. Do the best for your child. But never assume that you know every other situation.

But now, my request for Mr. Kennedy… I’m praying for this research study to be the best possible study ever. I think it would be important to stratify by autism severity. The causes may be quite different in those different categories. Clinically characterizing those ranges of severity will be critical… especially so that everyone feels seen.