The Missing-Group Test: Sorting ADHD Evidence by How Sure You Can Be

Type “is adhd coaching evidence based” into Google and read down the page. The fifth result reports a 2017 review in which 17 of 19 coaching studies found improvements. The eighth says there is limited concrete evidence that coaching works.

Both are accurate.

They sit at different levels of confidence and neither one says so, which is most of what is wrong with the phrase “evidence based”. One question sorts them. What nobody warns you about is the bill that arrives after you start asking it.

Ask who went without

Before you believe a study, ask who in it did not get the thing being tested, and what happened to those people. I call it the missing-group test, and it takes about ten seconds.

Ostinelli and colleagues pooled 113 randomized trials covering 14,887 adults, and every one of those trials had a missing group, which is why the result is both sturdy and narrow: medication cut core symptoms, and quality of life did not move. Yesterday’s post was built on that gap.

Run the same test on coaching and the page goes quiet. The newest dedicated paper is a single-arm study of 43 people through a 12-session engagement. Nobody went without, so nobody can say what those 43 would have done on their own. It is a fair description of the field and it settles nothing about cause.

The largest dedicated coaching trial is still Field 2013, in college students, and nothing bigger has come since. Sibley and colleagues say the rest plainly in JAMA Network Open: “Randomized clinical trials are essential to establish the safety and effectiveness of ADHD coaching.”

So the search results sort into three piles. Proven: medication moves symptoms and not quality of life. Plausible: coaching works on the daily follow-through that medication leaves alone. Unknown: whether it causes that for adults in general.

The sentence you lose

Two of those three piles are sentences no coach wants on a services page.

Watch what fills the space instead. “Has proven effective.” “There’s some research to back this up.” The second one is the one that gets me, because it promises a citation and then never produces one, and the claim always swells loudest where the evidence is emptiest.

Refusing those phrases costs you something real and pays you nothing I can point to. I went through 32 priced coaching offers and not one mentioned outcome data, because no client had asked. Nobody has tested whether saying “plausible” out loud wins the work or loses it, so anyone who tells you honesty sells is guessing, and that includes me.

Sort for whoever refers

Your client is not the only reader of that page. The same JAMA study found practitioners referring patients to coaching anyway, despite the limited evidence base, and the CHADD write-up of that survey of 481 coaches describes an unregulated field where most coaches work without clinical oversight and nearly three quarters have or suspect ADHD themselves.

Unregulated does not mean useless. It means nobody upstream is checking, so it all falls to the clinician deciding whether to send you a person, and that clinician reads your claims the way you just read Field 2013.

Four or more on Part A of a screener is a reason to look further, not an answer. Evidence works on the same setting: it tells you how big a claim you may make, and the size is something you can write down before anyone asks for it.

Say plausible where it is plausible. Your client will not notice. The person who sent them will.