Every preschooler is “on the go.” That’s practically the job description of being three. So when a pediatrician’s checklist asks whether your kid is constantly moving, checking “yes” tells you almost nothing. Everyone checks yes. A new study, Associations Between ADHD Symptoms and Diagnosis Across the Preschool Period, published August 29, 2026 in the Journal of Attention Disorders, is the first to actually rank the ten standard ADHD symptoms by how well each one predicts a real diagnosis in kids ages 3 to 5. And the symptom parents fixate on (hyperactivity, the running-jumping-climbing thing) turns out to be almost useless on its own.

That’s not a small finding. It lands right as fall pediatrician visits and preschool evaluations are stacking up on calendars everywhere, which means a lot of parents are about to sit across from a clinician filling out a symptom checklist that, until now, treated all ten items as roughly equal. They’re not.

TL;DR: What the Study Found

Question	Answer
The study	Associations Between ADHD Symptoms and Diagnosis Across the Preschool Period, Journal of Attention Disorders, published Aug. 29, 2026
Who ran it	Julia S. Feldman, Ph.D., postdoctoral fellow, University of Pittsburgh Department of Psychiatry
The data	ECHO Program national dataset, caregiver-reported symptoms at age 3 (N=1,430), age 4 (N=1,453), age 5 (N=2,029)
What’s new	First study to rank all 10 DSM ADHD symptoms by predictive strength across the preschool years
The finding	Hyperactive/impulsive symptoms (especially “on the go” and interrupting) beat inattentive symptoms as predictors, but “on the go” is so common it’s nearly worthless alone
My take	Frequency and predictive power are two different things, and most symptom checklists still confuse them

Who this is for: Parents heading into a fall pediatrician visit or preschool evaluation, anyone whose 3- or 4-year-old just got flagged by a teacher, and anyone who’s been told to “wait and see” and isn’t sure if that’s good advice.

What Did Researchers Actually Look At?

Feldman and her team pulled data from the ECHO (Environmental Influences on Child Health Outcomes) Program, a national research effort that pools cohorts from across the US into one dataset large enough to answer questions no single clinic ever could. For this study, that meant caregiver-reported ADHD symptoms collected at three separate ages (1,430 kids at age 3, 1,453 at age 4, and 2,029 at age 5), checked against which of those kids went on to receive an actual ADHD diagnosis.

The design matters here. This wasn’t a snapshot of symptoms in already-diagnosed kids. It was a look at symptoms before diagnosis, across the exact window (ages 3 to 5) where “is this just being a toddler or is this something else” gets asked the most and answered the least reliably.

Researchers assessed all ten symptoms that show up on standard ADHD checklists: the inattentive half (things like not following instructions, losing items, being easily distracted) and the hyperactive/impulsive half (running and climbing excessively, fidgeting, blurting things out, interrupting, being “on the go”). Every single one of the ten was significantly linked to an eventual diagnosis. That part isn’t the surprise. The surprise is how unevenly they were linked.

Which ADHD Symptoms Actually Predict a Diagnosis?

If you only take one thing from this study, take this: not all ten symptoms carry equal weight, and the two heaviest hitters aren’t the ones most parenting advice focuses on.

1. “On the go” (acts as if driven by a motor). The single strongest predictor in the study. Kids flagged for this were more than six times as likely to receive an eventual ADHD diagnosis as kids without it.
2. Interrupts or intrudes on others. The other standout. Also linked to more than six-fold higher odds of diagnosis, and one of the few symptoms that’s genuinely hard to write off as “typical for their age.”
3. The remaining hyperactive/impulsive symptoms (excessive running/climbing, fidgeting, difficulty waiting a turn, talking excessively), significant, but consistently weaker predictors than the top two.
4. The inattentive symptoms (losing items, not following instructions, being easily distracted, difficulty sustaining attention), significant too, but as a group, the weakest predictors of the ten.

That ranking cuts against a lot of casual assumptions. Inattention gets treated, culturally, as the “real” or more clinically serious half of ADHD. It’s the half that shows up in report cards later, the half most adult self-diagnosis conversations center on. In preschoolers, per this study, it’s actually the least discriminating cluster of symptoms. Hyperactive and impulsive behavior, especially the “on the go” and interrupting items, is where the signal is strongest at this age.

Why “On the Go” Is Also a Trap

Here’s the part that should reframe how you read a preschool symptom checklist: being the single strongest predictor and being a useful screening flag are not the same thing.

“On the go” was also the most commonly reported symptom in the entire dataset — endorsed for roughly two-thirds of 4-year-olds and about a third of 5-year-olds. Read that again. Two out of every three 4-year-olds in this national sample were rated by their own caregivers as constantly on the move. Most of those kids do not have ADHD.

So a symptom can be strongly associated with diagnosis in a statistical model and still be a weak screening tool in a pediatrician’s office, because it’s true of nearly everyone at that age. A checkbox that two-thirds of typically developing 4-year-olds trigger isn’t telling a clinician much by itself. It needs company (other symptoms, impairment across settings, duration, intensity) before it means anything.

That’s the nuance a ten-item checklist scored as a flat sum tends to flatten right back out. Feldman’s framing, per her interview with Managed Healthcare Executive, is that context and impairment level need more weight in early evaluations than raw symptom counts currently give them. A kid who’s on the go everywhere, all the time, in a way that gets them kicked out of every activity is a different picture than a kid who’s on the go at the playground and settled at the dinner table.

Diagnosis Rates Climbed Fast Across These Three Years

ADHD diagnosis rates in this sample rose noticeably across the study window, from about 6% of children at age 3 to about 12% by age 5. That’s not necessarily kids “becoming” more ADHD as they age. It’s more likely a mix of symptoms becoming clearer against typical development, more caregiver and teacher contact points (preschool, daycare, structured activities) generating more opportunities to notice a pattern, and clinicians growing more comfortable diagnosing as a child ages out of the “probably just being three” window.

That climb is exactly why this study’s timing matters. If a 3-year-old is showing some of these symptoms now, “wait and see” isn’t a neutral default. It’s a bet that the picture will get clearer on its own, and the data suggests it often does, but not without cost to the kid in the meantime.

What Researchers Are Actually Calling For

Feldman’s team isn’t arguing for stricter or looser diagnostic thresholds across the board. The recommendation is more specific: build screening tools that account for context, not just symptom frequency. A hyperactivity measure that only asks “is your child often on the go” is asking a question two-thirds of parents will answer yes to regardless of outcome. A better version asks where, how intensely, and with what consequences.

The other half of the message is aimed at how early concerns get handled in the exam room. When caregivers flag hyperactive or impulsive behavior early (the exact symptoms this study found most predictive), the researchers’ framing pushes back against defaulting to “they’ll grow out of it.” Some kids do. This isn’t a call to diagnose every energetic toddler. It’s a call to actually weigh early parental concern about these specific behaviors instead of shelving it by default, especially when it’s paired with real impairment across settings.

We’ve written before about how long that “grow out of it” reflex can echo into adulthood: the surge in adult ADHD diagnoses is, in no small part, a backlog of people whose early signs got waved off for years. Catching the pattern at 3 instead of dismissing it doesn’t guarantee a different outcome. But it at least puts the decision in front of someone qualified to make it, rather than defaulting to “wait and see” by habit.

What Should You Actually Do With This?

A few things follow directly if you’ve got a preschooler and a symptom checklist headed your way this fall.

Don’t panic over “on the go” alone. It’s the most common symptom in the entire dataset for a reason. Most 4-year-olds have it. On its own, it tells you your kid is a normal 4-year-old, not that something’s wrong.

Pay closer attention to interrupting and intruding. It’s less universal than constant motion and was one of the two strongest predictors in the study. If it’s happening across multiple settings (home, daycare, a grandparent’s house), name that explicitly at an evaluation instead of folding it into a general “he’s a handful” description.

Bring context, not just a checklist score. If you’re heading into a pediatrician visit or a preschool evaluation, come with specifics. Name the settings where the behavior shows up and where it doesn’t. Say what it’s costing your kid, socially or academically. That’s the exact gap Feldman’s framing is trying to close.

Don’t wait for a school problem to take it seriously. Research on childhood ADHD’s long tail makes a strong case that earlier isn’t just easier. It changes the trajectory. A preschool evaluation isn’t jumping the gun; the diagnosis rate in this study nearly doubled between ages 3 and 5 for a reason.

Know that evaluation tools are still catching up. Early identification research is moving fast right now: AI-based prediction models are already being tested to flag ADHD years before a formal diagnosis is typically made. Context-aware screening like Feldman’s team is calling for is part of that same shift toward catching this earlier and more accurately.

If a 504 or early intervention conversation comes up, you’re not overreacting. Once a preschooler ages into the school system, the practical difference between a 504 plan and an IEP becomes relevant fast. It’s worth understanding before you’re sitting in that meeting under time pressure.

Our Take

The easy headline version of this study is “hyperactivity predicts ADHD better than inattention in preschoolers.” True, but it undersells the actual point.

The real finding is about the gap between what’s common and what’s meaningful. “On the go” is simultaneously the single best predictor of ADHD in this dataset and something two-thirds of typically developing 4-year-olds also display. Both things are true at once, and a checklist that can’t hold that contradiction is going to either over-flag half of every preschool class or miss the kids where the pattern is real.

What actually moves the needle, per Feldman’s own framing, is context: not just whether the behavior shows up, but how intense it is and what it costs the kid, layered on top of the symptom count instead of standing in for it. If you’re a parent walking into a fall evaluation with a gut feeling that something’s different about your kid’s version of “on the go,” this study is a decent piece of evidence that your instinct is worth bringing up directly, in detail, instead of letting a generic checklist speak for you.

---

This post cites findings from Feldman, J.S., et al., “Associations Between ADHD Symptoms and Diagnosis Across the Preschool Period”, Journal of Attention Disorders, published Aug. 29, 2026 (also indexed on PubMed), using data from the ECHO Program. Additional context from Dr. Feldman’s interview with Managed Healthcare Executive. Symptom definitions referenced from the CDC’s ADHD signs and symptoms guidance. This is not medical advice. If you have concerns about your child’s development, talk to your pediatrician.