Why One Clinician Says ADHD and the Next One Says No

Two clinicians, two answers, one child. Here is what is actually different.

ADHD readers, or anyone whose attention is already drifting: the condensed version is at the bottom under The Cliff Notes. Skip ahead. That is what it is there for.

Parents bring me folders. Sometimes it is an accordion file with tabs. More often it is a manila folder with a school printout, a pediatrician's note, and a PDF someone printed at work. Inside there are two answers to the same question, and the parent wants to know which one is right.

The first document is three pages. It came from a telehealth visit that lasted about twenty minutes and ended with a prescription. The second is twenty-two pages and says the picture is more complicated than ADHD.

The question I get asked, usually about forty minutes in, is some version of this: did we just buy a diagnosis? Or the reverse, which is just as common. Did the second person miss it because they wanted to sell us more testing?

Fair questions. Here is the honest answer, and it starts with the part that makes clinicians uncomfortable.

Some of the disagreement is legitimate, and you should know that before you judge anyone

ADHD is dimensional. So is autism. Neither one is a switch. Attention, impulse control, and executive functioning are distributed across the population the way blood pressure is, and the diagnostic threshold is a line drawn across a continuous distribution. For a child sitting near that line, two careful clinicians using the same criteria and the same data can land on different sides of it. That is not incompetence. That is what happens at a boundary.

Informant agreement is also modest, and it always has been. Correlations between parent and teacher ratings of inattention run somewhere around .45. Low to moderate. That means the mother and the fourth grade teacher are frequently describing what looks like two different children, and neither of them is lying. Home and school make different demands, provide different scaffolding, and produce different behavior. A good evaluator expects that discrepancy, interprets it, and can tell you what it means. A weak one picks whichever informant confirms the hunch they walked in with.

Autism disagreement usually has a different source. One clinician is working from an outdated prototype, and the prototype was built decades ago on young boys with obvious presentations. Apply it to a twenty-nine year old woman who has spent her whole life studying other people in order to pass, and it fails. She makes eye contact because she taught herself to. She has friends because she learned the script. She gets told she is too social, too verbal, too successful. The criteria did not miss her. The clinician's mental image did.

And presentations change. A diagnosis given at six and not confirmed at sixteen is not automatically an error. Sometimes the child grew into a set of skills. Sometimes the environment changed. Sometimes what looked like ADHD at six was a sleep disorder, or an unrecognized language problem, or a household in the middle of something hard.

All of that is real. None of it is what most parents are actually running into.

Where the disagreement stops being clinical

There is a category of evaluation on the market right now where the business model and the clinical answer are pointed in the same direction, and it is worth understanding how that works before you assume you have received two opinions.

A direct-to-consumer platform whose revenue depends on people receiving a diagnosis, or on an ongoing prescription, is not neutral about the outcome. Nobody involved has to be dishonest for this to shape results. The visit gets short because short visits scale. The instrument becomes a self-report symptom checklist because checklists can be scored automatically. Collateral information is skipped because gathering it from a teacher takes a week of phone tag and cannot be automated. Rule-outs get thin because rule-outs are the expensive part.

What you end up with is not a competing clinical opinion. It is a much smaller amount of information, reported with the same confidence as a large amount of information.

That is the thing I most want parents and adults to hold onto. A fifteen minute visit with a checklist and a comprehensive evaluation are not two opinions. They are two different quantities of data. When they disagree, that is not a tie.

I will say the obvious counterpoint, because it is true. Access is a genuine problem. Waitlists for testing in Delaware can run months. Telehealth reached people who could not otherwise get seen at all, including adults in their forties who spent thirty years thinking they were lazy. The answer to a thin evaluation is not to sneer at the people who took the only option available to them. It is to know what you got.

The errors run in both directions

If this piece only warned about diagnoses handed out too easily, it would be leaving out the half of the problem I see more often.

His grades are fine, so it is not ADHD. She makes eye contact, so she is not autistic. He is polite and sits still, so he is neither. She is a girl, so probably anxiety.

Those are the same failure as the rubber stamp, just aimed the other way. Grades measure output, not effort, and a bright kid can burn four hours producing forty minutes of work and still bring home a B. Politeness is a social skill, not a measure of executive functioning. Eye contact is trainable and frequently trained. Anxiety and ADHD co-occur constantly, and stopping at the first one you find is not a differential, it is a guess with a word attached.

Underdiagnosis is quieter than overdiagnosis. Nobody writes a magazine piece about the kid who never got evaluated. He just gets a reputation instead.

You cannot buy a diagnosis, but you can buy a document

Here is the distinction that actually matters in practice.

A diagnosis is a clinical conclusion supported by convergent evidence across sources and settings, with the plausible alternatives ruled out and the reasoning written down where someone else can inspect it. A document is a piece of paper with a code on it.

Most of the time the difference is invisible. It becomes visible the first time the paper has to do work.

Bring a three page telehealth letter to an Appoquinimink or Smyrna IEP team and watch what happens. The team is not being obstructive. They are required to determine educational impact, and a letter with no teacher data, no observation, no cognitive or academic testing, and no discussion of what else was considered gives them nothing to act on. They will run their own evaluation, and in Delaware they have 45 school days from the signed consent to do it. That is most of a semester.

College disability services offices are stricter, not looser. Many want testing within a defined recency window, standardized scores, and a statement of functional limitation tied to the specific accommodation requested. Extended time is granted on evidence of a processing or attention deficit, not on a diagnosis alone.

Same story with a second opinion, a medication consult with a psychiatrist who wants to see the workup, or a disability claim years later. The document either survives contact with people who know what to look for, or it does not.

That is the whole difference, and it is worth paying attention to whether or not you ever set foot in my office.

What to ask before you schedule anything

Four questions. You can ask them on the phone in under five minutes, and the answers tell you almost everything.

How long is the evaluation, and what happens during it? You want specifics. Hours, sessions, what is administered. A comprehensive evaluation is measured in hours, not minutes, and the person on the phone should be able to describe it without hedging.

What information are you using besides what I tell you? Records, prior testing, report cards, standardized measures, direct observation. Self-report alone is a starting point, not an evaluation. This matters most for adults, who are often the only informant available and who are being asked to remember childhood.

Do you collect teacher or collateral information, and how? For a child, this is close to non-negotiable, because criteria require impairment across settings and you cannot assess across settings from one setting. For an adult, a partner, a parent, or an old report card can do similar work.

What happens if the answer turns out to be no? This is the one. A good evaluator answers it immediately and concretely, because they have said no before and they have a plan for it. You still get a formulation, an explanation of what is driving the difficulty, and a set of recommendations. If the answer is vague, or if it seems like nobody has considered the possibility, you have learned what you needed to know.

You are allowed to ask these questions. You are allowed to ask them of me. The clinicians worth your money will be glad you did, and the ones who are not will tell you something in the way they answer.

The Cliff Notes

You skipped down. Good instinct, and the whole point of putting this here. Here is the argument without the scenic route.

• When two clinicians disagree about ADHD or autism, the disagreement is usually not about the person. It is about how much was looked at.
• Some disagreement is legitimate. ADHD and autism are dimensional, and the diagnostic threshold is a line drawn across a continuous distribution, so a person near that line can land on either side with two careful evaluators.
• Parent and teacher ratings of inattention agree at only around .45. That gap is expected. A strong evaluator interprets it. A weak one picks the informant who confirms what they already thought.
• Most autism disagreement comes from one clinician using an outdated prototype built on young boys, which fails badly for women and for anyone who learned to mask by studying other people.
• A short telehealth visit with a self-report checklist and a comprehensive evaluation are not two opinions. They are two different amounts of information, and when they conflict, that is not a tie.
• Platforms whose revenue depends on the diagnosis or the prescription are not neutral about the outcome. No one has to be dishonest for the structure to shape the result.
• The errors run both directions. Good grades, eye contact, and politeness are routinely used to rule things out, and none of them are valid grounds for doing so.
• You cannot buy a diagnosis, but you can buy a document. The difference shows up the first time the paper meets an IEP team, a college disability services office, or a second opinion.
• A thin letter will not carry an IEP meeting. The district will run its own evaluation, and Delaware allows 45 school days from signed consent, which is most of a semester.
• College disability offices are stricter than school districts, not looser. Most want current standardized scores and functional limitations tied to the specific accommodation.
• Four questions before you schedule: how long is it, what sources besides me, do you gather teacher or collateral data, and what happens if the answer is no.
• The last question is the tell. If nobody can describe what happens when the answer is no, ask yourself what you are actually paying for.

If you are sitting with two answers and no idea which one to trust

That is a reasonable place to be, and it is worth a conversation before you spend money on anything else. At Center for Balanced Living, our evaluations are built to hold up: multiple sources, testing across domains, real rule-out reasoning, and a report written so a school team or a treating provider can use it. Sometimes the answer is that the diagnosis fits and the first evaluation was simply brief. Sometimes it is something else entirely.

If we are not the right fit, or if the wait does not work for what you are dealing with, tell us and we will point you somewhere else. The Parent Information Center of Delaware is also a genuinely useful place to start if the question is about school and you are not sure what to ask for.

This post is educational and is not a substitute for individual clinical advice or an evaluation.