You’ve probably argued with a doctor about ADHD treatment before. Or spent $80 on supplements because someone on Reddit said they helped. Or quietly wondered whether what you’re doing is actually backed by anything real, or just the loudest voice in an ADHD Facebook group.

I have. Multiple times.

In February 2026, researchers from Université Paris Nanterre, Institut Robert-Debré, and the University of Southampton published the most rigorous synthesis of ADHD treatment evidence ever compiled: 221 separate meta-analyses, covering 31 interventions, across 24 clinical outcomes. Then, instead of burying it in a journal, they built a free public website where you can look up any treatment, see how strong the evidence is, and filter by your age group and what outcome you actually care about.

The tool is at ebiadhd-database.org. Bookmark it before you read another word.

TL;DR

What	Details
The study	221 meta-analyses, 31 interventions, published in The BMJ (November 2025)
Strongest evidence	Stimulant meds (kids + adults); CBT (adults); atomoxetine (adults)
Free tool	ebiadhd-database.org — filterable by age, outcome, treatment
The catch	Almost all evidence is short-term only. Long-term data barely exists for any intervention
Who it’s for	Anyone making decisions about ADHD treatment — you, your doctor, your kid’s school

The actual breakthrough: For the first time, you can walk into a psychiatrist’s office with the same evidence database your doctor might use, and have a real conversation about why you’re choosing or reconsidering a particular treatment.

Why This Is Different From the Last ADHD Study You Saw

ADHD research gets shared constantly, and most of it is a single trial with 40 participants in one country, run over six weeks.

This isn’t that.

The research team didn’t run a new experiment. They synthesized 221 meta-analyses (each of which was itself a synthesis of multiple clinical trials). Then they re-ran the statistics across all of them using a standardized pipeline, so you can actually compare the evidence for, say, methylphenidate against the evidence for omega-3 supplements without comparing apples to very different oranges.

Professor Samuele Cortese called it “the first platform in the world” to provide “rigorous synthesis of the available evidence” in a publicly accessible format. That’s a big claim. It’s also accurate.

What makes it useful for non-researchers is that the interactive database translates statistical concepts into plain language. You don’t need to know what a GRADE rating is to understand that “high certainty evidence” means something different from “low certainty evidence.”

How to Actually Use the Tool (Step by Step)

The wizard at ebiadhd-database.org/wizard is the fastest way in. Here’s what the interface asks:

First: choose your mode. You can either compare two to five interventions side-by-side, or focus on a specific profile (your age group, your symptoms, your situation). If you’re evaluating a specific treatment your doctor mentioned, “compare interventions” is more useful. If you’re starting from scratch, “focus on a profile” builds a filtered view.

Second: select the age group. The evidence differs significantly between children and adults. A treatment with strong evidence in children may have weak evidence in adults, and vice versa. The tool separates these.

Third: navigate the matrix. The main evidence matrix is a grid where each cell is a treatment-outcome combination. Click any cell to see the effect size, the number of underlying studies, and the certainty rating. This is the part that takes five minutes to get used to but becomes intuitive fast.

I spent 20 minutes on the tool and came out with clearer answers than I’ve gotten from hours of reading individual studies. That’s the point.

The Part of the Evidence That Usually Gets Buried

The treatment rankings are covered in depth in the evidence-based ADHD productivity strategies guide. The short version: medications (stimulants and atomoxetine) and CBT for adults hold up. Most supplements don’t.

What’s worth highlighting here is the angle that gets underreported: the evidence matrix doesn’t just show you what works. It shows you what’s been studied, how rigorously, and for how long.

That last part matters. The study found no high-certainty, long-term evidence for any intervention. Zero. Not because long-term treatment doesn’t work, but because the research trials are overwhelmingly short: weeks to months, not years. If you’ve been on methylphenidate for five years, the clinical trial evidence for that specific situation is much thinner than you’d expect.

This isn’t a reason to panic. It’s a reason to have better conversations.

The Shared Decision-Making Frame (And Why It Changes Things)

The platform was built explicitly for shared decision-making. That framing is deliberate, and it’s worth sitting with.

Shared decision-making means the clinician brings the evidence, you bring your preferences and values, and the two of you arrive at a decision together. Not the doctor deciding for you. Not you demanding a specific medication because you read about it online. A real conversation with both sides informed.

For ADHD adults, this matters more than it might for other conditions. Treatment decisions are deeply personal. Medication side effects vary wildly by individual. CBT accessibility is uneven. Some people can’t take stimulants for reasons that have nothing to do with their ADHD severity.

The tool gives you something concrete to bring to that conversation. You can look up exactly which outcomes a treatment showed evidence for, and which ones it didn’t. If you care primarily about emotional regulation rather than attention scores, you can filter for that. If you’re a parent trying to evaluate options for a 10-year-old versus a 35-year-old, the age group filter changes the results substantially.

This is new. Before this tool existed, accessing that level of nuance required paying for journal access and knowing how to read a forest plot.

What This Means If You Manage ADHD Without Medication

I don’t take medication. The brand is medication-agnostic by design, because plenty of people manage ADHD without pharmaceutical support: by choice, by necessity, or because meds don’t work well for them.

The tool is just as useful in that situation. Look up the non-medication interventions and examine their evidence base honestly. Here’s what you’ll find:

CBT for adults: Moderate-to-high certainty evidence with medium effect sizes. This is the strongest non-medication option by a significant margin. If you haven’t engaged with ADHD-specific CBT, the evidence says this is worth pursuing.

Mindfulness: Large effect sizes in some studies, but low certainty evidence. The studies are small and inconsistent. Worth trying as an add-on. Not worth relying on as your primary strategy.

Exercise: Real mechanistic reasons to include it; evidence quality is limited. Treat it as a valuable part of a system, not a standalone answer.

Neurofeedback: Persistent in ADHD communities. Low certainty, effects largely disappear against active controls. Worth knowing this before spending $150/session on it.

Omega-3: Conflicting, low-certainty evidence. For core ADHD symptoms, the current research doesn’t justify it as a primary spend.

Knowing this doesn’t mean abandoning tools and strategies that work for you. The best AI coaching apps and working memory tools I cover elsewhere on this site aren’t clinical interventions. They’re accommodations. They operate on a different layer than what this database measures.

The One Thing the Tool Can’t Do

The evidence matrix ranks treatments at a population level. It tells you that, across thousands of study participants, methylphenidate showed X effect on attention symptoms with Y certainty.

Your brain is not thousands of study participants. It’s yours.

ADHD presentation varies more than almost any other diagnosis. Inattentive subtype responds differently than hyperactive-impulsive. Adults diagnosed in their 30s often have decades of compensatory strategies layered on top. Coexisting anxiety, depression, or autism changes the treatment picture significantly.

The tool handles this partially (you can filter by outcome type, age group, and intervention), but it can’t account for your specific combination of factors. That’s not a flaw in the research. It’s just the limit of population-level evidence applied to individual decisions.

Use the tool to inform the conversation. Let the conversation account for you specifically.

A 15-Minute Exercise Worth Actually Doing

Step 1: Go to ebiadhd-database.org/wizard. Select your age group.

Step 2: Look up whatever you’re currently doing or considering: medication, CBT, supplements, neurofeedback, anything. Note the certainty rating and effect size for the specific outcome you care about (attention symptoms, emotional regulation, daily function are all different).

Step 3: Write one sentence about what you learned. “The evidence for [X] is [strong/moderate/low] for [outcome] in [my age group].” That sentence is what you bring to your next psychiatry or therapy appointment.

Step 4: If you found a gap between what you’re doing and what the evidence supports, that’s the conversation to have with a clinician, not by cold-stopping anything on your own.

That’s it. Fifteen minutes. One sentence. One better conversation.

Where This Fits in Your Broader Stack

The evidence in this database is about clinical interventions. What this site covers (gamified task apps, body doubling tools, AI task-breakers) operates at the accommodation layer. Both matter. Neither replaces the other.

Think of it this way: CBT might help you understand why task initiation keeps failing. An AI task-breaker helps you on Tuesday morning when initiation is failing right now. You need both the structural understanding and the practical tool.

The database tells you which structural interventions have the evidence to back them. What you build around them (your apps, your systems, your routines) is where you have more flexibility to experiment based on your specific brain.

---

Start here: Bookmark ebiadhd-database.org. Use the wizard. Look up one thing you’re currently spending money or time on. That’s the action. Everything else follows from having better information.

---

The study, “Benefits and harms of ADHD interventions: umbrella review and platform for shared decision making,” was published in The BMJ in November 2025. The interactive platform launched publicly in February 2026. Full methodology and data available at ebiadhd-database.org.