What to look for in a patient-finding platform

I talk with a lot of teams evaluating patient-finding platforms, and most demos look impressive. That’s exactly the problem: the demo is the easy part. What separates a platform you can build a program on from one that disappoints six months in shows up in questions you have to ask deliberately.

Here’s the checklist I’d use if I were on the other side of the table.

1. Is it reproducible?

Ask: if you run the same definition twice, do you get the same patients? If the answer is anything but an unqualified yes, stop there. A cohort that drifts between runs can’t support a clinical or regulatory decision. Reproducibility is the price of admission, not a feature.

2. Who governs the definitions, and can you see the change history?

A disease definition encodes real clinical judgment. Ask who curates it, who signs off, and whether changes are versioned and auditable. “An AI generates it from a prompt” is a red flag. You want criteria reviewed by key opinion leaders and a record of how they’ve evolved.

3. Where does the data go?

The right answer is nowhere it doesn’t already live. Favor platforms that analyze data in place, inside your governed environment, so protected health information isn’t copied out. Ask specifically about the data foundation: does it work with standard FHIR and the real-world data you already have, or does it demand a migration project first?

4. Can it explain itself?

A ranked list with no reasoning is a liability. For every patient surfaced, you should be able to see why: the specific evidence that matched the definition. Explainability is what lets a clinician trust the output and act on it.

5. Is it built to last?

Finally: is this a one-time analysis or an ongoing capability? Definitions need to stay current, and value should compound. A platform that’s only as good as its launch-day demo isn’t worth the integration effort.

None of these questions are about who has the cleverest model. They’re about whether you can stand behind the results, which in rare disease is the only thing that matters. If you’re weighing these trade-offs for your own program, we’re happy to compare notes.

— Travis Musick, Co-founder & CRO, Sagacity Diagnostics

This article reflects the author’s perspective and is not medical or investment advice.