All articles
Recruitment StrategyNovember 13, 2026·3 min read

Why Your Patient Registry Isn't Producing Enrollments

Most research sites have a patient registry. Far fewer have one that produces enrollments. Here's how to design fields around real protocol criteria, handle consent, and fight data decay.

By Trialflow Team

Most sites have a registry. Far fewer have a registry that produces enrollments. The difference usually isn't the size of the list — it's whether the data in it maps to the questions protocols actually ask, and whether anyone in the database still remembers who you are.

A registry is a matching tool, not a mailing list

The common failure mode is collecting name, email, phone, and a vague "interested in diabetes studies." Six months later a protocol lands requiring A1c between 7.5 and 10.5, metformin monotherapy, BMI under 40, and no GLP-1 in the prior 90 days. You can't query any of that, so you blast the whole diabetes segment, get a flood of replies, and burn coordinator hours on pre-screens that fail on criteria you could have captured upfront.

Build your fields backward from the protocols you actually get. Pull the last two years of feasibility questionnaires in your top therapeutic areas and list the criteria that appeared repeatedly. In most site datasets a handful of variables do nearly all the filtering work:

  • Diagnosis plus approximate diagnosis date
  • Current medications in the relevant class, and prior washout history
  • Most recent key lab value and roughly when it was drawn
  • Height, weight, smoking status
  • Prior trial participation and the date of last study exit
  • Travel radius and realistic visit availability

That's maybe fifteen structured fields. Fifteen well-chosen fields beat fifty free-text notes every time.

Consent for the registry, separately

Registry enrollment needs its own IRB-approved consent or authorization — it is not covered by the consent for the study a person screen-failed out of. Make that document do real work: permission to retain data, permission to re-contact for future unspecified studies, permission to share within your site network if applicable, and a clear, easy withdrawal path. Get it signed at the moment of highest engagement, which is usually at the end of a screening visit or a screen-fail conversation, not three weeks later by email.

Decay is the real enemy

Registry data goes stale fast. Phone numbers change, patients move, labs drift, medications get swapped. A record older than about eighteen months with no contact in between is closer to a guess than a lead.

Practical countermeasures:

  1. Timestamp every data point, not just the record. Knowing the A1c is from February matters more than knowing the record was created in February.
  2. Run a light refresh cycle. A short annual check-in — two or three questions, one click to confirm nothing changed — keeps records usable without feeling like surveillance.
  3. Send something of value between asks. A brief update on what studies ran at your site, plain-language results when sponsors permit, a seasonal health note. People who only hear from you when you want something stop opening your emails.
  4. Tag outcomes, not just contacts. Record why someone didn't enroll last time. "Declined due to visit frequency" and "ineligible on renal function" lead to very different future outreach.

Measure it like a channel

Track registry-sourced referrals separately from ads, physician referrals, and walk-ins. Watch pre-screen-to-consent rate specifically. If registry contacts convert worse than cold advertising, your fields are wrong or your data is too old — and more contacts won't fix either.

A registry built this way quietly becomes the cheapest enrollment source you have, and the one competitors can't buy.

Enrolling studies shouldn't be this hard

Trialflow gives research sites one platform for lead management, AI trial matching, and recruitment analytics.

Request a Demo