Turning Screen Failures Into Next Quarter's Enrollments

Most sites treat a screen failure as a dead end. The chart gets closed, the sponsor log gets a reason code, and the person who drove 40 minutes and gave eight tubes of blood never hears from the site again. That's a waste of the most expensive asset in recruitment: a consented, phenotyped, motivated human being who already knows where your parking lot is.

Screen failures are the warmest pool you will ever have. They understand research, they tolerated a visit, and you have real data about them. The work is building the habit and the infrastructure to use that.

Capture the reason, not just the code

Sponsor logs want a category. Your own database needs detail. "Lab out of range" tells you nothing in six months. "HbA1c 6.3, needed 6.5" tells you this person may qualify next year, or qualifies today for a prediabetes protocol.

At minimum, record:

• The specific value or finding that failed, with the threshold it missed
• Whether the failure is permanent (prior organ transplant), temporary (recent steroid course, washout not met), or borderline (one point outside a lab window)
• Medications that caused exclusion and whether the prescriber might adjust them
• Diagnoses discovered during screening that weren't in the original referral

That last one matters more than people expect. Screening frequently surfaces comorbidities a participant never mentioned at pre-screen. A failed cardiology screen often produces a confirmed sleep apnea or CKD diagnosis that makes someone a strong candidate elsewhere.

Classify into three buckets at closeout

Build the triage into your visit-closeout checklist so it happens while details are fresh.

1. Re-screenable for this study. Washout, transient lab, scheduling conflict. Set a concrete recall date and own it. Many protocols allow one re-screen; sites commonly leave those slots unused simply because nobody tracked the eligibility date.
2. Candidate for another current or upcoming study. Tag by condition and key values so this person surfaces in your next feasibility review.
3. Not eligible for research now. Still worth a respectful close and, with permission, retention in your registry for future indications.

Make the closing conversation count

How you deliver the news determines whether they answer the phone later. Script it plainly: explain the specific reason, state clearly that it reflects the protocol's narrow rules and not their health or effort, and ask directly whether they would like to be contacted about other studies. Document that preference. Give them their screening labs if the protocol allows — it is the single most appreciated thing sites do at this moment.

Avoid vague promises like "we'll keep you in mind." Say what you will do: "We have two studies in this area starting in the spring. May I call you in March?"

Build the recall workflow

A tagged database only pays off if someone queries it. Make a new-study startup checklist item: before spending on advertising, run the screen-failure pool against the new inclusion criteria. In most site datasets the first few enrollments of a new protocol can come from prior screen failures, at a fraction of the cost per randomization.

Also review your failure patterns quarterly. If 60% of failures come from one lab or one medication exclusion, you have a pre-screening problem, not a bad luck problem. Move that question earlier and stop paying for visits that were never going to work.

Screen failures are data. Treat them that way and they become next quarter's enrollments.