How do I reduce screen failures in my clinical trial?
By Bryan Manning, Founder · Updated August 2026
Direct answer: Most screen failures are preventable, because most of them happen on criteria that a two-minute questionnaire or a short phone call could have caught. The fix is layered screening before the site ever sees the patient: a web screener built from the actual inclusion/exclusion criteria, then a live phone screen that checks again. In our programs that double-screen cut screen failures by 18%, and 89% of the patients we walk into a site go on to sign the consent form.
Why screen fails hurt more than they look like they do
A screen failure isn’t just a lost patient. It’s a coordinator’s prep, a screening visit slot, lab costs, and an afternoon of site goodwill, spent on someone who was never going to randomize. Sites remember which vendor sent them referrals that fail. When your best sites start deprioritizing your study, the screen-fail line item has quietly become an enrollment problem.
The three layers that catch failures early
Layer one: a web screener that isn’t afraid to say no
Build the screener from the real criteria, not a softened version designed to maximize passes. In our programs the web screen removes about 60% of applicants. That feels brutal until you realize every one of those people would have failed later, at higher cost, after wasting more of everyone’s time, including their own.
Layer two: a phone screen that checks again
A human screener walks through the criteria a second time, catches the ambiguities a form can’t (medication histories, comorbidities people forget to type, willingness to actually travel), and answers questions. The phone screen removes another 77% of those remaining in our programs. Two independent filters, two chances to catch a disqualifier before a site visit gets scheduled.
Layer three: records before the visit
Where possible, retrieve medical records before the on-site screen, so the site can verify the hard criteria in advance. It also means the visit starts with confirmation instead of discovery.
What you can’t prevent, honestly
Some disqualifiers only surface on site: labs, imaging, in-person assessments. In one of our recent programs, 23.9% of on-site screens still failed even after double-screening. That’s the irreducible part. The goal isn’t zero screen fails, it’s making sure every failure is one that genuinely could not have been caught earlier.
The incentive problem underneath it all
If your vendor is paid per referral, screen failures are your problem, not theirs. Every marginal referral is revenue for them and burned capacity for you. We’re paid on enrolled patients only, which means a screen failure costs us, not just you. Ask any vendor what a screen fail costs them. The answer tells you whose problem it is.
Frequently asked questions
What’s a normal screen failure rate?
It varies enormously by indication and criteria tightness, which is why the trend matters more than the number. If your rate isn’t falling month over month, nobody is learning from the failures. Ask for fail reasons, ranked, every month.
Won’t aggressive pre-screening slow down enrollment?
The opposite. Site capacity is the bottleneck in most trials. Filling screening slots with patients who can actually randomize is faster than filling more slots with patients who can’t.
Should we loosen criteria to reduce screen fails?
That’s a protocol and science question, not a recruitment tactic. Sometimes an amendment genuinely is the answer, but decide that from ranked fail-reason data, not frustration.
Enrollment on the line?
Most firms recruit patients. We deliver enrollments. We take ownership of the whole path, from first click to site visit, and we only get paid when patients enroll.