A screen failure occurs when a patient who has begun formal screening — typically after informed consent and at least some screening assessments — is found to be ineligible for the trial. In Phase II and Phase III trials, screen failure rates of 30–40% are typical at sites without structured pre-screening. At sites relying primarily on physician referrals without EHR-based eligibility filtering, rates above 50% are not unusual.
The operational cost is not trivial. Depending on protocol requirements, a screen visit may involve a physical exam, ECG, fasting blood draw, patient questionnaires, and 60–90 minutes of coordinator time for consent and data entry. A screen failure represents the full cost of that effort with no enrollment outcome. Sites running 50–60 screen failures to enroll 30 subjects are consuming substantial operational capacity on unproductive visits.
Classify before you reduce
Screen failures fall into three categories. Protocol-ineligible failures occur when a patient did not meet objective I/E criteria that were identifiable from EHR data before the screen visit. Screening-ineligible failures occur when the patient met EHR-confirmable criteria but failed a criterion assessable only at the screen visit — an ECG finding, a live physical exam measurement, a fasting lab value not available pre-visit. Patient-withdrawn failures occur when an eligible patient declines to continue after consent.
Protocol-ineligible failures are directly addressable through pre-screening workflow changes. If a patient fails screen because they had prior insulin use within the exclusion window, and that information was available in the EHR before the screen visit was scheduled, that is a pre-screening failure. The intervention is upstream: better EHR data extraction before the outreach call, not better screening form design.
Classifying failures by reason takes approximately one hour per trial per quarter if data is captured in the CTMS. Sites not currently capturing screen failure reason codes should add that field — it is among the most actionable data elements in enrollment operations.
EHR pre-screening filters: sequencing by discriminating power
Not all I/E criteria are equally efficient as pre-screening filters. The most efficient are those with low pass rates in the target population, confirmable from EHR data without live assessment, and evaluable without clinical judgment. For a T2DM trial with HbA1c inclusion and eGFR exclusion criteria, the eGFR threshold is often the more powerful first filter — a proportion of the T2DM population has CKD stage 3 or higher, and querying eGFR first eliminates that fraction before any other criteria are evaluated.
Medication exclusions are often the most powerful pre-screening filters in cardiometabolic and rheumatology trials. A patient on a GLP-1 receptor agonist is excluded from a GLP-1 mechanism trial regardless of lab values — running the medication exclusion filter first reduces the population requiring lab value evaluation. EHR MedicationRequest and MedicationStatement resources should be queried with a full list of RxNorm identifiers for the excluded drug class, not just a few brand-name matches.
The efficient pre-screening workflow applies filters in order of discriminating power, not in the order criteria appear in the protocol document. Each filter reduces the population evaluated by the next, minimizing the chart review time per ineligible patient identified.
The pre-screen outreach call
The outreach call to a candidate patient is the most efficient point to surface exclusion criteria requiring patient-reported information. An outreach call that spends four minutes on scheduling and one minute on eligibility screening misses the opportunity to prevent a screen failure before the screen visit is booked.
Patient-reported information is most useful for: current medication use the patient may not have in the EHR, recent clinical events since their last encounter, and schedule or travel constraints affecting protocol visit compliance. A coordinator who discovers on the call that a patient started insulin two months ago has prevented a screen failure at zero additional cost. That same information discovered at the screen visit after consent has cost the site a full visit workup.
The outreach script should be reviewed against the trial's historical screen failure reasons. If prior failures included a disproportionate number due to a specific exclusion criterion that patients could have reported on the phone, that question should be added to the script.
Coordinator judgment stays central
Structured pre-screening gives coordinators better information faster — it does not replace coordinator judgment. An EHR-based pre-screening tool that returns a ranked candidate list is providing the coordinator with criterion-level match detail: which criteria are confirmed from EHR data, which have data gaps, which require manual review. The coordinator evaluates that information alongside their clinical knowledge of the patient population and makes outreach and prioritization decisions accordingly.
A patient with an eligibility confidence score of 78 because their BMI was last recorded 14 months ago might be high priority (if the coordinator knows from prior encounters that the patient's weight has been stable) or lower priority (if the coordinator knows the patient has been actively working on weight loss). That judgment is irreplaceable and should not be bypassed by the tool.
Protocol amendment impact on screen failure patterns
Protocol amendments that modify I/E criteria have direct effects on screen failure rates. An amendment that broadens an inclusion criterion will increase the candidate pool but may not decrease screen failure rates if the newly eligible population has characteristics that trigger other exclusion criteria at higher rates.
Before investing outreach effort in an expanded pool post-amendment, sites should re-run the population analysis against the updated criteria. The CTMS screen failure data is the most valuable resource for protocol amendment discussions with the sponsor: a site that can show 35% of its screen failures in the prior quarter were due to a single exclusion criterion — with specific counts — is in a much stronger position to support a targeted amendment request than a site that reports enrollment is behind target.
Realistic expectations
Structured pre-screening will not eliminate screen failures. Some are unavoidable — criteria requiring live assessment cannot be pre-screened, patient health status changes between pre-screening and the screen visit, and some criteria genuinely cannot be evaluated from EHR data alone. A site that reduces its protocol-ineligible screen failure rate by 50–60% through better pre-screening has achieved a meaningful operational improvement even if the overall screen failure rate remains above 20%.
The benchmark is not a zero screen failure rate. It is a distribution where protocol-ineligible failures account for a small minority of total failures, and the majority occur due to assessments genuinely not evaluable before the screen visit. Getting there requires accurate screen failure classification in the CTMS and a pre-screening workflow that systematically applies EHR-confirmable criteria before the outreach call.