Screen Failure in Clinical Trials: Why One Average Rate Misleads

Of 72,178 participants screened for phase 3 and 4 industry-funded trials of chronic conditions, 24,733 (34%) never entered a treatment group [1]. For a sponsor, each of those is screening work with no randomized participant at the end, and a pooled average says little about whether one study's rate is a problem. A rate becomes actionable once its denominator is fixed and each failure is coded to protocol design, site execution or recruitment targeting, and because the cost of failures can fall on sites unless the contract says otherwise, the budget belongs in the same conversation. Refero does not run recruitment campaigns. We vet and match the clinical research marketing agencies that do, so what follows is written for the buyer, from what those vendors need and see.

Screen failure in clinical trials at a glance

Screen failure is concentrated in a few criteria and reasons, yet the screening work behind it is largely unfunded.

• 54.9% of screen failures across 15 randomized trials at one oncology unit were people who missed an inclusion criterion [2]
• 55 of 59 people re-screened in a multiple sclerosis trial (93%) had resolved the original exclusion and were randomized [3]
• Broadening three lung cancer criteria would have cut exclusions from 47.7% to 1.5% of 10,500 patients [4]
• Eligibility screening cost $129.15 to $336.48 per enrolled patient at one cancer center, and trial contracts typically did not pay for it [5]
• 88% of sites say screen failures are not sufficiently covered, in a survey by a site trade association [6]

What Is a Screen Failure in Clinical Trials?

A screen failure is a person who has consented to a study and been assessed against its protocol but does not enroll, because they miss an eligibility criterion or withdraw before randomization. In common operational practice the term starts at consent, so people turned away earlier, at a records check or a short questionnaire, are pre-screen failures.

That gives the funnel four points: referred, pre-screened, consented and screened, randomized. The pre-screen line and the consent line mark where each kind of exclusion falls, but the two lines are a convention rather than a rule. Staged recruitment guidance separates pre-screening, a lighter check against records or a brief questionnaire, from formal screening [7], while the 2024 BMJ Medicine meta-analysis counts any failure to enroll after entering screening [1]. Because the two sources differ on whether a pre-screen exclusion counts, every rate depends on where its denominator starts.

How to Reduce Screen Failure in Clinical Trials

Screen failure falls when each failure is coded to its source (protocol design, site execution or recruitment targeting) and the matching lever is pulled, because a restrictive criterion, a site-process gap and a poorly aimed outreach message each respond to a different fix.

Calculate the rate on the consent line so every study compares

Use one formula on every screening log:

Screen failure rate (%) = participants who screen failed ÷ participants consented and screened × 100

For the denominator, count everyone who signed consent and began screening, and report pre-screen exclusions as a separate number beside the rate. In a hypothetical study with 160 referrals, 100 are excluded at pre-screening, 60 consent and are screened, 21 of those 60 fail and 39 are randomized. The rate is 21 ÷ 60 × 100 = 35%, yet dividing the same 21 by all 160 referrals gives 13%, and because only the starting point of the count separates the two figures, any vendor's or site's low figure needs its denominator.

A re-screened person should be counted once, under one participant ID, and as a failure only if the final attempt fails. That convention matters because many failures are temporary: in the CombiRx multiple sclerosis trial, 55 of 59 people re-screened (93%) had resolved the original exclusion and were randomized [3]. Record the reason code in the same screening log row as the outcome, since the rate says how many failed and only the codes say why.

Code every failure reason, then sort it into one of three buckets

If the log has no reason codes, add them before changing anything, because the bucket shows where the fix lives. For each coded failure, mark one bucket and pull its lever:

• Protocol design: the person missed a lab value, biomarker, imaging or history criterion. Lever: review that criterion.
• Site execution: the failure was logistical, such as an inadequate tissue sample, an out-of-window test or missing paperwork. Lever: fix the process, or screen for it earlier.
• Recruitment targeting: the person declined, would not travel or was referred without a basic fit. Lever: change who is reached and how the study is described.

Published audits show why the buckets matter, though each figure holds only for its own setting. Across 15 randomized trials at one medical oncology unit in India, 54.9% of screen failures missed an inclusion criterion and 22.2% declined to participate [2], so the first lever there is the criteria and the second is targeting. In breast cancer trials at 14 Indian sites, 80% of failures missed selection criteria, mostly HER2-negative status, while 10.5% failed for logistical reasons, almost all inadequate tissue samples [8]; the selection-criteria failures sit in design and targeting, the tissue-sample failures in site execution. At one phase 1 site in China, abnormal laboratory results accounted for 43.25% of healthy-volunteer failures [9], which only a post-consent sample could reveal.

Review criteria that exclude without protecting safety

If most failures trace to one lab, biomarker or imaging criterion, review that criterion first. In its 2020 cancer-trial eligibility guidance, FDA states that unnecessarily restrictive eligibility criteria may slow patient accrual, limit access to trials and produce results that do not fully represent the patients who will use the drug [10], and it applies the same logic across therapeutic areas in its current guidance, "Enhancing Participation in Clinical Trials: Eligibility Criteria, Enrollment Practices, and Trial Designs" (December 2025) [11]. ASCO and Friends of Cancer Research name treatment washout periods, concomitant medications, prior therapies, laboratory reference ranges and test intervals, and performance status as the criteria to modernize [12].

The effect of broadening criteria can be large: in a retrospective analysis of 10,500 patients with advanced lung cancer, reported from an ASCO 2019 abstract, traditional criteria on brain metastases, prior cancer and creatinine clearance would have excluded 47.7%, against 1.5% under the broadened criteria [4]. That shows how much exclusion sits in a few criteria, though it measures patients in records rather than a trial's screen failure rate.

Fund protocol-specific pre-screening, and treat its effect as unproven until shown

Sponsors often expect pre-screening and rarely fund it, and the evidence supports a bounded version. One position, common among sites and vendors, holds that a scripted, IRB-reviewed eligibility interview before consent removes avoidable failures cheaply. FDA guidance supports part of that: a limited review of records to establish eligibility does not require consent if it records only eligibility and contact information, although collecting more may [13]. Recruitment materials need IRB review [14], and the HHS Office of Inspector General recommends IRB review of trial-specific pre-screening questionnaires that collect health information [15].

The second position is that none of this shows pre-screening lowers failures by a known amount or lets non-clinical staff decide eligibility, so clinical calls stay with the investigator or delegated staff and the effect must be shown per protocol. Some failures also cannot be pre-screened away, because eligibility rests on post-consent labs or imaging, as the phase 1 and multiple sclerosis data show [9][3]; a zero target is therefore the wrong goal. This section is not legal advice; confirm pre-screening procedures with regulatory counsel and the IRB of record.

What Screen Failure Rate Is Normal?

No single rate is normal, because published single-setting rates run from 15.6% in one multiple sclerosis trial [3] to 61.5% among healthy volunteers at one phase 1 site [9], and the pooled rate for chronic-condition trials describes neither.

Published screen failure rates by setting; each row uses its own authors' definition, so the rows show range rather than a target.

A rate well above the range for your indication points to protocol criteria or targeting, so run the three-bucket reason-code sort; a rate inside the range means the screen failure cap and the screening budget should allow for it rather than treat each failure as an exception.

Who Pays for Screen Failures in Clinical Trials

Once the informed consent form is signed, the screening bill can fall on the site unless the contract says otherwise: in one prospective study at an academic cancer center, the contracts supporting the trials typically did not compensate the cost of eligibility screening [5]. That study put the cost at $129.15 to $336.48 per enrolled patient by trial type, so the figure spreads the effort spent on people who never enrolled across those who did.

A published site-budget template lists screening procedures as their own section and adds a separately invoiced screen-fail fee (its $1,000 is a sample figure, not a benchmark), and it states that sponsors must fund protocol requirements that exceed standard care [16].

A screen failure cap limits how many failures, or what ratio of them, the sponsor will pay for. In one practitioner's view, sponsors use the cap as a safety net against paying for random screenings, and the cap in turn pushes sites toward careful pre-screening, often with a sponsor-provided IRB-approved script [17]. Sites see the other side: the Society for Clinical Research Sites (SCRS) says a site that hits a cap and is paid in part or not at all is out of pocket for work already done [6].

For a buyer, the survey points to two budget lines to fund explicitly: pre-screening (38%) and screen failures (36%) were the areas most often named as underfunded in the 2024 SCRS global survey, whose respondents were 63% sites, 21% CROs and 17% sponsors [18]. The figures are self-reported by a site trade association, so they show where sites feel the gap rather than measure it, and they are why the contract questions below ask about both lines.

What a Screen Failure Costs Sites and Patients

A screen failure costs the site coordinator time, clinic and imaging slots and lab work that cannot be recovered: in one case reported in an industry analyst comment, site staff spent 25 hours screening a single patient who was then rejected for not meeting inclusion criteria, with no compensation for failed screening [19]. A run of failures at a new site can also read to a principal investigator as poor site performance when the protocol or referral is the cause.

Patients bear a quieter cost: someone turned away after consent has already given time, travel and sometimes time off work, and in the Indian oncology audit 6.2% of screen failures were unwilling to stay near the site [2]. Fewer avoidable failures spare patients that cost and leave more of the referred pool to reach randomization, which bears on the enrollment speed described in the clinical trial recruitment benchmarks.

What to Ask and Budget Before You Sign

The rate a site or vendor reports depends on where it starts counting and on who funds the work in between, so the questions belong before the contract.

Ask every site network and recruitment vendor for pre-screen and screen failure counts separately

Ask this when a reported rate is low but randomization is slow, because that pattern suggests exclusions are happening before the consent line. Request:

• Counts at each funnel point: referred, pre-screened, consented and screened, randomized
• Pre-screen exclusions and screen failures with reasons by code
• How pre-screening is scripted, who performs it, and whether the IRB reviewed it
• How re-screens are counted

A vendor offering remote or hybrid pre-screening should show enrollment-speed and dropout evidence of the kind summarized in the decentralized trial evidence.

Budget pre-screening and set the screen failure cap from the indication's range

Use this when a draft contract caps screen failures but has no pre-screening line. Fund pre-screening explicitly, then set the cap from the expected range for the indication and protocol, and leave room for failures that only post-consent assessments reveal. A cap set below the realistic failure count leaves sites out of pocket: in the same 2024 SCRS survey, 42% of sites said screen failure terms never or rarely cover the actual number of failures [18].

If you would rather not run that evaluation alone, Refero screens healthcare marketing and patient recruitment agencies against five published criteria and introduces up to three that fit your brief. It is free for buyers, and there is no obligation to hire anyone we introduce. Tell us what you need.

Frequently asked questions

What does failed screening mean in a clinical trial?

Failed screening means a person consented to a study and was assessed but did not meet one or more protocol criteria, withdrew or declined before randomization, so they are not enrolled. It says nothing about their health outcome, only that they do not fit this protocol.

What is the goal of patient screening?

Screening confirms eligibility and safety before randomization, so the trial enrolls the population its protocol was designed for. It protects participants from a treatment they are unsuited to receive, and it protects the data, because a population that strays from the protocol weakens the results.

Can a patient who screen fails be re-screened?

Only when the protocol allows it, usually after the disqualifying condition has changed. The new attempt is linked to the first under the same participant ID. Many failures are temporary, as one multiple sclerosis trial showed when most people it re-screened had resolved the original exclusion [3].

Does a pre-screen exclusion count as a screen failure?

Usually not, because it happens before consent and before formal screening begins. Usage varies, however: some published analyses count any failure after entering screening [1], while staged recruitment guidance separates the two stages [7]. Ask any site or vendor which denominator its rate uses before comparing it with a benchmark.

What screen failure rate is normal for a clinical trial?

No single rate is normal. Published single-setting rates range widely by indication, phase and definition [1][3][9], so a blended average is a weak target. Set expectations from the indication, the protocol and the denominator, then use the reason codes to judge whether the excess comes from criteria, site execution or targeting.

Sources

• BMJ Medicine. Participant characteristics and exclusion from phase 3/4 industry funded trials of chronic medical conditions: meta-analysis of individual participant level data. BMJ Medicine, 2024. Link
• ecancermedicalscience. Audit of screen failure in 15 randomised studies from a low and middle-income country. ecancermedicalscience, 2022. Abstract, Full text
• Recruitment of participants to a multiple sclerosis trial: The CombiRx experience. Clinical Trials, 2014. Link
• Amorosi D. Broader inclusion criteria could double the number of patients with lung cancer eligible for clinical trials (reporting Harvey et al., ASCO 2019 abstract LBA108). Healio, 2019. Link
• Penberthy et al. Effort Required in Eligibility Screening for Clinical Trials. Journal of Oncology Practice, 2012. Link
• Society for Clinical Research Sites. Site Finances Whitepaper (2023 survey). SCRS, 2024. Link
• CASRAI. Clinical Trial Patient Recruitment Guide. CASRAI, 2026. Link
• Reasons why patients fail screening in Indian breast cancer trials. Perspectives in Clinical Research, 2015. Link
• Analysis of the reasons for screening failure in phase I clinical trials in China. Annals of Translational Medicine, 2021. Link
• U.S. Food and Drug Administration. Cancer Clinical Trial Eligibility Criteria: Minimum Age Considerations for Inclusion of Pediatric Patients. FDA, 2020. Link
• U.S. Food and Drug Administration. Enhancing Participation in Clinical Trials: Eligibility Criteria, Enrollment Practices, and Trial Designs. FDA, December 2025. Link
• The ASCO Post. ASCO and Friends of Cancer Research Recommend Expanding Patient Access to Cancer Clinical Trials by Further Broadening Eligibility Criteria. The ASCO Post, 2021. Link
• U.S. Food and Drug Administration. Informed Consent: Guidance for IRBs, Clinical Investigators, and Sponsors. FDA, August 2023. Link
• U.S. Food and Drug Administration. Recruiting Study Subjects. FDA, 1998. Link
• Office of Inspector General, U.S. Department of Health and Human Services. Clinical Trial Web Sites: A Promising Tool to Foster Informed Consent (OEI-01-97-00198). HHS OIG, May 2002. Link
• Developing an Investigator Site Budget for Clinical Trials. Journal of Oncology Practice 3(2):94-97, 2007. Link
• Gordon L. Payment for Screen Failures. ACRP Community blog, 2014. Link
• Society for Clinical Research Sites. 2024 Global Site Landscape Survey presentation. SCRS, 2024. Link
• Clinical Trials Arena. Clinical trial sites bear the brunt of failures (GlobalData Healthcare analyst comment). 2024. Link