Patient recruitment and patient enrollment are not the same problem, and mixing them up is why marketing teams and clinical operations blame each other when a trial falls behind. Recruitment is sourcing: getting qualified candidates into the pipeline, owned by marketing, measured in leads and cost-per-lead.
Enrollment is conversion: moving those candidates through screening, informed consent, and randomization, owned by clinical operations, measured in screen-fail rate and randomization rate. Confusing the two wastes budget in the wrong place: pouring more spend into recruitment ads does nothing if the funnel is actually leaking at the screening desk.
OneStudyTeam's widely cited explainer on this exact distinction ranks near the top of search results for the query, but it stops at the definition. It does not show what a normal screen-fail rate looks like, what a normal randomization rate looks like, or whether the oft-repeated claim that 80% of clinical trials miss their enrollment timeline actually holds up. This article draws that funnel-stage line clearly, benchmarks the drop-off at each stage using a real, publicly documented screening case, and answers the 80% timeline claim with an honest look at where that number actually comes from.
What Is Patient Recruitment in a Clinical Trial?
Patient recruitment is the top-of-funnel work of finding and engaging eligible candidates for a specific trial before any clinical screening happens. A marketing director or a dedicated patient-recruitment vendor owns this stage end to end, from ad creative through the moment a candidate submits an interest form. That work typically runs through paid social, search ads, site referral networks, and patient registries.
The output of recruitment is a lead: someone who has expressed interest and provided basic contact and eligibility information, not someone who has been clinically screened. Recruitment success is measured in volume and efficiency metrics: leads generated, cost-per-lead, and lead-to-prescreen conversion rate, not in how many patients ultimately get treated. A recruitment campaign can hit every one of its lead targets and still leave a trial behind schedule if the leads it generates do not survive clinical screening, which is exactly the gap enrollment measures instead.
What Is Patient Enrollment in a Clinical Trial?
Patient enrollment is the clinical-operations-owned process that converts a recruited lead into a randomized study participant through screening, informed consent, and randomization. It starts where recruitment ends: a lead who has already expressed interest gets pre-screened against eligibility criteria online first, then follows up with a phone or in-person clinical screen from site staff or a coordinator.
Candidates who pass clinical screening move to informed consent, an IRB-governed step where the site confirms the patient understands the trial's risks and procedures, and finally to randomization, the point at which a patient is formally assigned to a study arm and counted as enrolled. Enrollment is measured in conversion terms, not volume: screen-fail rate, randomization rate, and time-to-randomization, because the metric that matters is how many recruited leads actually become dosed, randomized patients, not how many people initially raised a hand. This is the stage most sponsors underestimate: it is entirely possible to overperform on recruitment leads and still miss the enrollment target.
Recruitment vs. Enrollment: Where the Funnel Actually Breaks
The funnel between a patient recruitment lead and a randomized, enrolled patient loses candidates at every screening step, and the biggest loss usually happens at clinical screening, not at the top of the funnel where most trial teams are looking. AutoCruitment, a clinical trial recruitment vendor, published one of the more detailed public breakdowns of this funnel from a Phase III uterine fibroid study: nearly 200,000 interested individuals narrowed to roughly 40,000 who completed online screening, then to just over 10,000 who completed a phone screen, and finally to 600 randomized participants. The company reported an overall screen-fail rate of around 90%.
That single case is useful precisely because it is granular, not because it is universal. A screen-fail rate from one Phase III trial in one indication tells a program little about what to expect in a Phase II oncology study or a Phase III diabetes trial, where eligibility criteria, comorbidities, and site density all shift the numbers. What a program needs is not one snapshot but an aggregate view of where its own funnel is losing candidates, stage by stage, measured against a realistic range rather than a single company's best case. The table below lays out the funnel in the order candidates actually move through it, using the AutoCruitment case as the clearest publicly available reference point and naming plainly where broader cross-trial benchmark data is thin rather than filling gaps with invented numbers.
Read stage-over-stage rather than cumulatively: every stage after the initial lead loses at least three-quarters of the candidates ahead of it, and the steepest single-stage loss is actually the last one, where on-site screening, consent, and randomization combined removed 94% of the candidates who had already cleared a live phone screen. Even the milder-looking step from online pre-screen to phone screen still removed three-quarters of otherwise-interested candidates before a coordinator ever spoke with them. That scale of drop-off is concentrated almost entirely in clinical screening, not recruitment intake, and no amount of additional ad spend fixes it. The final row combines on-site screening, consent, and randomization into one step because AutoCruitment's published case does not break those three sub-stages out separately, a reminder that even the most detailed public funnel data available still leaves gaps a sponsor's own program data has to fill.
Screen-Fail and Randomization Rate Benchmarks
Screen-fail and randomization rates vary far more by phase and indication than any single case study can show, which is why the AutoCruitment case should be read as one detailed data point, not a target to hit. Screen-fail rate is the share of clinically screened candidates who do not qualify for randomization, whether because of a failed diagnostic, an excluded comorbidity, or a consent decline; randomization rate is the inverse, the share of screened candidates who do qualify and get assigned to a study arm. In the AutoCruitment case, phone-screened candidates converted to randomization at roughly 6%, with the company reporting an overall screen-fail rate near 90% for that specific Phase III protocol.
Rare-condition and tightly enrolled indications, such as biomarker-gated trials and rare genetic disorders, tend to run narrower eligibility criteria by design, and narrower criteria mechanically produce higher screen-fail rates regardless of how well recruitment sources candidates. Broader indications with looser inclusion criteria generally see more candidates clear clinical screening, though sponsors rarely publish a single, standardized cross-trial figure for this. That is exactly why benchmarking against a program's own historical screen-fail rate is the more reliable practice, since a clean, citable industry average does not exist. Screen-fail rates vary more by therapeutic area than by any single rule of thumb, and that variation is broken out in the screen-fail data by therapeutic area, which carries the indication-level detail this article deliberately does not duplicate.
A high screen-fail rate is not automatically a recruitment failure. It often means the recruitment channel is doing exactly what it was asked to do, generating volume, while the eligibility criteria written into the protocol are the actual constraint. Distinguishing between the two before reallocating budget is the entire point of tracking recruitment and enrollment as separate lines rather than one blended "enrollment" number, which is the practice most trial dashboards still default to.
Why So Many Clinical Trials Miss Their Enrollment Timeline
The claim that around 80% of clinical trials fail to meet their original enrollment timeline traces to a named primary trail: a 2020 systematic review and meta-analysis in the Journal of Medical Internet Research (Brøgger-Mikkelsen et al., "Online Patient Recruitment in Clinical Trials," JMIR 22(11):e22179), which attributes the figure to a 2015 clinical-trial feasibility study (Johnson O., "An evidence-based approach to conducting clinical trial feasibility assessments," Clinical Investigation 5(5):491-499). Enrollment delay itself is well established as the norm rather than the exception, and it traces back to a small number of repeatable causes.
Overly restrictive eligibility criteria are the most common cause: protocols written to minimize confounding variables also shrink the pool of people who can pass clinical screening, which is exactly the mechanism behind the AutoCruitment case's 90% screen-fail rate. Overly narrow criteria carry a second-order effect too: they also shape who ends up representing the trial's results, since the same restrictive filters that cut screen-fail candidates also cut demographic diversity from the randomized population.
Site underperformance is the second cause: a small number of high-enrolling sites typically carry a disproportionate share of randomizations, and a protocol that activates too many low-volume sites instead of concentrating budget on proven ones adds time without adding patients. Recruitment-channel mismatch is the third: a channel that generates leads efficiently for a common condition, such as a broad primary-care population, does not necessarily generate leads that pass screening for a narrow, biomarker-defined trial, and sponsors that don't match channel to protocol complexity lose time re-sourcing mid-trial. None of these three causes require a new statistic to diagnose. A sponsor's own screen-fail log, broken down by reason code, usually shows which one is dominant within a single enrollment cycle.
How to Track Enrollment Without Confusing the Two Metrics
Tracking recruitment and enrollment on the same dashboard, under the same metric, is the single most common reason teams misdiagnose a stalled trial. The fix is to track four numbers separately and never blend them.
Cost-per-recruited-lead measures marketing efficiency: total recruitment spend divided by the number of leads who submit an interest form, regardless of whether they ever get screened. Cost-per-randomized-patient measures the whole funnel's efficiency: total program spend, recruitment and clinical operations combined, divided by the number of patients actually randomized. A program can have an excellent cost-per-lead and a terrible cost-per-randomized-patient at the same time, and that gap is the enrollment problem hiding behind a recruitment metric that looks fine.
Time-to-first-patient-in tracks how quickly a site or program gets its first randomized patient after activation, a recruitment-and-startup metric. Time-to-last-patient-in tracks how long the full enrollment window takes to close, a metric that reflects both how well recruitment kept the funnel full and how efficiently clinical operations converted it. When time-to-last-patient-in blows past projection while time-to-first-patient-in was on schedule, the delay happened in enrollment, not recruitment, and budget conversations should follow the data instead of defaulting to more ad spend. None of these four numbers require special software: a shared spreadsheet that separates recruitment spend from clinical operations spend, updated weekly against actual randomizations, catches the divergence early enough to act on it.
Improving Enrollment Once Recruitment Is Working
Fix enrollment only after confirming recruitment is generating enough qualified leads to fill it. Diagnose which stage is broken first, using cost-per-lead, cost-per-randomized-patient, and the two time-to-patient metrics, before touching either budget. If cost-per-recruited-lead is reasonable and leads are flowing but cost-per-randomized-patient is climbing, the fix lives in clinical operations, not the ad account: faster pre-screening turnaround, tighter coordination between the call center and the site, and eligibility criteria reviewed for restrictions that don't actually protect the study's scientific validity.
If recruitment is the actual bottleneck instead, no amount of clinical-operations tuning will move the timeline. That is a sourcing problem, and it gets solved with better channel selection, sharper eligibility pre-qualification in the ad creative itself, and site-level referral programs, not with a faster clinical screening process. The entire value of separating these two metrics in the first place is naming which half of the funnel is actually broken before reallocating a single dollar.
Conclusion
Recruitment and enrollment fail for different reasons, get fixed by different teams, and get measured with different numbers, which is the whole argument of this article. A trial that is behind schedule is not automatically a recruitment problem: the AutoCruitment funnel data shows screen-fail rates can run as high as 90% in a single documented case, and that kind of drop-off happens entirely downstream of a successful recruitment campaign. The 80% enrollment-timeline statistic, traceable to a 2020 JMIR systematic review, is less useful on its own than the diagnostic habit this article makes the case for: track cost-per-recruited-lead separately from cost-per-randomized-patient, watch time-to-first-patient-in against time-to-last-patient-in, and let that data, not a repeated stat, tell a sponsor which half of the funnel actually needs fixing.
FAQ
What does enrollment mean in a clinical trial?
Enrollment is the clinical-operations stage that converts a recruited lead into a randomized study participant, covering pre-screen, phone or in-person clinical screening, informed consent, and randomization into a study arm. It is the conversion half of the funnel, not the sourcing half; a candidate becomes enrolled only once randomized, not simply because they expressed interest or submitted a lead form.
What is the difference between patient recruitment and patient enrollment?
Recruitment sources candidates: it is the marketing-driven work of generating leads and is measured by cost-per-lead. Enrollment converts those candidates into randomized patients through clinical screening and consent, and is measured by screen-fail rate and randomization rate. The two use different owners, different budgets, and different success metrics, which is why a trial can be strong on one and still stall on the other.
What percentage of clinical trials miss their enrollment timelines?
Around 80% of clinical trials fail to meet their original enrollment timeline, according to a 2020 systematic review and meta-analysis in the Journal of Medical Internet Research, which traces the figure to a 2015 clinical-trial feasibility study (Johnson, Clinical Investigation). AutoCruitment's published Phase III screening case documents a screen-fail rate near 90%, and drop-off at that scale is enough on its own to push a timeline past its original projection without any single funnel stage being mismanaged.
How do you increase patient enrollment in a clinical trial?
Broaden eligibility criteria wherever the protocol allows it without compromising the study's scientific validity, and pre-screen candidates earlier using online screeners so clinical staff spend time only on likely-qualified candidates. Cut avoidable screen failures by training site coordinators on the exact inclusion and exclusion criteria before phone screening starts, and activate patient and physician referral networks alongside paid recruitment channels, since referred candidates typically arrive further along the eligibility path than cold leads.
What does it cost to enroll a patient in a clinical trial?
Per-patient enrollment cost varies too widely by phase, indication, and geography for a single published figure to be reliable, and sponsors rarely disclose their own numbers publicly. The more useful number is internal: cost-per-randomized-patient, or total recruitment and clinical operations spend divided by patients actually randomized. Tracked against a program's own prior trials in the same therapeutic area, it is a more reliable benchmark than any industry-wide average.