ClinicalTrials.gov had 404,637 registered interventional trials on file as of March 2025, and more than 80% of trials still fail to meet their original enrollment timeline, per a 2020 review of enrollment barriers. Enrollment delay drives more added cost and lost time in a clinical trial than any other single operational variable, with Phase III trial conduct now running $55,716 a day per Tufts CSDD's 2024 analysis and site-level enrollment output declining through the 2010s. This roundup gathers sourced data on enrollment timelines, cost per day by phase and therapeutic area, screen-failure and dropout rates, site-level performance, and how patients actually learn about open trials. Every figure in this piece traces to a named source and publication year, not a restated vendor-blog claim, and any figure that could not be independently verified during this research carries an explicit flag.
Key clinical trial enrollment statistics at a glance
Clinical trials miss their enrollment timeline more than 80% of the time, sites enrolled a median of just six participants each by the late 2010s, and Phase III trial conduct now runs $55,716 a day, the highest of any phase. This table pulls one headline figure from each major theme in this piece, with a named source and year for every row.
Clinical trial enrollment timelines and missed-deadline rates
More than 80% of clinical trials fail to meet their original enrollment timeline, and that gap typically forces a study extension, added sites, or both, according to a 2020 review of enrollment barriers. The same review found that more than 40% of trials amend their protocol before the first subject visit, a change that adds roughly four months to the timeline. Average enrollment efficiency for Phase III and IV trials runs below 40%, meaning most late-stage trials fall short of their planned enrollment pace. That 2020 review also found that 55% of terminated trials cited low accrual as the single highest reason for stopping the study, which shows enrollment shortfalls threaten a trial's survival, not just its schedule.
Enrollment duration has also grown over the past decade. A peer-reviewed analysis of 5,672 industry-sponsored Phase III trials found the median recruitment period rose from 13 months in 2008-2011 to 18 months in 2016-2019, a 38% increase.
PeriodMedian Phase III enrollment duration2008-201113 months (IQR 7-23)2016-201918 months (IQR 11-28)
More recent data suggests the trend has leveled off. IQVIA's Global Trends in R&D 2025 report found that enrollment duration, described as the largest opportunity to improve trial cycle times, rose between 2021 and 2023 but stabilized in 2024. The same report put the inter-trial interval, the time between successive trials in a development program, at roughly 17 months in 2024, down from a 32-month peak during the 2022 pandemic period.
Cost per patient and per randomized patient by phase and therapeutic area
No primary source publishes an audited cost-per-randomized-patient figure broken out cleanly by phase and therapeutic area. The closest verified benchmark is Tufts CSDD's direct daily trial cost data, drawn from 447 protocols with budget data from 2016 to 2021 inflated to 2023 dollars. This section reports cost per day of trial conduct, the nearest defensible proxy, rather than an unverified per-patient number.
Tufts CSDD's 2024 white paper puts the mean direct cost to run a Phase II or Phase III trial at approximately $40,000 a day. Phase III carries the highest daily cost of any phase. Phase I, the least expensive of the four phases Tufts measured, still runs $7,829 a day, which means a schedule slip adds real cost even in the cheapest phase.PhaseMean direct cost per dayPhase I$7,829Phase II$23,737Phase III$55,716Phase IV$14,091Phase II and III combined averageAbout $40,000
Cost per day also varies sharply by therapeutic area. Immunology and respiratory trials run the most expensive per day of the areas Tufts measured; cardiovascular and oncology trials run the least expensive of the group.
Therapeutic areaMean direct cost per dayImmunology$51,340Respiratory$50,351Oncology$33,365Cardiovascular$30,657
The same Tufts study estimates that a single day of trial delay costs sponsors roughly $800,000 in lost or delayed prescription sales, a figure the researchers say replaces an outdated $4 million to $5 million estimate that had circulated in the industry for more than 25 years.
Market-size estimates for clinical trial patient recruitment services vary widely by vendor methodology, so no single number should be read as a settled consensus. One industry analysis estimates the global market at $1.06 billion in 2025, growing to $2.29 billion by 2035, an 8.01% compound annual growth rate, and Mordor Intelligence's 2025 estimate independently lands on the same $1.06 billion. At the other end of the spread, Roots Analysis values the same market at $11.8 billion in 2025, projecting $26.2 billion by 2035, an elevenfold gap that comes down to which services each firm counts.
Screen-failure and dropout rates in clinical trials
Screen-failure rates range from roughly 1 in 5 candidates in some solid-tumor trials to more than 2 in 5 in biomarker-gated Alzheimer's studies, and no single average figure applies across all therapeutic areas. A review of 50 Phase II and III genitourinary cancer trials conducted between 1999 and 2016 found screen-failure rates of approximately 20% to 30% overall. Rates varied by cancer type: prostate cancer trials screen-failed 26% of candidates (range 12% to 45%), kidney cancer trials 25% (range 21% to 29%), and bladder cancer trials 19% (range 4% to 28%).
More recent data from early Alzheimer's disease trials shows even wider variation. A 2025 analysis of biomarker-gated enrollment found that a ptau-181 biomarker cutoff screen-failed 36.2% of prescreened candidates in the ADNI cohort and 42.1% in the NACC cohort. Biomarker-gated neurology trials screen out a substantially higher share of candidates than genitourinary cancer trials, where screen-failure rates run roughly 20% to 30%.
Screen-failure rate depends heavily on therapeutic area and protocol design. Claims of one flat industry-wide dropout or retention rate are not supported by a verifiable primary source and are avoided in this piece.
Site-level enrollment performance: how many sites actually deliver
One in 10 investigator sites that are ready to recruit fails to enroll a single patient, and 48% of all sites selected for a trial either enroll nobody or under-enroll, per a Tufts CSDD analysis of nearly 16,000 investigative sites. The study, reported by Tufts CSDD's Kenneth Getz in Applied Clinical Trials in 2012, covered 151 Phase II and III global trials conducted between 2008 and 2010, with performance data supplied by 10 pharmaceutical companies and two contract research organizations. Non-enrollment varied by region: 13% of North American sites failed to enroll a single patient, against 7% in Western Europe and 20% in Latin America.
A higher figure, that about 27% of investigator sites cannot enroll a single qualified patient, circulates widely in industry content attributed to Tufts CSDD. No Tufts publication supporting that number could be located during fact-checking for this piece, and Tufts' own published site-level figure is the lower 11%, so treat the 27% claim with skepticism.
Site-level performance has also declined over time in the same 5,672-trial Phase III dataset spanning 2008 to 2019. Median enrolled participants per site fell from 7 in 2008-2011 to 6 in 2016-2019, and the median recruitment rate fell from 0.6 participants per site per month to 0.4 over the same period, a statistically significant decline.
Nearly half of selected sites enrolling nobody or missing their target is a site-level performance problem, separate from trial-level enrollment timelines, and it is the specific gap addressed by the agencies that close enrollment gaps at underperforming sites.
How patients actually find clinical trials: recruitment channel data
Most people who actually hear about a clinical trial hear about it online, 67% versus just 7% through their own doctor, yet 58% say they would prefer to learn about a trial from a doctor in the first place. CISCRP's 2023 global Perceptions and Insights survey asked respondents where they had seen or heard about a clinical research study. Among the 3,810 respondents who had never participated in a trial but had heard about one in the past six months, 67% heard about it online and just 7% heard about it from their own doctor. Among the 7,459 respondents who had never participated in a trial, 58% said they would prefer to learn about a trial from a doctor, versus 46% who preferred an online source.
Within the online channel alone, based on the 2,538 respondents who heard about a trial online, social media accounted for 61% of that awareness, online ads for 29%, patient advocacy group websites for 20%, and government trial registries for 16%.
That gap, between doctors as the preferred source and social platforms as the actual one, is central to how healthcare marketers reach patients across today's most-used channels.
Recruitment vs. enrollment: what the terms actually measure
Recruitment and enrollment mark different points on the same funnel, and the federal rulebook behind ClinicalTrials.gov draws the line at informed consent. Under 42 CFR 11.10(a), the regulation that defines terms for ClinicalTrials.gov reporting, a person counts as enrolled once they or their legally authorized representative agree to participate after completing the informed consent process. Candidates screened for eligibility who never join the study are not counted as enrolled. Recruitment covers everything before that point: identifying candidates, outreach, and eligibility screening.
ClinicalTrials.gov's glossary defines a study's enrollment simply as the number of participants in it, with estimated enrollment as the target researchers need. The two terms get used interchangeably in casual industry conversation, but they describe different stages of the same funnel. That is why the definition behind a figure matters as much as the figure itself when comparing a sponsor's internal recruitment-rate metric against a published enrollment benchmark.
What the enrollment data adds up to
Enrollment remains the single biggest lever on clinical trial cost and timeline, and sponsors are still missing enrollment targets on a clear majority of trials. More than four in five trials miss their original timeline, site-level performance declined across the 2010s, and Phase III trial conduct now costs $55,716 a day. Patients say they want to hear about trials from a doctor, who, in practice, is rarely the source they actually hear from. Every figure in this piece traces to a named, dated primary source, and any figure that could not be independently verified during this research is explicitly flagged as such.