Site growth

How sponsors and CROs select clinical trial sites

Sponsor clinical operations team reviewing site maps and performance charts on a large screen
Short answerSponsors and CROs shortlist sites using historical performance data and epidemiology, send feasibility questionnaires to the shortlist, and select based on patient access, past enrollment, startup speed, and investigator experience. Because sites often overestimate enrollment, sponsors discount projections and increasingly rank sites with shared performance data and predictive models.

What are the steps in site selection?

  1. Shortlisting. Sponsors identify sites with relevant research experience and direct access to the target population, often using epidemiological and geographic analysis (PLOS One).
  2. Feasibility questionnaires. Shortlisted sites estimate how many patients they can enroll and describe their capabilities.
  3. Evaluation. Sponsors weigh estimates against historical performance, investigator experience, and operational readiness.
  4. Selection and startup. Contracts, budgets, IRB approval, and the site initiation visit.

Why don't sponsors trust feasibility estimates?

Researchers note that feasibility questionnaires often result in overestimated recruitment, likely because investigators lack full protocol information and time for a thorough assessment (PLOS One). Ken Getz described how historically sites gave figures reflecting their abilities, but sponsors, burned by optimistic projections, began routinely discounting them, and even adjusted estimates are reported accurate less than 10% of the time (Applied Clinical Trials). Sites feel the problem too: in an Advarra survey, over half of site respondents said feasibility questionnaires do not accurately represent their capabilities (Advarra).

How are data and models changing selection?

Sponsors now share site level enrollment data through platforms that record site open dates, first and last patient enrolled, and patients enrolled per study. Researchers have built machine learning models on that data that outperform baseline methods at ranking sites by expected recruitment (PLOS One). A site's past performance increasingly travels with it.

Do familiar sites have an advantage?

Yes. Tufts CSDD found 28% of engaged sites are new relationships, and familiar sites finish initiation 9.9 weeks faster (Tufts CSDD). Site type matters too: physician practices reached first patient in fastest, while academic and government funded sites took longest (Applied Clinical Trials).

What does this mean for your site?

  • Your enrollment history is your resume. Every study counts.
  • Patient access you can document beats patient access you estimate.
  • Community practices have a speed advantage worth highlighting.

A physician referral network gives you documented access: named specialists, engagement data, and past referral volumes. See how to answer a feasibility questionnaire.

Get your study in front of the right local physicians

TrialNotice builds a physician referral pipeline around one active study. We identify relevant local physicians within driving distance of your site, send study aligned direct mail, follow up by email and LinkedIn, track engagement with recipient level QR codes, and route warm responses into your site team's workflow.

Talk to TrialNotice

Sources

  1. "Enhancing site selection strategies in clinical trial recruitment using real world data modeling," PLOS One
  2. Getz K., "Is Investigative Site Feasibility Feasible?" Applied Clinical Trials
  3. Advarra, "Modernizing Site Feasibility and Selection"
  4. Tufts CSDD, first comprehensive clinical site initiation benchmark (2018), via GlobeNewswire
  5. Applied Clinical Trials, "New Benchmarks for Trial Initiation Activities" (Tufts CSDD)