Winning More Studies With Feasibility Data Sponsors Believe

Most feasibility questionnaires get filled out in twenty minutes by whoever has a free afternoon. The numbers are estimates, the estimates are optimistic, and the sponsor knows it. Sites that consistently win competitive studies tend to do one thing differently: they answer feasibility questions with real data they already had on hand, and they answer faster than everyone else.

What sponsors are actually trying to learn

A feasibility questionnaire looks like a form. It's really a risk assessment. The sponsor is asking three questions behind every field:

• Can this site find the patients, or are they guessing from a billing code count?
• Will this site start up on time, or will it sit in contracting for five months?
• Will the data come back clean, or will the monitor live here?

Every answer you give should be read as evidence for one of those. "We see approximately 400 type 2 diabetes patients" answers nothing. "Our last three diabetes studies randomized 9, 14, and 6 of 31, 44, and 19 consented, over an average of 7 months" answers all three.

Build the dataset before the questionnaire arrives

You cannot assemble credible feasibility data in the 48 hours a sponsor gives you. Build it once, maintain it quarterly.

At minimum, keep a standing record by therapeutic area of:

1. Consented, screened, randomized, and completed counts for every study in the last three years, with dates.
2. Screen failure reasons, grouped. Knowing that 40% of your failures in a cardiology study came from one lab threshold is a powerful thing to hand a sponsor at protocol review.
3. Time from site selection to first patient in, broken into contracting, IRB, and activation segments. If your bottleneck is central IRB turnaround, say so — it's not your fault and the sponsor knows it.
4. Database source counts for your actual, reachable patient pool: not all patients with the diagnosis, but those seen in the last 18 months with current contact information.
5. Retention rates on long studies, with dropout reasons.

Separate the pool from the projection

The most common credibility killer is confusing database size with enrollment capacity. Be explicit about your funnel math and show the steps.

If you have 600 patients in the diagnosis category, state how many were seen recently, how many survive the obvious I/E screens you can query, what your historical contact-to-consent rate runs in that population, and what your consent-to-randomization rate has been. Land on a number you'd be comfortable defending in a monitoring visit. A site that projects eight and enrolls nine gets called again. A site that projects thirty and enrolls four gets quietly dropped.

Volunteer the problems

When you review a protocol for feasibility, flag the specific requirements that will slow you down — the washout nobody will accept, the imaging you have to send out, the visit schedule that conflicts with working-hour patients. Then propose what you'd need to make it work: a recruitment budget line, a mobile phlebotomy vendor, an extra coordinator FTE.

Sponsors interpret this as competence, not weakness. The sites that say yes to everything are the sites that renegotiate in month four.

Make speed part of the pitch

Turnaround time is itself a feasibility signal. If your questionnaire comes back in two days with clean, sourced numbers and a named PI who has actually read the synopsis, you've already demonstrated the operational discipline the rest of the form claims. Assign one person to own feasibility responses, give them query access to your patient data, and set an internal 72-hour standard.

That combination — fast, specific, honestly caveated — is what moves a site from the long list to the short one.