Recruiting for Studies Where Almost Nobody Qualifies

Narrow-eligibility studies break the habits that work everywhere else. Broad protocols reward volume: advertise widely, pre-screen fast, book anyone plausible. A protocol that needs a specific genotype, a specific washout history, or a lab value inside a tight window punishes volume. You burn staff hours on calls that were never going to convert, and your screen-fail ratio starts hurting your reputation with the sponsor.

Here is how experienced sites handle them differently.

Identify the one or two criteria that actually do the filtering

Every restrictive protocol has a small number of criteria responsible for most exclusions. It is rarely the ones the sponsor highlights. Read the I/E list and ask: which of these can I not influence, not work around, and not determine cheaply?

Common culprits:

• A required prior therapy or documented failure on a specific drug class
• A lab or imaging value inside a narrow band at a specific timepoint
• Washout periods that conflict with standard-of-care prescribing in your area
• Age plus comorbidity combinations that are statistically uncommon together

Once you have named the binding constraints, your whole recruitment plan should be designed to test for them first, as early and as cheaply as possible.

Flip the pre-screen order

Most pre-screening scripts move chronologically: demographics, diagnosis, medications, then the hard stuff. For narrow protocols, invert it. Ask the binding question in the first ninety seconds. If a candidate needs documented failure on two prior biologics, that is question one, not question fourteen.

This feels abrupt. It is also kinder. Nobody wants to spend twenty minutes on the phone to be excluded on something that could have been established immediately.

Stop advertising and start searching

For tight protocols, outbound beats inbound almost every time. Paid advertising generates volume, and volume is exactly what you do not need. Sites commonly find that a narrow study fills faster from three sources:

1. Structured EMR or registry queries built on the binding criteria, not on diagnosis alone. Query the lab value and the medication history together.
2. Referring physician conversations where you describe the patient in one sentence a clinician can recognize. "Looking for patients who flared after switching off their first biologic" works. A protocol summary does not.
3. Your own prior screen-fail and completed-study records, filtered on the specific criterion rather than the therapeutic area.

If you do advertise, write copy that pre-excludes. Naming the restrictive requirement in the ad costs you clicks and saves you hours.

Renegotiate the enrollment math up front

When feasibility arrives, be explicit with the sponsor about your expected funnel. If you believe your population yields a handful of qualified candidates per month, say so and commit to a number you can hit. Sites that over-promise on narrow studies end up as the site that underperformed; sites that forecast honestly and deliver become the site the sponsor calls for the next hard protocol.

Also ask whether any criterion is negotiable. Tight windows sometimes exist because of a draft assumption nobody revisited. Protocol amendments do happen, and site feedback during startup is when they are cheapest.

Track pre-screen efficiency, not just screen fails

For these studies, the metric that matters is how many pre-screens it takes to produce one consented participant, and which question killed each one. Keep a simple tally of exclusion reasons by criterion. Within a few weeks you will know whether your sourcing channel is wrong, your script order is wrong, or the population simply is not there — and each of those has a different fix.