Protocol Complexity Is Eating Your Eligible Population

Protocols have grown steadily more complex — more endpoints, more procedures, more eligibility criteria. Each criterion looks reasonable in isolation; multiplied together, they can shrink an eligible population to a sliver of the initial estimate.

The multiplication problem

Suppose 1,000 database patients have the target condition. Age range keeps 70%. The A1c window keeps 60% of those. Excluding a common comorbidity keeps 65%. Excluding two frequent medications keeps 55%. Requiring a washout keeps 80% of the remainder. You're at roughly 120 candidates before considering interest, logistics, or competition — an 88% reduction from the number everyone remembered from feasibility.

What sites can do

Model criteria impact before committing. Run each major criterion against your actual database during feasibility and share the math. Sponsors increasingly listen — protocol amendments driven by enrollment failure cost them far more than a feasibility conversation.

Flag the killer criterion early. In most under-enrolling studies, one or two criteria cause the majority of exclusions. Documenting that pattern in the first month — "80% of our screen-fails hit the creatinine bound" — gives the sponsor evidence to consider an amendment while it can still help.

Screen in the right order. Structure phone screens so the highest-elimination criteria come first. Respecting a candidate's time is also respecting your coordinator's.

Keep the near-misses. A candidate excluded by one criterion is a strong candidate for the next protocol. Tag the reason for exclusion in your database; "failed X only" lists are gold when a similar study opens with looser bounds.

Complexity is mostly outside a site's control. What's controllable is refusing to be surprised by it — measuring the funnel it creates and reporting that reality upstream, early, with numbers.