A feasibility count is an estimate of how many patients or specimens at a site appear to meet a study’s stated criteria. It answers one question early: is the cohort likely to exist? A researcher uses feasibility counts to decide whether to pursue a study design, adjust the criteria, or look elsewhere, before committing time, budget, or a protocol to it.
What does a feasibility count tell a researcher?
A feasibility count tells a researcher where matching material appears to exist and roughly how much. For a request such as “Stage III colorectal cancer: FFPE blocks with matched whole-slide images and clinical data,” one site might return 118 matching cases, another 42, and another none (illustrative counts).
Read that way, a set of feasibility counts answers three practical questions:
- Does the cohort exist at all? A count of zero across every site is an early signal to revisit the criteria.
- Where is it concentrated? Counts by site show which institutions to approach first.
- Is the target size realistic? If a study needs 300 patients and the combined counts are near that number, attrition during review may leave the study short.
An FFPE block is tissue preserved in formalin and embedded in paraffin wax. A whole-slide image is a high-resolution digital scan of an entire microscope slide. A request that needs both, linked to clinical data for the same patient, is narrower than a request for any one of them, and feasibility counts make that difference visible.
What does a feasibility count not tell you?
A feasibility count is not a reservation, a quality guarantee, or a release decision. Researchers who treat a count as a commitment often plan around material they will never receive.
A feasibility count does not tell you:
- Whether every specimen passes quality review. Tissue quantity, preparation, and image quality are assessed later against the study’s acceptance criteria.
- Whether the material can be released for your use. Consent terms, institutional review board (IRB) review, and the institution’s release conditions are separate decisions. An IRB is the committee that reviews research involving human subjects.
- Whether every linked field is complete. A case can match on diagnosis and stage but lack a treatment history or outcome field the study needs.
- Exact numbers. Counts depend on how each site records the criteria. Small differences in coding can move a count up or down.
The National Cancer Institute’s best practices for biospecimen resources treat specimen quality and associated clinical information as separate from availability. A feasibility count speaks to availability.
What is a count actually counting?
A feasibility count can refer to patients, cases, specimens, blocks, slides, or image files, and those answer different questions. Ten slides cut from one block are not the same as material from ten independent patients.
Before relying on a feasibility count, confirm the unit. A study that needs 100 independent donors should read counts in patients. A study that needs serial sections for several assays may care about blocks and the slides each can yield. The biospecimen request guide covers how to state that unit in the request itself.
How do you read feasibility counts across several sites?
Feasibility counts from several sites should be read as a distribution, not a total. Adding them gives a planning estimate, not a forecast: review can remove cases, and a site that does not record an attribute may hold more than its count shows.
Three habits make multi-site feasibility counts more useful:
- Compare like with like. Confirm that each site counted the same unit against the same criteria.
- Expect attrition. Quality review, missing fields, and release decisions remove cases between feasibility and delivery. Plan for a margin.
- Look at the zeros. A site that returns none may not record the attribute at all. That is different from a site that records it and has no matches.
How do you get a feasibility count before committing to a study?
A researcher gets a feasibility count by describing the cohort in enough detail for a site to check its records. The description should name:
- The indication: the diagnosis and, where it matters, stage or subtype.
- The material: FFPE blocks, unstained slides, whole-slide images, clinical data, or a combination.
- The linked data: which clinical fields must be connected to the same patient, such as treatment history or outcomes.
- The unit and target size: patients or specimens, and how many.
On Biobase, a researcher describes the cohort in plain English or as precise criteria. Agents check every partner site and return counts and attributes from participating institutions, without any data changing hands. Partner data stays at each institution; the network carries the query and the governed result.
If the counts support the study, the next step is a formal request, where quality, documentation, IRB review, chain of custody, and pricing are worked through. Chain of custody is the documented record of who held a specimen or dataset, when, and under what conditions. The five-step overview shows where feasibility sits in that sequence.
When should a researcher change the criteria?
A researcher should change the criteria when feasibility counts show the cohort is too small or too scattered to deliver. Change one criterion at a time and request counts again, so it is clear which requirement is limiting the cohort.
Common adjustments include widening a stage range, accepting an alternative preparation format, or separating required clinical fields from preferred ones. Each change alters the research question slightly. Record why it was made, so the final cohort definition still matches what the study can claim.
Do not include patient identifiers in a feasibility request. Start with the research requirements on the Life Sciences page or through a research request.