If contact classification is wrong, endpoint selection is wrong. That means the entire biological evaluation strategy can be mis-scoped from the start, even when the later writing looks polished.
ISO 10993 planning starts with two questions: what does the device contact, and for how long? Those answers drive the endpoint matrix, the evidence strategy, and whether testing or waivers make sense.
Many biological evaluations go off track before any testing, literature review, or waiver logic begins. The root problem is often simple: the device was classified too broadly, too narrowly, or as a single contact profile when the design actually contains multiple contact situations.
Why Contact Classification Comes First
Contact classification is the first real decision in a BEP. It determines which biological endpoints need to be addressed and how demanding the evaluation will be. If that first step is off, the rest of the plan becomes harder to defend because every later decision depends on it.
Contact Nature Categories
ISO 10993 distinguishes the device by what it physically contacts. In broad terms, that includes surface-contact devices, external communicating devices, and implants. Within those groups, the biological implications differ substantially depending on whether the device touches intact skin, mucosa, blood path indirect tissue, circulating blood, bone, or implant sites.
- Surface contact devices: intact skin, mucosal membranes, or breached/compromised surfaces.
- External communicating devices: blood path indirect, tissue or bone communicating, or circulating blood contact.
- Implants: long-term contact with tissue, bone, or blood.
Contact Duration Categories
The second dimension is duration. Even the same contact nature can drive a very different endpoint set depending on whether exposure is limited, prolonged, or permanent.
- Limited: cumulative exposure of 24 hours or less.
- Prolonged: more than 24 hours and up to 30 days.
- Permanent: more than 30 days.
The Classification Drives Endpoint Selection
Once contact nature and duration are established, the endpoint table becomes meaningful. A limited-contact skin device may need a relatively narrow endpoint set. A permanent implant usually requires a much broader evaluation that can extend into chemistry, toxicology, implantation, degradation, and chronic-risk considerations.
Where Teams Commonly Get It Wrong
- Using the product name as the classification: for example, calling something a catheter without unpacking the actual tissue and blood-contact profile.
- Ignoring cumulative duration: repeated short uses can still push the device into a longer-duration category.
- Treating the whole device as one contact profile: different components may contact different tissues and require different endpoint logic.
- Skipping change impact: a design, indication, or material change can alter the classification and therefore the whole endpoint strategy.
Multi-Component Devices Need More Care
Combination or multi-material devices often need separate classification logic for distinct components. A device with one part in circulating blood and another part implanted in tissue should not automatically inherit a single simplified endpoint set. That kind of shortcut is a common source of reviewer questions.
When in doubt, classify the patient-contacting components explicitly and show the reasoning. Reviewers are much more comfortable with a visible, defensible classification path than with a short conclusion that skips the logic.
What a Strong BEP Does with Contact Classification
A strong BEP does not just state the classification result. It explains how the contact profile was determined, how cumulative duration was considered, whether different components were assessed separately, and how that led into the final endpoint strategy.
Need help confirming the right contact classification and endpoint scope before drafting the BEP?
Discuss Your Project Back to Insights