Considered a first in FDA guidance, the document provides clarity on the "principal factors FDA considers when making benefit – risk determinations" in hopes of improving "predictability, consistency and transparency of the pre‐market review process." manufacturers of both diagnostic and therapeutic devices can apply this guidance to its device development and testing plans.
FDA’s statutory mandate, per the FD&C Act, has not changed; FDA approves PMA and 510k de novo applications based upon "reasonable assurance of safety and effectiveness" as contrasted with the probable benefit to health versus risk associated with the use of the device.
Valid scientific evidence remains the foundation from which sponsors provide supportive data upon which their claims of safety and effectiveness for the indications and conditions for use are based. FDA has now formalized the type and scope of factors leading to these determinations. In the determination of approval, several key, common factors are used to assess both safety and efficacy:
1. Extent of benefit and risk, inclusive of:
| Benefit | Risk |
| Type | Type |
| Magnitude of benefit | Rates of risk, Relatedness of risk to device or procedure, Seriousness of risk experienced |
| Probability of experiencing benefit | Probability of experiencing harm |
| Duration of benefit | Duration of harm |
| Risk from false positive or false negative (diagnostics) |
2. Uncertainty regarding risk‐benefit in a representative population
3. Patient Tolerance for Risk
4. Risk Mitigation
5. Postmarket Data
6. Novel technology addressing unmet clinical need
FDA has taken its guidance one step further by providing both hypothetical cases as well as accompanying worksheets complete with representative determination information from prior approvals. The hypothetical as well as sample cases offer a practical range of experience to understand the factors FDA Reviewers consider in their decision to approve/disapprove PMA approvals and de novo classifications. These factors and cases will be reviewed in further detail in our next installment.