Identify the measurement chain
A wearable report is more useful when a reader can follow the path from a physical signal to a displayed metric. Name the signal, the processing step and the metric separately. Where the device presents a score, explain what the score represents and the context in which it is intended to be read.
Record the configuration
An evidence record should identify the hardware, firmware, algorithm and application versions involved. Describe relevant placement, accessories and settings. This allows a later reviewer to determine whether the report concerns the same configuration as the current product.
Describe the evaluation
Use the following headings to organise a technical review:
- Question: the specific performance question being examined.
- Comparison: the reference or comparator and the reason for using it.
- Participants and setting: who contributed data and under what conditions.
- Data handling: missing observations, exclusions and processing decisions.
- Results: the measures reported, their uncertainty and relevant variation.
- Limits: circumstances, users or functions the work did not examine.
These headings help expose gaps in a record; they do not prescribe a universal test protocol or acceptance threshold.
Explain transfer to intended use
Ask whether the evaluation reflects the conditions described in the product's intended purpose. Document the basis for applying the findings elsewhere and identify where additional work is needed. Keep technical performance and claims about a health outcome as distinct questions.
Maintain the link after change
When a component or algorithm changes, record which evidence remains relevant and the assessment supporting that conclusion. A version number alone does not explain the significance of a change.
For an example of a digital measurement resource that makes inclusion criteria explicit, see the DiMe digital endpoints library. Its criteria belong to that library and are not adopted here as product approval criteria.