Test record keeping — documenting what was tested, results, and tester is standard quality management practice.
Class B applicability added by A1:2015. Explicit content requirements for repeatability: test case specifications, equipment records, and test environment documentation. [Class B, C]
Per-test-case pass/fail with anomaly list, tester identification, and environment documentation sufficient to reproduce the test — all three elements must be present in every integration test record.
Maps to
IEC 62304: §5.6.7 Integration test record contents
ISO 13485: §7.3.6 Design and development verification
Pre-QMSR Part 820 (legacy QSR): §820.30(f) Design verification.
Requirement text
The manufacturer shall: (a) document the test result (pass/fail and a list of anomalies); (b) retain sufficient records to permit the test to be repeated; and (c) identify the tester. [Class B, C] NOTE (non-normative): requirement (b) could be implemented by retaining, for example, test case specifications showing required actions and expected results, records of equipment, and records of the test environment (including software tools) used for testing.
Why this clause exists
Test record content requirements define the minimum evidentiary standard for integration test records. Without clause 5.6.7, manufacturers could produce test reports that confirm testing was performed without providing the information necessary to evaluate whether the testing was valid, to reproduce a test to investigate a failure, or to confirm that test results are attributable to a known tester. The repeatability requirement is particularly important for regulatory submissions and problem investigations: if a test failure occurs in the field and integration testing records cannot support reproducing the test, the manufacturer cannot determine whether the field failure would have been detectable at the integration test stage.
What changed
Clause 5.6.7 was present in IEC 62304:2006 and maintained in A1:2015 with the addition of Class B applicability. The three required record elements (pass/fail, anomaly list, tester identification) are unchanged. The repeatability examples in the note are also carried forward from the 2006 version.
Common gaps (what we see in audits)
- Test records capture aggregate pass/fail but not per-test-case results — Integration test reports state 'all tests passed' or provide a summary count without documenting individual test case results. Clause 5.6.7(a) requires documentation of the test result per test, and (b) requires sufficient records to repeat the test — a summary cannot support either requirement.
- Tester not identified in test records — Integration test reports do not identify the individual tester by name and role. Clause 5.6.7(c) requires tester identification — anonymous test records cannot be evaluated as controlled quality evidence and cannot support investigation of test validity.
- Test environment not documented in test record — Integration test reports record test results but do not document the hardware configuration, operating system, and software tool versions used. Without this information, the test cannot be repeated in the same environment, and the environmental validity of the test results cannot be confirmed.