Data Quality Controls in Biotechnology Need Failure Semantics
Holding
Biotechnology data quality is not a single score; it is the explicit treatment of missing, conflicting, delayed, and out-of-range information.
Issue
A dashboard can show high completeness while concealing values that are stale, duplicated, imputed, or outside the method’s valid range. Quality controls must describe what a failure means operationally.
Analysis
The practical issue is ambiguous status. If a missing measurement, failed control, and delayed upload all appear as blank cells, downstream users cannot distinguish absence of evidence from evidence of failure.
Resolution
The system should define field-level quality rules, preserve raw values, attach status codes, route exceptions to accountable owners, and prevent invalid records from silently entering decision workflows.
Evidence Package
The evidence should include rule definitions, test results, exception histories, remediation records, and reports showing quality status by sample, batch, instrument, and process stage.
Conclusion
A quality control is useful only when its failure changes system behavior or creates an accountable review.