An IVD's analytical and clinical performance must be good enough for the use the manufacturer claims, established by sound science and engineering, and the design must deal, as relevant, with trueness and precision, sensitivity and specificity, stability, known interferents and the uncertainty of measurement. Where results depend on calibrators or control materials, the quality management system must secure the metrological traceability of the values given to them. As far as reasonably practicable the user must be able to check at the time of use that the device will perform as intended. A self-test IVD must perform suitably given lay users' skills, means and the variation expected in technique and environment, its information and instructions must be easy to understand and apply, and its design must reduce the risk of error in use, sample handling and reading results.
The graph holds this control, the 0 it maps to, and the evidence behind each claim, over MCP and REST.