IVD Performance Studies
For IVDs, clinical evidence takes the form of performance evaluation under RDC 36/2015.
Analytical performance: Accuracy/trueness, precision (repeatability & reproducibility), linearity and measuring range, LOD/LOQ, interference, cross-reactivity, stability.
Clinical performance: Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV).
For Class III and IV IVDs, a comprehensive Performance Evaluation Report (PER) documenting both analytical and clinical performance validation is required as part of the Registro technical dossier submission. Point-of-care and self-test IVDs require usability testing with representative lay users (minimum 30–50 subjects per AAMI/IEC standards) reflecting the intended user population (e.g., patients, healthcare workers without laboratory training), and performance data collected under real-world conditions (home environment, clinic setting, field deployment). Usability reports must document error rates, user instruction comprehension, and identification of critical use failures.
Verify all information against official ANVISA sources before making regulatory decisions.
For Class III/IV IVD performance studies, minimum sample sizes should follow CLSI or ISO 13000 series standards: typically ≥30 samples for analytical performance studies and ≥100–200 clinical samples (split between disease and non-disease populations) for clinical performance validation. Statistical acceptance criteria must include 95% confidence intervals, sensitivity/specificity >95% for most diagnostic applications, and predefined cut-off values with documented rationale.