Validation & Transparency
Validated. Transparent. Trustworthy.
Every Zetyra calculator passes 51 automated validation tests against industry gold standards. Our Group Sequential Design module matches gsDesign R package within 0.005 z-score units—10× more accurate than typical commercial software tolerances.
At a Glance
Open Source
Unlike proprietary alternatives, our validation is public.
Our complete validation suite runs continuously on GitHub Actions. Anyone can verify our accuracy, examine our methodology, and reproduce our results. No black boxes—just glass boxes.
View Live Validation SuiteResults Summary
Table 1
Validation Results by Calculator
| Calculator | Tests | Reference | Max Deviation | Status |
|---|---|---|---|---|
| Group Sequential Design | 30 | gsDesign R package | 0.0046 z-score | Pass |
| CUPED | 12 | Analytical formulas | < 0.001 | Pass |
| Bayesian | 9 | Conjugate priors | < 0.01 | Pass |
Real-World Validation
HPTN 083
Phase 3 HIV Prevention Trial
HeartMate II
LVAD Clinical Trial
Boundary Accuracy
23 boundary comparisons across 7 design configurations, all within 0.05 z-score tolerance.
Table 2
GSD Boundary Accuracy by Design
| Design | Looks | Max Deviation | Status |
|---|---|---|---|
| O'Brien-Fleming | 2 | 0.0000 | Pass |
| O'Brien-Fleming | 3 | 0.0001 | Pass |
| O'Brien-Fleming | 4 | 0.0017 | Pass |
| O'Brien-Fleming | 5 | 0.0046 | Pass |
| Pocock | 2 | 0.0000 | Pass |
| Pocock | 3 | 0.0002 | Pass |
| Pocock | 4 | 0.0008 | Pass |
Methodology
gsDesignGroup Sequential Design
Validated against the gold-standard gsDesign R package. O'Brien-Fleming and Pocock spending functions computed to match FDA submission standards.
VRF = 1 - r²CUPED
Variance reduction validated against analytical formulas. Sample size reduction proportional to baseline-outcome correlation squared.
Beta(a+x, b+n-x)Bayesian
Conjugate posteriors validated analytically. Beta-Binomial and Normal-Normal models with exact predictive probability calculations.
References
- 1. GSD: Jennison & Turnbull (2000) Group Sequential Methods with Applications to Clinical Trials
- 2. CUPED: Deng et al. (2013) Improving the Sensitivity of Online Controlled Experiments (WSDM)
- 3. Bayesian: Gelman et al. (2013) Bayesian Data Analysis
- 4. gsDesign: Anderson (2022) gsDesign R package
The only clinical trial design platform with public, continuously validated accuracy.
Try our validated calculators today.