twoCoprimary 1.0.0
Initial Release
This is the first release of twoCoprimary, providing comprehensive
tools for sample size and power calculation in clinical trials with two
co-primary endpoints.
Features
- Two Continuous Endpoints
ss2Continuous(), power2Continuous(),
twoCoprimary2Continuous()- Based on Sozu et al. (2011)
- Supports known and unknown variance cases
- Two Binary Endpoints (Asymptotic Methods)
ss2BinaryApprox(), power2BinaryApprox(),
twoCoprimary2BinaryApprox()- Based on Sozu et al. (2010)
- Four test methods: AN, ANc, AS, ASc
- Two Binary Endpoints (Exact Methods)
ss2BinaryExact(), power2BinaryExact(),
twoCoprimary2BinaryExact()- Based on Homma and Yoshida (2025)
- Five exact tests: Chisq, Fisher, Fisher-midP, Z-pool, Boschloo
- Mixed Continuous and Binary Endpoints
ss2MixedContinuousBinary(),
power2MixedContinuousBinary(),
twoCoprimary2MixedContinuousBinary()- Based on Sozu et al. (2012)
- Supports biserial correlation structure
- Mixed Count and Continuous Endpoints
ss2MixedCountContinuous(),
power2MixedCountContinuous(),
twoCoprimary2MixedCountContinuous()- Based on Homma and Yoshida (2024)
- Handles overdispersed count data with negative binomial
distribution
Utility Functions
corrbound2Binary() - Calculate valid correlation bounds
for binary endpointscorrbound2MixedCountContinuous() - Calculate valid
correlation bounds for count and continuous endpointsdesign_table() - Create comprehensive design comparison
tablesplot.twoCoprimary() - Visualize sample size vs
correlation relationships
Documentation
- Six comprehensive vignettes covering all methodologies
- Complete function documentation with examples
- Validation against published results
Testing
- Comprehensive test suite with testthat
- Tests for all major functions
- Validation against published tables and results