Effect Size
Effect Size
Running CompARE
For each composite component, the user is prompted for the anticipation of:
- The probability of having the event in the reference group
- The relevant effect (HR) to detect
- The correlation between endpoints
CompARE computes the HR*(t) along time depending on the input parameters.
Input Items
- P01: Probability of event for the relevant endpoint in control group
- P02: Probability of successes for the additional endpoint in control group
- HR1: Hazard Ratio on the Endpoint 1
- HR2: Hazard Ratio on the Endpoint 2
- ρ: Correlation between endpoints (assumed equal in both groups)
Output Items
- Hazard ratios along time: Constant HR1 (dark blue) and HR2 (light blue) over time, as well as the HR* (orange line) over time.
- Measures of non-proportionality: Different measures to asses if the HR* could be considered reasonably constant:
-
- Max. HR*(t). Maximum value of the HR for the composite endpoint along time
- Average HR*(t). Averaged value of the HR for the composite endpoint along time
- Min HR*(t). Minimum value of the HR for the composite endpoint along time
- D. Difference between maximum and minimum value of the HR for the composite endpoint along time. Assuming that both effects are in the same direction, this value ranges between 0 and 1. Values higher than 0.05 could be considered as high, specially if HRs are close to 1.
- R. Sample size ratio between the minimum detectable effect (considering the Max. HR*(t)) and the averaged effect (considering the average HR*(t)). This value ranges between 1 and ∞
. Values higher than 1.25 imply that Naive method to calculate sample size could be wrong.
References
- Statistical considerations when using a composite endpoint for comparing treatment. 2013. Statistics in Medicine. 32(5):719-38. doi: 10.1002/sim.5547.