Skip to main content

Table 4 Power of tests under different scenarios: simulation results

From: An approach to trial design and analysis in the era of non-proportional hazards of the treatment effect

Test

Treatment effect

 

Constant (PH)a

Increasing (HR1)b

Decreasing (HR2)c

Logrank

0.874

0.719 (0.006)

0.820 (0.005)

Joint

0.800

0.854 (0.005)

0.855 (0.005)

  1. aUnder proportional hazards with HR = 0.75.
  2. bSimulation mean HR = 0.790.
  3. cSimulation mean HR = 0.767.
  4. Values under PH are exact theoretical figures. Other values are the proportions of 5,000 simulation replicates in which the joint null hypothesis was rejected. Values in parentheses are Monte Carlo standard errors.
Back to article page

Trials

ISSN: 1745-6215