TABLE 1.
Parameter | Estimate | Simulation (n = 1000) | Bias | |
|
||||
Median | 95% confidence interval | |||
Emax, % | −3.92 | −3.92 | [−3.92, −3.92] | 0 |
ET50, weeks | 3.6 | 3.6 | [0.297, 9.89] | 0 |
ωEmax | 0.106 | 0.098 | [0.003, 0.339] | −0.075 |
ωET50 | 0.867 | 0.376 | [0.003, 2.117] | −0.566 |
ε | 0.01 | 0.01 | [0.01, 0.01] | 0 |
95% confidential interval is shown with 2.5th, 97.5th percentile; Emax is the maximal effect; ET50 is the treatment duration to reach half of Emax; ωEmax is the inter-study variability of Emax; ωET50 is the inter-study variability of ET50; and ε is the residual error; bias = (median-estimate)/estimate.