Table 1 Summary of the ASPREE trial results for five endpoints using regression-based modelling approaches assuming PH or allowing for TD treatment effects
From: Examining evidence of time-dependent treatment effects: an illustration using regression methods
Endpoint | Estimation | HR (95% CI), p-value | Estimation | \(\Delta\)RMST (95% CI), p-value |
|---|---|---|---|---|
| Â | model | Â | Model | Â |
Primary | Cox PH | 1.01 (0.92,1.11), 0.79 | FPM PH | − 0.006 (− 0.047, 0.035), 0.79 |
|  | Weibull PH | 1.01 (0.92,1.11), 0.79 | FPM TD | − 0.005 (− 0.046, 0.036), 0.81 |
|  | FPM PH | 1.01 (0.92,1.11), 0.79 | GLM p-obs | − 0.007 (− 0.049, 0.035), 0.75 |
MACE | Cox PH | 0.89 (0.77,1.03), 0.12 | FPM PH | 0.021 (− 0.006, 0.049), 0.13 |
|  | Weibull PH | 0.89 (0.77,1.03) 0.12 | FPM TD | 0.021 (− 0.006, 0.048), 0.12 |
|  | FPM PH | 0.89 (0.77,1.03), 0.12 | GLM p-obs | 0.021 (− 0.008, 0.050), 0.16 |
Clinically | Cox PH | 1.38 (1.18,1.62), \(<0.001\) | FPM PH | − 0.050 (− 0.075, − 0.026), \(<0.001\) |
significant | Weibull PH | 1.38 (1.18,1.62), \(<0.001\) | FPM TD | − 0.052 (− 0.077, − 0.027), \(<0.001\) |
bleeding | FPM PH | 1.38 (1.18,1.62), \(<0.001\) | GLM p-obs | − 0.057 (− 0.084, − 0.029), \(<0.001\) |
Cancer | Cox PH | 1.05 (0.95,1.15), 0.32 | FPM PH | − 0.020 (− 0.059, 0.019), 0.32 |
incidence | Weibull PH | 1.05 (0.95,1.15), 0.32 | FPM TD | − 0.018 (− 0.058, 0.021), 0.36 |
|  | FPM PH | 1.05 (0.95,1.15), 0.32 | GLM p-obs | − 0.024 (− 0.068, 0.020), 0.29 |
Cancer | Cox PH | 1.36 (1.13,1.63), 0.001 | FPM PH | − 0.032 (− 0.047, − 0.013), 0.001 |
mortality | Weibull PH | 1.36 (1.13,1.63), 0.001 | FPM TD | − 0.029 (− 0.048, − 0.010), 0.003 |
|  | FPM PH | 1.36 (1.13,1.63), 0.001 | GLM p-obs | − 0.033 (− 0.055, − 0.012), 0.003 |