Table 4 Typical inputs required for simulation-based power calculations for an IPD meta-analysis of randomised trials with a binary or a time-to-event outcome, using a two-stage IPD framework
When considering the power of a summary (overall) treatment effect: | |
• Number of IPD meta-analysis datasets to generate | |
• Number of trials in the IPD meta-analysis | |
• Number of patients in each trial, and proportion treated | |
• Analysis model and method for estimating the treatment effect in each study separately | |
• Distribution and magnitude of treatment effects across all trials, e.g. normal with a particular mean (summary) effect and between-trial variance | |
• Approach to use in second stage of the two-stage IPD meta-analysis: e.g. fixed effect model (equation 5) or random effects model (equation 9) | |
• Approach to derive confidence intervals and p-values (e.g. conventional method, Hartung-Knapp Sidik-Jonkman, etc) | |
Binary outcomes | |
• Baseline event risk in the control group in each trial (and any correlation between baseline risk and treatment effect across trials, if relevant) | |
Time-to-event outcomes | |
• Maximum length of follow-up in each trial | |
• Distribution of event times in the control group in each trial, and whether these are related or change across trials (corresponding to the shape of the baseline hazard function in each trial and whether they are the same, different but proportional, or completely distinct) | |
• Censoring mechanism and amount of censoring over time | |
• Magnitude of any non-proportional hazards in treatment effect | |
Additionally, when considering the power of a treatment-covariate interaction: | |
• Analysis model and method for estimating the interaction effect in each study separately | |
• Distribution and magnitude of covariate values in each trial; e.g. normal with chosen mean and variance for a continuous covariate, or Bernoulli for a binary covariate with a chosen probability of being a 1. | |
• Between-trial distribution and magnitude of treatment-covariate interaction effect, e.g. normal with a particular (summary) mean effect and between-trial variance | |
• Magnitude of any non-proportional hazards in interaction effect |