Table 1 Example of inputs required for simulation-based power calculations for an IPD meta-analysis of randomised trials with a continuous outcome
When considering the power of a summary (overall) treatment effect with model (1) or (2) used as the data generating model in the first stage: | |
• Number of simulations to conduct (recommend at least 1000) | |
• Number of trials in the IPD meta-analysis | |
• Number of patients in each trial, and proportion treated | |
• Method for estimating the treatment effect in each study separately | |
• Magnitude of control group mean outcome in each trial (‘baseline risk’) | |
• Between-trial distribution and magnitude of treatment effects, e.g. normal with a particular mean (summary) effect and between-trial variance (plus any between-trial correlation of baseline risks and treatment effects, if considered relevant) | |
• Magnitude of residual variance in each trial | |
• For ANCOVA model: distribution and magnitude of baseline continuous values in each trial e.g. normal with particular mean and variance | |
• For ANCOVA model: between-trial distribution and magnitude of the prognostic effect of the baseline continuous values, e.g. normal with particular mean and variance | |
• Approach to use in second stage of the two-stage IPD meta-analysis to pool effect estimates: e.g. fixed effect model or random effects model | |
• Approach to derive confidence intervals and p-values (e.g. standard normal-based method, Hartung-Knapp Sidik-Jonkman, etc) | |
Additionally, when considering the power of a treatment-covariate interaction with models (3) or (4) used as the data generating model in the first stage: | |
• 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 |