Table 4 Overall study design of the simulated trial in comparison with the original trial.

From: Exploring the feasibility of using real-world data from a large clinical data research network to simulate clinical trials of Alzheimer’s disease

Component Target trial (NCT00478205) Simulated trial
Aim Assess the safety and effectiveness of 23 mg SR compared to 10 mg IR Assess whether the simulated trial can generate similar results to the “real” trial in terms of its safety profile
Eligibility 36 eligibility criteria 25 are computable or partially computable
Treatment strategies Randomized allocation of 23 mg:10 mg ratio is 2:1 One-arm simulation of the 10 mg control arm using random sampling and proportional sampling with the same sample size as the original trial.
Two-arm simulations considering different case-to-control ratios. Propensity score matching was performed on the following baseline covariates: sex, race, age, and Charlson Comorbidity Index (CCI).
Sampling strategies N/A Bootstrap with replacement was repeated 1000 times to randomly generate the sample population, and mean value and 95% confidence interval were reported.
Follow-up The outcomes were measured from the first dose to 24 weeks after the first dose.
Outcome SAE and cognition function measures SAEa
Statistical analysis Compare the average number of SAEs per patient, and the SAE rates (i.e., how many patients have SAE).
  1. aSAE: serious adverse event.