SAVVY: estimation of AE probabilities


Regina Stegherr and Kaspar Rufibach


The assessment of safety is an important aspect of the evaluation of new therapies in clinical trials, with analyses of adverse events being an essential part of this. Standard methods for the analysis of adverse events such as the incidence proportion, i.e. the number of patients with a specific adverse event out of all patients in the treatment groups, do not account for both varying follow-up times and competing risks. Alternative approaches such as the Aalen-Johansen estimator of the cumulative incidence function have been suggested. Theoretical arguments and numerical evaluations support the application of these more advanced methodology, but as yet there is to our knowledge only insufficient empirical evidence whether these methods would lead to different conclusions in safety evaluations. The Survival analysis for AdVerse events with VarYing follow-up times (SAVVY) project strives to close this gap in evidence by conducting a meta-analytical study to assess the impact of the methodology on the conclusion of the safety assessment empirically. Three papers summarize the rationale and results of the project (for all SAVVY papers see publications):

  • Stegherr, Beyersmann, et al. (2021): Statistical analysis plan, presenting the rationale and statistical concept of the empirical study conducted as part of the SAVVY project. The statistical methods are presented in unified notation and examples of their implementation in R and SAS are provided. arxiv | publication
  • Stegherr, Schmoor, et al. (2021): 1-sample case, compares estimators of AE risk in one treatment arm. arxiv | publication
  • Rufibach et al. (2023): 2-sample case, compares relative risk estimators comparing two treatment arms based on (1) AE probabilities and (2) hazards. arxiv | publication

SAS code

The original SAS macros that were used by each sponsor to generate the summary data for the meta-analysis is available as supplement to Stegherr, Beyersmann, et al. (2021), see here, or direct download link for zip-file.

R code

All code for SAVVY has been implemented in the savvyr R package (also on github). See the vignette for an introduction on its use.


Rufibach, K., R. Stegherr, C. Schmoor, V. Jehl, A. Allignol, A. Boeckenhoff, C. Dunger-Baldauf, et al. 2023. “Comparison of Adverse Event Risks in Randomized Controlled Trials with Varying Follow-up Times and Competing Events: Results from an Empirical Study.” Statistics in Biopharmaceutical Research 15 (4): 767–80.
Stegherr, R., J. Beyersmann, V. Jehl, K. Rufibach, F. Leverkus, C. Schmoor, and T. Friede. 2021. “Survival Analysis for AdVerse Events with VarYing Follow-up Times (SAVVY): Rationale and Statistical Concept of a Meta-Analytic Study.” Biometrical Journal 63 (March): 650–70.
Stegherr, R., C. Schmoor, J. Beyersmann, K. Rufibach, V. Jehl, A. Brückner, L. Eisele, et al. 2021. Survival analysis for AdVerse events with VarYing follow-up times (SAVVY)-estimation of adverse event risks.” Trials 22 (1): 420.