SAVVY (Survival analysis for AdVerse events with VarYing follow-up times)
SAVVY is a method designed to improve how we assess the safety of medical treatments by better analyzing adverse events in clinical trials. Unlike simple methods that just count how many patients had side effects, SAVVY accounts for how long each patient was followed and the fact that some patients might experience other events (like death) that affect the chance of seeing a side effect. It uses advanced statistical techniques to give a more accurate picture of treatment risks over time.
At a glance
Use when
Assessing safety in clinical trials where follow-up times vary across patients or when competing risks (e.g., death) are present and could influence adverse event observation.
Avoid when
When only aggregated adverse event counts are available without timing information, or when simplicity and rapid reporting are prioritized over methodological accuracy.
Inputs
Adverse event data from clinical trials, including event times, censoring times, competing events, and treatment group assignments.
Outputs
Comparative safety assessments using both standard incidence proportions and advanced time-to-event methods, including cumulative incidence functions and hazard estimates.
How it works
SAVVY is a meta-analytic method that evaluates the impact of using time-to-event analyses—specifically the Aalen-Johansen estimator of the cumulative incidence function—for adverse event assessment in clinical trials. It addresses limitations of the incidence proportion by accounting for varying follow-up times and competing risks. The method involves re-analyzing adverse event data using survival analysis techniques, presented in unified statistical notation, with implementations provided in R and SAS. The project empirically compares conclusions from standard versus advanced methods across multiple trials.
- HTA domains
- Clinical Effectiveness, Safety
- Assumptions
- Adverse events and competing risks follow time-to-event processes; data are sufficiently detailed to support survival analysis (e.g., event timing); trial follow-up times vary across patients.
- Strengths
- Accounts for varying follow-up times across patients,Incorporates competing risks in safety analysis,Provides more accurate and dynamic assessment of adverse event risk over time,Empirically evaluates methodological impact through meta-analysis,Includes practical implementations in R and SAS
- Limitations
- Requires detailed individual patient data, which may not always be available,More complex than simple incidence calculations, requiring statistical expertise,Findings depend on the availability and quality of data from included trials
- Also known as
- SAVVY, Survival analysis for AdVerse events with VarYing follow-up times
Questions this answers
- › How do varying follow-up times affect the assessment of adverse events in clinical trials?
- › Do advanced methods like the Aalen-Johansen estimator lead to different safety conclusions compared to simple incidence proportions?
- › What is the impact of competing risks on adverse event reporting?
- › Can survival analysis methods improve the accuracy of safety evaluations?
- › How can adverse event data be re-analyzed using time-to-event methods in practice?
- › Are current safety assessments potentially biased due to methodological limitations?
References & sources
Similar by meaning
Beta record. Generated from the primary source via AI extraction and independent audit, pending final human review.