Abstract:
|
Conducting observational medical product effectiveness and safety studies are important for comparing available medical products and for detecting rare adverse events not identified pre-licensure. New systems have been built using electronic healthcare data that keeps the individual patient data within the health plan and establishes a distributed data network to share deidentified data to answer important safety questions about new medical products. Such networks include PCORnet and FDA Sentinel Initiative. We will present survival methods tailored to these networks to estimate hazard ratios using different Cox Proportional Hazard models that account for confounding using approaches such as regression, stratification and exposure matching. To assess the performance of such methods we will conduct a simulation study comparing methods in terms of bias, power, and coverage. We will focus our simulation comparison on the rare event setting with strong across health plan/site confounding.
|