Use of Historical Individual Patient Data in Analysis of Clinical Trials
share
Historical data from previous clinical trials, observational studies and health records may be utilized in analysis of clinical trials data to strengthen inference. Under the Bayesian framework incorporation of information obtained from any source other than the current data is facilitated through construction of an informative prior. The existing methodology for defining an informative prior based on historical data relies on measuring similarity to the current data at the study level that can result in discarding useful individual patient data (IPD). This paper proposes a family of priors that utilize IPD to empower statistical inference. IPD-based priors can be obtained as a weighted likelihood of the historical data where each individual's weight is a function of their similarity to the current study population. We demonstrate that the proposed prior construction approach can considerably improve estimation accuracy and precision in compare with existing methods.
READ FULL TEXT
