All Times EDT
Keywords: Real world data, real world evidence, propensity score, data matching, individual patient data, inverse probability weight
Real World Data (RWD) have been emerging as important data sources for deriving Real World Evidence (RWE) for medical and healthcare policy research and decision making. The innovative use of RWD can lead to efficient trial design, enhancement in statistical inference and answer to questions that cannot be addressed using data from randomized clinical trials (RCTs). Nonetheless, many methods for analyzing data from a RCT cannot be directly applied and the usage of causal inference framework is preferrable in RWD analysis. Comparability of patients across treatments and sources is a key requirement for a valid assessment of treatment effect. Propensity score based on individual patient data (IPD) is often used for patient matching to ensure the comparability. However, sharing sensitive IPD is subject to strict regulations and is logistically prohibitive. In such a scenario, propensity scores for individual patients will not be available to the user and the alternative methods should be applied to match RWD or count for heterogeneity across multiple data sources. In this research, we propose methodologies for data matching however without the need of IPD to the user. Based on the formulations of the data analysis, we can see the required components of summary statistics or functions needed from the individual data owners. As demonstrated, through assembling of these components in data analysis, we can achieve the purpose of data matching and valid analysis.