Patient Recruitment Using Electronic Health Records Under Selection Bias: a Two-phase Sampling Framework
share
Electronic health records (EHRs) are increasingly recognized as a cost-effective resource for patient recruitment for health research. Suppose we want to conduct a study to estimate the mean or mean difference of an expensive outcome in a target population. Inexpensive auxiliary covariates predictive of the outcome may often be available in patients' health records, presenting an opportunity to recruit patients selectively and estimate the mean outcome efficiently. In this paper, we propose a two-phase sampling design that leverages available information on auxiliary covariates in EHR data. A key challenge in using EHR data for multi-phase sampling is the potential selection bias, because EHR data are not necessarily representative of the target population. Extending existing literature on two-phase sampling designs, we derive an optimal two-phase sampling method that improves efficiency over random sampling while accounting for the potential selection bias in EHR data. We demonstrate the efficiency gain of our sampling design by conducting finite sample simulation studies and an application study based on data from the Michigan Genomics Initiative.
READ FULL TEXT