Mining electronic health records for adverse drug effects using regression based methods

摘要: The identification of post-marketed adverse drug events (ADEs) is paramount to health care. Spontaneous reporting systems (SRS) are currently the mainstay in pharmacovigilance. Recently, electronic records (EHRs) have emerged as a promising and effective complementary resource SRS, they contain more complete record patient, do not suffer from biases inherent SRS. However, mining EHRs for potential ADEs, which typically involves statistical associations between drugs medical conditions, introduced several other challenges, main one being necessity techniques that account confounding. objective this paper present demonstrate feasibility method based on regression techniques, designed automated large scale ADEs EHR narratives. To best our knowledge first its kind approach combines both use data, methods order address confounding identify ADEs. Two separate experiments conducted. results, validated by clinical subject matter experts, great promise, well highlight additional challenges.