Risk of Liver Injury With Potentially Hepatotoxic Drugs—Reply

In Reply We thank Björnsson for their Letter to the Editor about our study in JAMA Internal Medicine. Our study introduced innovative methods for improving the understanding of the risk for hepatotoxic effects associated with medications. We developed a systematic, reproducible approach that used electronic health record data to measure rates of hospitalization for severe acute liver injury (ALI) following medication initiation among patients without liver or biliary disease. We reduced the likelihood of capturing non–drug-related severe ALI by censoring as nonevents any patients with diagnoses of hepatobiliary disease, ischemic injury, alcoholic hepatitis, viral hepatitis, or other conditions that could precipitate ALI, encompassing more than 700 diagnoses from the International Classification of Diseases, Ninth and Tenth Revisions (ICD-9 and ICD-10) (included in eTables 2 and 3 in the supplementary materials of our article). We acknowledged that some severe ALI events might have been caused by non–drug-related conditions that evaded censoring because they were not recorded as discharge diagnoses. However, we do not believe that these conditions were differentially distributed across the medication initiator cohorts. We would welcome others to apply our approach to their data to evaluate the generalizability of these findings.

from JAMA Internal Medicine Online First https://ift.tt/1Ja7mqf