Big Data Can Prevent Adverse Drug Effects

By Katie Wike, contributing writer

Using EHR data, researchers from Stanford University identified two heartburn medications they believe increase the risk of heart attacks.

Analysis of EHR data has the potential to uncover adverse drug reactions not previously identified. For example, researchers from Stanford University mined data from the EHRs of nearly three million patients to come to the conclusion that certain PPIs may increase the risk of heart attack.

PPIs, or heartburn medications, are sold to nearly 20 million US residents both over the counter and by prescription. iHealth Beat explains certain brands of these medications were identified by an algorithm as causing an increased risk of heart attack in patients.

Dr. Nick Leeper, a cardiologist at Stanford and senior author of the study, stressed the research could not prove causation, “but I will say the association looks to be fairly compelling. What we have seen in this study is that PPI usage is very clearly associated with increased risk of heart attack.”

“We figured out a way to analyze the notes that doctors write about their patients during the course of care,” said Nigam Shah, a study author and professor of bioinformatics at Stanford, “and created a data structure which we (could) then analyze for associations that may have clinical significance.”

Others are skeptical, however, that the results of this study should change the way we treat heartburn. Dr. Alan Go, director of the Comprehensive Clinical Research Unit at Kaiser’s Division of Research in Oakland, found the association between PPIs and heart attack risk to be “very weak overall” in the study. “More rigorous data are needed before we make a clinical recommendation to stop using these medications,” he said.