Researchers Investigate the Effect of Labeling Changes on PPI Use in Sentinel

In the first-ever study using the US Food and Drug Administration's (FDA) Sentinel system by non-FDA stakeholders, a team of academic and industry researchers found that a 2010 label change for proton pump inhibitors (PPIs) led to less long-term use of the drugs.

The study is also the second to be conducted using Sentinel to evaluate the impact of an FDA-mandated label change after a study by FDA officials and other researchers published in 2016 found that the use of long-acting β2-agonists (LABA)  decreased after FDA added new warnings to the drugs' labels.

Sentinel

Sentinel was initially developed as a tool for active postmarket safety surveillance for FDA as required by Congress under the 2007 Food and Drug Administration Amendments Act (FDAAA). In the decade since its launch, FDA has expanded Sentinel to include new tools and access to data for more than 223 million US patients.

As a means of providing third party researchers in academia, industry and other government agencies with access to the bulk of Sentinel's resources, the Innovation in Medical Evidence Development and Surveillance (IMEDS) was formed in collaboration with the non-profit Reagan-Udall Foundation.

Study

In May 2010, FDA issued a class-wide label change for PPIs due to a potential increased risk of hip and other fractures with prolonged use of the drugs reported in several observational studies and meta-analyses.

However, according to the authors of the study, it is often difficult to determine the impact label changes have on prescriber and patient behavior.

"The impact of label changes on physician and patient behavior is understudied with few examples, and when assessments are done, as with US [risk evaluation and mitigation strategies] REMS, many are not publicly disclosed or of sufficient scientific interest or rigor to be publishable," they wrote.

The authors also said that healthcare databases, such as those available through Sentinel are "underused" compared to cross-sectional patient and prescriber surveys.

Using Sentinel, the authors were able to identify nearly 1.5 million patients that took PPIs in the three years before the label change and nearly 2.25 million patients that took the drugs in the five years after.

Among those patients, the authors found that the proportion taking PPIs for longer than a year decreased from 8.4% to 7.5% and that the number of days patients were supplied PPIs fell from 130.4 days to 113.7 days in the period following the label change.

While the authors found a decrease in osteoporosis screening and interventions following the label change, they found the proportion of patients that had fractures decreased from 4.4% to 3.1% across the two periods.

"Taken at a population level, our results overall are at least suggestive that label changes and communication tools to alter certain prescriber behaviors as measured by dispensing patterns," the authors wrote, noting that their finding is consistent with FDA's finding in its study of label changes for LABAs.

Pharmacoepidemiology & Drug Safety