Find Datasets from
UMB Researchers

Reporting of Drug Benefit in FDA-Approved Prescription Drug Labeling
UMB Dataset

UID: 96

Author(s): Peter Doshi*, Bansri Desai, Kyungwan Hong, John H. Powers, III * Corresponding Author
Clinicians prescribing therapeutic agents newly approved by the FDA must rely on the information reported through drug labels (also known as “package inserts”). Ideally, labels should incorporate details regarding key efficacy information, including the magnitude of treatment benefit, and measures of precision and random error (confidence intervals and p values). This study assessed the degree to which they actually provide this information. Original drug labels for all new molecular entities (NMEs) approved by the FDA between January 2015 and March 2018 were downloaded from the Drugs@FDA online database. NMEs not treating or preventing diseases or conditions and not tested in controlled trials were excluded. For each indication, the primary efficacy outcome was extracted and the presence or absence of the following attributes describing treatment effects were recorded: point estimates, confidence intervals, and p values. This dataset includes all data extracted and coded as well as copies of the original drug labels used in the study.
2015 - 2018
Subject Domain
Access via Zenodo

Extracted and coded data and copies of original drug labels

Access Restrictions
Unrestricted access
Access Instructions
Available to download from the Zenodo site
Associated Publications
Data Type
Study Type
Dataset Format(s)
Microsoft Excel, PDF
Dataset Size
80.2 MB (PDF), 123.0 kB (Excel)
Grant Support
Faculty Research Funds/University of Maryland, Baltimore