Collaboration Details

Print  Close


Title of Collaborative Activity:

Extraction of adverse drug events from MEDLINE indexing to support a multimodal model for pharmacovigilance.

Description of Collaborative Activity:

Renewal (2019) of an IAA between Center for Drug Evaluation and Research (CDER) of the Food and Drug Administration (FDA) and the Lister Hill National Center for Biomedical Communications (LHNCBC) at the National Library of Medicine (NLM). The collaboration delivers monthly MEDLINE dataset files derived from the Medical Subject Heading (MeSH) indexing (resulting from a 2012 FDA/CDER – NIH/NLM/LHNCBC IAA research collaboration) to harness MEDLINE as a source to support prospective detection of drug-adverse event safety signals from published scientific literature. CDER is responsible for protecting the public health by regulating over-the-counter and prescription drugs, biological therapeutics, and generic drugs. CDER is developing novel, integrative, validated analytic informatics tools to access and analyze new data sources for drug-adverse event signal detection. PEARL (an acronym for ‘Prospective Detection of Emerging Drug-Adverse Event Safety Signals from Relevant Scientific Literature through Quantitative Data Mining of MEDLINE Indexing Terms’) is a prototype, automated, web-based data mining and visual analytics tool that supports the detection of drug-adverse event (ADE) safety signals from published literature reports in MEDLINE through quantitative data mining of Medical Subject Heading (MeSH) indexing terms. MEDLINE dataset file refreshes are needed on a regular monthly basis so that the data mining and safety signal detection outputs from PEARL reflect the most current subset of MeSH-indexed citations in PubMed.

Type of Collaborative Activity:

Resource Development

Year the Collaborative Activity Originated:

2012

NIH Participating Institutes/Centers/Office of the Director:

NLM

HHS Agency Collaborators on this Activity:

FDA