Elsevier

Journal of Biomedical Informatics

Volume 59, February 2016, Pages 42-48
Journal of Biomedical Informatics

Early identification of adverse drug reactions from search log data

https://doi.org/10.1016/j.jbi.2015.11.005Get rights and content
Under an Elsevier user license
open archive

Highlights

  • Study use of search logs for the early detection of known adverse drug reactions (ADRs).

  • Employ a time-indexed reference standard as ground truth in our prospective analysis.

  • Propose novel methods to control for the advance availability of information online.

  • Show that logs can detect new ADRs, the central challenge in pharmacovigilance.

Abstract

The timely and accurate identification of adverse drug reactions (ADRs) following drug approval is a persistent and serious public health challenge. Aggregated data drawn from anonymized logs of Web searchers has been shown to be a useful source of evidence for detecting ADRs. However, prior studies have been based on the analysis of established ADRs, the existence of which may already be known publically. Awareness of these ADRs can inject existing knowledge about the known ADRs into online content and online behavior, and thus raise questions about the ability of the behavioral log-based methods to detect new ADRs. In contrast to previous studies, we investigate the use of search logs for the early detection of known ADRs. We use a large set of recently labeled ADRs and negative controls to evaluate the ability of search logs to accurately detect ADRs in advance of their publication. We leverage the Internet Archive to estimate when evidence of an ADR first appeared in the public domain and adjust the index date in a backdated analysis. Our results demonstrate how search logs can be used to detect new ADRs, the central challenge in pharmacovigilance.

Keywords

Pharmacovigilance
Search log analysis
Adverse drug reactions

Cited by (0)

1

Work done while employed as an intern at Microsoft.