Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Filters applied. Clear all
. 2013 May 1;20(3):404-8.
doi: 10.1136/amiajnl-2012-001482. Epub 2013 Mar 6.

Web-scale Pharmacovigilance: Listening to Signals From the Crowd

Affiliations
Free PMC article

Web-scale Pharmacovigilance: Listening to Signals From the Crowd

Ryen W White et al. J Am Med Inform Assoc. .
Free PMC article

Abstract

Adverse drug events cause substantial morbidity and mortality and are often discovered after a drug comes to market. We hypothesized that Internet users may provide early clues about adverse drug events via their online information-seeking. We conducted a large-scale study of Web search log data gathered during 2010. We pay particular attention to the specific drug pairing of paroxetine and pravastatin, whose interaction was reported to cause hyperglycemia after the time period of the online logs used in the analysis. We also examine sets of drug pairs known to be associated with hyperglycemia and those not associated with hyperglycemia. We find that anonymized signals on drug interactions can be mined from search logs. Compared to analyses of other sources such as electronic health records (EHR), logs are inexpensive to collect and mine. The results demonstrate that logs of the search activities of populations of computer users can contribute to drug safety surveillance.

Figures

Figure 1
Figure 1
Venn diagram showing the different user groups in our analysis (not drawn to scale).
Figure 2
Figure 2
Percentage of users in each of the three user groups searching for hyperglycemia-related terms. Percentage is computed per week over 12 months of search log data. Background refers to the fraction of all searchers who search for hyperglycemia-related symptoms or terminology independent of the presence of the drugs in the users’ search histories.
Figure 3
Figure 3
Receiver operating characteristic curve for the identification of drug pairs known to be associated with hyperglycemia using search log data. Red (dashed) line denotes the performance when using all hyperglycemia-related terminology in our set. Yellow (solid) line denotes the performance of a more narrowly focused set of symptoms strongly connected to hyperglycemia.
Figure 4
Figure 4
Influence of removing symptoms and conditions on the classification performance as measured by change in AUCAll. AUC, area under the curve.

Similar articles

See all similar articles

Cited by 68 articles

See all "Cited by" articles

MeSH terms

Feedback