A systematic assessment of the availability and clinical drug information coverage of machine-readable clinical drug data sources for building knowledge translation products. (3rd July 2018)

• Record Type: Journal Article
• Authors: Grandy, Catherine Anne; Donnan, Jennifer R; Peddle, Justin T; Romme, Kristen; Kim, Satpyul; Gamble, John-Michael
• Is Part Of: Journal of the American Medical Informatics Association. Volume 25:Number 9(2018)
• Page Start: 1240
• Page End: 1247
• Publication Date: 2018-07-03
• Subjects: data sources -- databases -- information retrieval -- machine-readable data files -- knowledge translation; Medical informatics -- Periodicals; Information Services -- Periodicals; Medical Informatics -- Periodicals; Médecine -- Informatique -- Périodiques; Informatica; Geneeskunde; Informatique médicale; Computer network resources; Electronic journals; 610.285
• DOI: 10.1093/jamia/ocy074
• Languages: English
• ISSNs: 1067-5027

Abstract

Abstract: Objective: To identify and describe clinical drug data sources that have the potential to serve as a repository of information for developing drug knowledge translation products. Methods: Two reviewers independently screened citations from PubMed and Embase, websites from the web search engine Google, and references from selected journals. Publicly licensed or non-proprietary data sources containing clinical drug information accessible in a machine-readable format were eligible. Data sources were assessed for their coverage across 18 pre-specified domains and 74 elements of clinical drug information. Results: Of the 3369 unique citations or webpages screened, 44 drug information data sources were identified. Of these, 22 data sources met the study inclusion criteria. There was a mean of 4.5 (SD = 5.19) domains covered by each source and a mean of 10.9 (SD = 18) elements covered by each source. None of the data sources covered all domains and eight elements were not addressed by any source. All of the data sources identified by the study are government or academic databases. Conclusion: Our study demonstrated the availability of machine-readable clinical drug data that could help facilitate the creation of novel drug knowledge translation products. However, we identified clinical content gaps in the available non-proprietary drug information sources. Further evaluation of the quality of each data source would be necessary prior to incorporating these sources into any knowledge translation products intended for clinical use.