Drug Information Journal - March 2009 - (Page 119) Web Tools to Improve PubMed Searches MEDICAL INFORMATION 119 nia, Los Angeles, allows the extraction of bibliometric and statistical data from the PubMed database on the basis of search queries performed by the user. The tool answers the scientific community’s need for information about the impact of a study in a specific field or given sector, the prominence of the researcher or author in a discipline, and the development of drugs over time in a geographical area. Every query entered into PubFocus also generates statistics relating to the evolution of publications over time, that is, frequency of use in quantitative terms and by citations, and publication types. Currently, the main limitations of PubFocus are its slow data processing (explained, however, on the site) and the small number of records used to generate statistics: 2,500. It is nevertheless an excellent instrument, versatile and also adaptable to other databases. Table 2 shows the performance evaluation of the above-mentioned PubMed third-party tools. In addition to this analysis, we developed two tools to help PubMed end users to carry out fast and exhaustive searches. The first one, PubMed Qualified, was created with a template using Microsoft FrontPage software. The prototype simplifies access to PubMed, reorganizing the contents and tailoring them to topics previously selected. Figure 2 shows cancer-related searches using six different strategies: prostate cancer, urological journals with the highest impact factors, presence of full texts, publications in Italian language, multidisciplinary journals, and ahead-ofprint publications. Each strategy reduced information overload, and results were free of the redundancy that usually arises when consulting the whole database. For data protection reasons, the tool was installed in the intranet and protected by passwords. The second tool exploits RSS (Really Simple Syndication) technology. RSS is a web feed format used to publish frequently updated digital content (30). We set up customized feeds using this technology freely available on the web (Figure 3). By means of a special reader we have aggregated results from searches performed on the web, particularly on PubMed. We choose Google Reader, among other tools such as Bloglines (31), FeedReader (32), and Blog Navigator (33). In order to keep the scientific community abreast of the latest developments, we have customized the new aggregator Google Reader (34). The responses show the versatility and highlight the power offered by flexible, specific, low-cost, user-friendly tools in retrieving useful information. Below is a list of the RSS PubMed search queries applied to the reader: • PubMed searches 1: PubFocus OR eTBlast OR Medkit OR XplorMed OR “PubMed assistant” OR PubFinder OR ClusterMed OR GoPubMed OR HubMed OR SLIM OR Ali Baba OR AnonMed OR iHOP OR Twease OR JADE OR PubCrawler OR PubMatrix OR Anne O’Tate OR BiometaCluster • PubMed searches 2: PubMed[ti] OR medline[ti] C O N T I N U I N G E D U C A T I O N CONCLUSION The continuous proliferation of new tools with the aim of improving PubMed searching for the biomedical community is remarkable. The selected and evaluated tools differ in their performance, depending on the users’ profile and customization, the applied information technology to the database imported from the NLM. They made the tools the first qualified instruments able to guide users through the large amount of information on tap. This is their real plus value: they try to meet the user’s information-searching needs. Fast responses and userfriendly search strategies do not necessarily guarantee adequate and exhaustive results. PubMed remains our best reference standard. Despite the high performance levels of these tools, the quality of the results still depends on the syntactic and semantic correctness of the search strategies set and on the proper use of the PubMed MeSH Terms. Acknowledgments—We thank Mrs. Baggot for her valuable English language revision. Drug Information Journal
For optimal viewing of this digital publication, please enable JavaScript and then refresh the page. If you would like to try to load the digital publication without using Flash Player detection, please click here.