Usability of Selected Databases for Low-Resource Clinical Decision Support

Summary

Background: Smartphones are increasingly important for clinical decision support, but smartphone and Internet use are limited by cost or coverage in many settings. txt2MEDLINE provides access to published medical evidence by text messaging. Previous studies have evaluated this approach, but we found no comparisons with other tools in this format.

Objectives: To compare txt2MEDLINE with other databases for answering clinical queries by text messaging in low-resource settings.

Methods: Using varied formats, we searched txt2MEDLINE and five other search portals (askMED-LINE, Cochrane, DynaMed, PubMed PICO, and UpToDate) to develop optimal strategies for each. We then searched each database again with five benchmark queries, using the customized search-optimization formats. We truncated the results to less than 480 characters each to simulate delivering them to a maximum of three text messages. Clinicians with practice experience in low-resource areas scored the results on a 5-point Likert scale.

Results: Median scores and standard deviations from 17 reviewers were: txt2MEDLINE, 3.2±0.82 (control); askMEDLINE, 3.2±0.90 (p = 0.918); Cochrane, 3.8±0.58 (p = 0.073); DynaMed, 3.6±0.65 (p = 0.105); PubMed PICO, 3.6±0.82 (p = 0.005); and UpToDate, 4.0±0.52 (p = 0.002). Our sample size was sufficiently powered to find differences of 1.0 point.

Conclusions: Comparing several possible sources for texting-based clinical-decision-support information, our results did not demonstrate one-point differences in usefulness on a scale of 1 to 5. PubMed PICO and UpToDate were significantly better than txt2MEDLINE, but with relatively small improvements in Likert score (0.4 and 0.8, respectively). In a texting-only setting, txt2MEDLINE is comparable to simulated alternatives based on established reference sources.

Keywords

Library information systems - handheld devices - ambulatory care - general healthcare providers - economic barriers

 

Conflict of Interest

The authors declare that they have no conflicts of interest in the research.

• References

• 1 Cisco Systems.. Global Mobile Data Traffic Forecast Update, 2011–2016. San Jose: Cisco Systems; 2012. 29 p.
• 2 Lorence DP, Park H, Fox S. Racial disparities in health information access: resilience of the Digital Divide. J Med Syst 2006; 30 (04) 241-249.
• 3 Economist Newspaper Ltd.. A special report on telecoms in emerging markets. The Economist. 2009 Sep 24 3-19.
• 4 Fontelo P, Liu F, Muin M, Tolentino T. txt2MEDLINE: text-messaging access to MEDLINE/PubMed. AMIA Annual Symposium Proceedings 2006: 259-263.
• 5 Ely JW, Osheroff JA, Chambliss ML, Ebell MH, Rosenbaum ME. Answering physicians’ clinical questions: obstacles and potential solutions. J Am Med Inform Assoc 2005; 12 (02) 217-224.
• 6 Marcelo AB, Fontelo P. An Evaluation of the Usability of Accessing MEDLINE over Short Messaging System. MEDINFO 2007: 235-236.
• 7 Armstrong K, Liu F, Seymour A, Mazhani L, Littman-Quinn R, Fontelo P. et al. Evaluation of txt2MED-LINE and Development of Short Messaging Service-Optimized, Clinical Practice Guidelines in Botswana. Telemed J E Health 2012; 18 (01) 1-4.
• 8 txt2MEDLINE: search by SMS [Internet]. Bethesda, MD: US National Library of Medicine; c2005–06 [updated c2005–06; cited 2012 Mar 1–20]. Available from: http://sms.nlm.nih.gov/tbl/txt.php.
• 9 askMEDLINE [Internet].. Bethesda, MD: US National Library of Medicine; c2005–06 [updated c2005–06; cited 2012 Mar 1–20]. Available from: http://askmedline.nlm.nih.gov/ask/ask.php.
• 10 Search MEDLINE/PubMed via PICO [Internet].. Bethesda, MD: US National Library of Medicine; c2005–06 [updated c2005–06; cited 2012 Mar 1–20]. Available from: http://pubmedhh.nlm.nih.gov/nlmd/pico/piconew.php.
• 11 The Cochrane Library [Internet].. New York: John Wiley & Sons; c1999 [updated 2012 Mar 1; cited 2012 Mar 1–20]. Available from: http://www.thecochranelibrary.com.
• 12 DynaMed [Internet].. Ipswich, MA: EBSCO Publishing; c1995 [updated 2012 Mar 1; cited 2012 Mar 1–20]. Available from: http://www.ebscohost.com/DynaMed.
• 13 UpToDate [Internet].. Alphen aan den Rijn: Wolters Kluwer; c1992 [updated 2012 Mar 1; cited 2012 Mar 1–20]. Available from: https://www.uptodate.com.
• 14 Tom O, Fontelo P, Liu F. Do computer-generated summaries, ,,The Bottom Line (TBL)“ accurately reflect published journal abstracts?. AMIA Annu Symp Proc Oct 2007; 11: 1135.
• 15 Reddy M, Gill SS, Kalkar SR, Wu W, Anderson PJ, Rochon PA. Oral Drug Therapy for Multiple Neglected Tropical Diseases: A Systematic Review. JAMA 2007; 298 (16) 1911-1924.
• 16 Bauer, DF.. Constructing confidence sets using rank statistics. J Am Stat Assoc 1972; 67: 687-690.
• 17 Hollander M, Wolfe DA. Nonparametric Statistical Methods. New York: John Wiley & Sons; 1973. p. 27-33, 68-75.
• 18 McGraw KO, Wong SP. Forming inferences about some intraclass correlation coefficients. Psychological Methods 1996; 1: 30-46.
• 19 Shrout PE, Fleiss JL. Intraclass correlation: uses in assessing rater reliability. Psychological Bulletin 1979; 86: 420-428.
• 20 R Development Core Team.. R: A language and environment for statistical computing. Vienna: R Foundation for Statistical Computing; 2011
• 21 UpToDate International Grant Subscription Program [Internet].. Boston: Global Health Delivery Project; c2010 [updated c2010; cited 2012 May 6]. Available from: http://www.ghdonline.org/uptodategrant.