1. Using the Internet to Answer Clinical Questions STFM Spring Conference - 2006 Brian S. Alper, MD, MSPH Anne-Marie Lozeau, MD, MS Beth Potter, MD

2. Disclosure “A funny thing happened to me on the way to rural family practice…” Dr. Alper is Editor-in-Chief of DynaMed™ and Medical Director of EBSCO Publishing

3. No Disclosures Anne-Marie Lozeau, MD, MS Beth Potter, MD University of Wisconsin - Madison Department of Family Medicine

4. Learning Objectives Recognize that using the Internet to meet clinical information needs is practical Meet many useful web sites Learn efficient strategies for selecting useful resources for specific needs Use a free Internet portal to organize Internet use during patient care

5. Our Goal in Medicine Provide the best care… Provide patients the best information to guide health care decisions… Improve health outcomes… …based on the “truth” – separating medical knowledge from folklore

6. How do we know medicine? WE = society MEDICINE = clinical knowledge Scientific investigation Original research published in journals Systematic reviews EVIDENCE-BASED MEDICINE

7. How do we know medicine? WE = individual clinicians MEDICINE = clinical knowledge Consultants Colleagues Lectures Textbooks Rounds Precepting Guidelines CME Experts Experience PRACTICAL choices selected for efficiency

8. “Just-in-Case” vs. “Just-In-Time” Apprentice-styleMemorization Soak up information now Retrieve information during practice Knowledge Mastery Find information when you need it Adult learning –Problem-oriented –Clinical application

9. Using Traditional Textbooks Not always written for clinical practice May not provide best research evidence (Are opinions well supported, or selectively supported?) May be years out-of-date Finding specific information within the text can be time-consuming

10. Using Journal Articles Individual journal articles –may not relate to specific information need –may not provide complete picture Finding one journal article can be time- consuming, let alone finding all the relevant articles Articles are written to promote research findings, not often written for clinical application BUT THIS IS OUR EVIDENCE SOURCE

11. Evidence-Based Medicine Definition: Integration of best research evidence with clinical expertise and patient values. –Sackett DL, Straus SE, Richardson WS, Rosenberg W, Haynes RB. Evidence-Based Medicine. How to Practice and Teach EBM. 2nd ed. London: Harcourt Publishers Ltd. 2000. p. 1.

12. Best Research Evidence Comprehensive – evidence only known to be best if all the available evidence is known Valid – critical appraisal determines potential for bias Systematic – selection and evaluation of evidence by protocol reduces author bias, investigator bias Current – every day new evidence could be best Synthesized – one study vs. the whole picture

13. Ideal Sources for Clinicians Comprehensive – covering all relevant articles AND information needs. Valid/Systematic – driven by best available evidence. Relevant – focus on clinically relevant information. Easy to use – organized for use by clinicians. Affordable

14. Validity (“Truth”) Consider validity of original research PLUS validity of translation –Was the best evidence identified? –Was the best evidence selected? –Was the evidence evaluated well? –Is the summary accurate? –Are conclusions based on the underlying evidence?

15. Evidence-Based = conclusions based on best available evidence “Evidence-based clinical reference” requires the following: 1. Systematically identifying all applicable evidence 2. Systematically selecting the best available evidence from that identified 3. Systematically evaluating the selected evidence (critical appraisal) 4. Accurately summarizing the evidence and its quality 5. Making conclusions dependent on the evidence 6. Synthesizing multiple bits of evidence for overall conclusion 7. Changing the conclusions when new evidence alters the best available evidence

16. Systematic Evidence Identification Systematic Literature Search Commonly called Systematic Review Start with question, search for best available answers Systematic Literature Surveillance Also called Research Monitoring Start with literature being published, determine how best to use

17. Systematic Literature Search Start with clinical question Define inclusion criteria Search for all relevant information Assess methodology (validity) Combine results Determine clinical conclusion Apply to practice

18. Using Systematic Reviews to Answer Clinical Questions Cochrane Reviews –may be “gold standard” when available and current –many sources use Cochrane reviews Other Reviews - check whether process (search, evaluation) is systematic

19. Using Systematic Reviews to Answer Clinical Questions More difficult for non-RCT answers Consider additional info Most valid method for answering an individual question Not adequate to meet most information needs

20. Systematic Literature Surveillance Start with defined literature set Define inclusion criteria Select all relevant information Assess methodology (validity) Summarize results Determine clinical conclusion

21. Systematic Literature Surveillance - Uses Alert clinicians to critical info (EBM article summaries) Catalog for future searches (EBM search engine) Update clinical reference information (Integrated EBM)

22. Systematic Literature Surveillance - Advantages Cover all types of information - Diagnosis - Treatment - Etiology - Prognosis Keep up with current research Address current questions

23. Systematic Literature Surveillance - Disadvantages Can miss the big picture –1 study may be “rebuttal” –“ping-pong” between sources – overemphasis of study results Need to read collection of studies Miss older research (unless retrospective literature surveillance)

24. Ideal Sources Comprehensive – covering all relevant articles AND information needs. Valid/Systematic – driven by best available evidence. Relevant – focus on clinically relevant information. Easy to use – organized for clinician uses. USE SYSTEMATIC REVIEWS AND SYSTEMATIC LITERATURE SURVEILLANCE

25. Internet Portals: Useful Links for Patient Care Organized for effective searches A starting point for use during patient care Access from any computer with internet connection Links to all resources in this presentation You can make your own!

26. Examples of Portals My HQ: www.myhq.com www.myhq.com “Dr Alper’s links” www.myhq.com/public/a/l/alper www.myhq.com/public/a/l/alper I keep bookmarks www.ikeepbookmarks.com www.ikeepbookmarks.comwww.ikeepbookmarks.com

30. * Asterisk - Requires Registration - MAY Require $$ Subscription

34. Clinical Scenario 52-year old male presents with Shingles (zoster) for 2 days. Clinical Concerns: –short-term symptom relief –prevention of Post-Herpetic Neuralgia

35. EVIDENCE-BASED MEDICINE Highly Valid EBM sources Relevant - many have direct clinician input FIRST CHOICE if fast enough (many are) FIRST CHOICE when research likely –Quality research is possible? –Research is present? (difficult to guess)

38. Using EBM Search Engines Examples: TRIP, InfoRetriever Broader than Systematic Reviews alone Limitations: –individual article summaries may miss the “big picture” –may have to read multiple items –No subject overview

40. TRIP Database Turning Research Into Practice Identifies free websites with EBM and other medical content Catalogs titles and URLs to each page Inclusion based on Relevance and Rigor 300+ new articles added monthly Searches synonyms, cues misspelling

41. TRIP Database EBM Resources: systematic reviews, CATs, journal clubs Clinical Guidelines Query-answering Medical Images: over 80,000 eTextbooks: background information Patient Information Journals – automatically searches PubMed with EBM filters

42. TRIP Database Comprehensiveness: high Validity: variable Relevance: moderate Convenience: easy to use Affordability: will be free

52. InfoRetriever - 2005 2828 POEMs 2393 Cochrane Abstracts 734 Clinical Guidelines 229 Clinical Decision Rules 1373 Diagnostic Test Calculators 1345 H&P calculators ICD-9 search tool 5-Minute Clinical Consult www.InfoPOEMs.com

61. InfoRetriever Comprehensiveness: moderate Validity: variable –High for most content –Low for 5-Minute Clinical Consult Relevance: high Convenience: takes practice Affordability: $249/year

63. DynaMed Integrated EBM Resource combines best available evidence with clinical focus Updated continuously with systematic literature surveillance –> 500 Journals directly or via 10 journal reviews services –Cochrane reviews –National Guideline Clearinghouse Template system for rapid browsing

64. DynaMed Peer Review in process Includes ICD-9 Codes and patient handout links for each topic Easy to navigate –Alphabetical topic browse –Standardized templates –Some research-heavy topics can get difficult to navigate UNTIL new topics created

66. shing

75. DynaMed Comprehensiveness: shown to answer 69% of clinical questions in primary care in 2004 Validity: variable & explicit Relevance: high, clinical focus Convenience: easy to use Affordability: $200/year or effort

76. Ideal Sources Comprehensive – covering all relevant articles AND information needs. Valid/Systematic – driven by best available evidence. Relevant – focus on clinically relevant information. Easy to use – organized for clinician uses. USE SYSTEMATIC REVIEWS AND SYSTEMATIC LITERATURE SURVEILLANCE

77. Key sources for Systematic Reviews Cochrane Library DARE Major benefit : best evidence (at time of review) Major limitation : limited scope of coverage Clinical Evidence AHRQ

78. Highlights 200 common clinical topics Lists treatments for those topics Qualitatively describes Systematic Reviews and RCT data for and each treatment Compiles results for easy use Includes relevant background information Updates searches every 12 months Peer Reviewed

84. Clinical Evidence Comprehensiveness: modest Validity: high - limited to RCTs Relevance: high - clinical focus Convenience: easy to use Affordability: $97/year –free for US physicians via United Health Foundation for 6 months

85. PubMed Clinical Queries Use search filters that have been researched by Haynes, et al. who are involved with the McMasters EBM group Help to refine EBM searches in PubMed

93. Clinical Queries Comprehensiveness: moderate Validity: High Relevance: Low –Often finds many articles –Still need to sift through a lot of articles Convenience: takes practice –New feature allows you to receive updates when new info is added to your search Affordability: Free

94. UpToDate Review Articles Heavily Referenced Rapid Search Engine Broad Coverage

95. UpToDate Frequently Updated Electronic Textbook Limitations –reference-based is not “evidence-based” –specialty-oriented –varying quality

103. UpToDate Comprehensiveness: shown to answer 34% of internists’ clinical questions Validity: varies with author Relevance: high, clinical focus Convenience: variable, non-standard presentation Affordability: $495 then $395/year