FEATURE ARTICLE

Translating Evidence-Based Decision Making intoPractice: EBDM Concepts and Finding the Evidence

Jane L. Forrest, EdD, RDH1and Syrene A. Miller, BA2

1University of Southern California, School of Dentistry, DEN 4330, Los Angeles, CA 90089-06412The Center for Oral Health, Deer Park, WA

Corresponding AutCalifornia, School ofoffice, 213-740-8669J Evid Base Dent1532-3382/$34.00� 2009 Elsevier Inc.doi:10.1016/j.jebdp.

This is the first of 2 articles that focuses on strategies that can be used to in-tegrate an evidence-based decision making [EBDM] approach into practice.The articles will focus on EBDM methodology and enhancing skills, includinghow to find valid evidence to answer clinical questions, critically appraisethe evidence found and determine if it applies. In addition, online resourceswill be identified to supplement information presented in each article.

The purpose of this article is to define evidence-based decision making anddiscuss skills necessary for practitioners to efficiently adopt EBDM. It will pro-vide a guide for finding evidence to answer a clinical question using PubMed’sspecialized searching tools under Clinical Queries.

Keywords: Evidence-based Decision Making (EBDM), PubMed, PubMed Clinical Queries, Systematic Reviews.

EBDM DEFINED

EBDM isdefined as the formalizedprocess of using the skillsfor identifying, searching for, and interpreting the results ofthe best scientific evidence, which is considered in conjunc-tion with the clinician’s experience and judgment, the pa-tient’s preferences and values, and the clinical/patientcircumstances when making patient care decisions. EBDMis not unique to medicine or any specific health discipline,but represents a concise way of referring to the applicationof evidence to the decision-making process (Fig. 1).

An evidence-based approach has emerged in responseto the need to improve the quality of health care and toclose the gap between research and practice.2,3 Evidence-based practice has been defined as, ‘‘the integration ofbest research evidence with clinical expertise and patientvalues.’’1 This definition is further clarified by definingeach of the 3 major concepts—best evidence, clinical ex-pertise, and patient values. For example, ‘‘best research evi-dence refers to clinically relevant research, especially frompatient-centered clinical research. Clinical expertise means

hor: Jane L. Forrest, EdD, RDH, University of SouthernDentistry, DEN 4330, Los Angeles, CA 90089-0641,; E-mail: jforrest@usc.edu.Pract 2009;9:59-72

All rights reserved.2009.03.017

the ability to use clinical skills and past experience to rap-idly identify each patient’s unique health state and diagno-sis, individual risks and benefits of potential interventions,and personal values and expectations. Patient values refersto the unique preferences, concerns, and expectationsthat each patient brings to a clinical encounter and thatmust be integrated into clinical decisions if they are toserve the patient.’’1 In other words, the use of currentbest evidence does not replace clinical skills, judgment,or experience, but rather provides another dimension tothe decision-making process that also considers the pa-tient’s preferences. Thus, scientific evidence is consideredby the practitioner in the context of an individual patient’scircumstances when it is appropriate.

Several professions have adapted this definition to makeit specific to their discipline. For example, the AmericanDental Association (ADA) defines evidence-based dentistry(EBD) as an approach to oral health care that requires thejudicious integration of systematic assessments of clinicallyrelevant scientific evidence, relating to patient’s oral andmedical condition and history, with the dentist’s clinical ex-pertise and the patient’s treatment needs and preferences.4

EVIDENCE-BASED PRINCIPLES AND LEVELSOF EVIDENCE

EBDM is about solving clinical problems and involves2 fundamental principles:

Figure 1. Evidence-based decision-making process.

JOURNAL OF EVIDENCE-BASED DENTAL PRACTICE

1. Evidence alone is never sufficient to make a clinical de-cision, and

2. A hierarchy of evidence exists to guide clinical decisionmaking.5

The hierarchy of clinical evidence for treatment ques-tions is based on demonstrating that the intervention ortreatment caused the effect, and the ability to control orminimize bias (Fig. 2).6,7

The highest levels of evidence or ‘‘gold standard’’ fortreatment questions are the systematic review and meta-analysis (synthesis of 2 ormore randomized controlled trials[RCTs] answering the same question). Also consideredLevel 1 evidence is an individual RCT. These are followed re-

Figure 2. Study types and levels of clinical evidence.

60

spectively by cohort studies (Level 2), case-control studies(Level 3), and case reports (Level 4) to studies not involvinghuman subjects. An excellent short, graphic review of eachresearch method can be found at the SUNY Downstate Med-ical Center, Evidence-Based Medicine Course, Guide toResearch Methods - The Evidence Pyramid:8 http://library.downstate.edu/EBM2/2100.htm

Systematic reviews (SRs) and meta-analyses (MAs) areconsidered the gold standard for evidence because oftheir strict protocols to reduce bias and to synthesizeand analyze already conducted studies that address thesame question. A meta-analysis is a statistical tool com-monly used with SRs. It involves combining the statisticaldata from the individual studies and conducting anotheranalysis of the combined data. When the data are pooledfrom individual studies, the sample size and powerusually increase. As a result, the combined effect can in-crease precision in estimating the effects of treat-ment.9,10

SRs and MAs facilitate decision making by providinga clear summary of the current state of the existing evi-dence on a specific topic. With more than 2 million arti-cles published annually, SRs provide a way of managinglarge quantities of information10 and make it easier tokeep current with new research. SRs also support the de-velopment of clinical practice guidelines by putting to-gether all that is known about a topic in an objectivemanner. Most recently, evidence-based (EB) methodol-ogy was used by the American Heart Association to updatethe Guidelines for the Prevention of Infective Endocardi-tis,11 by the ADA for the Guidelines on ProfessionallyApplied Topical Fluoride: Evidence-Based Clinical Rec-ommendations,12 and by the Canadian Dental Hygienists’Association on Toothbrushing and on CommerciallyAvailable OTC Rinsing Products.13

Although they are a secondary source of informationbecause they synthesize already conducted research, SRscontain clear descriptions of the aims of the review, thematerial and methods, and a summary of the individualand combined results of the studies.9 Systematic reviewsinclude evidence from RCTs, as well as other well-con-trolled methods. As valuable as SRs can be, their useful-ness and the strength of the evidence derived from theSR depends on the quality of the previously publishedoriginal studies, ie, their ability to meet the eligibility re-quirements for inclusion in the SR.

Systematic reviews differ from traditional literature re-views in that they concentrate on answering a specific clin-ically focused question, making them narrower in scopethan a literature review. A multidisciplinary team of ex-perts in a given area generally conducts the review.9 Theyuse formal and explicit methods, and specify criteria for in-cluding or excluding studies in the review, which is de-signed to reduce bias. The methods used to conducta systematic review surpass what can reasonably beexpected of any one individual.

June 2009

TABLE 1. Skills needed to apply the evidence-baseddecision-making process15

� Convert information needs/problems into clinicalquestions so that they can be answered

� Conduct a computerized search with maximumefficiency for finding the best external evidence withwhich to answer the question

� Critically appraise the evidence for its validity andusefulness (clinical applicability)

� Apply the results of the appraisal, or evidence,in clinical practice

� Evaluate the process and your performance

JOURNAL OF EVIDENCE-BASED DENTAL PRACTICE

In contrast to SRs, traditional literature or narrative re-views are generally directed by an individual using per-sonal experience to define the hypothesis or researchquestion, and select and summarize the literature. Thetraditional approach deals with a broad range of issueson a given topic rather than answering a specific ques-tion in depth. For example, a very specific question an-swered in an SR would be, ‘‘Is use of the nicotinepatch alone as compared to use of the nicotine patchwith behavioral counseling as effective in permanentlyquitting smoking?’’ whereas a broad question in a litera-ture review would be, ‘‘What measures are available forassisting smokers in permanently quitting smoking?’’The broad review could include many types of smokingcessation strategies and may not make comparisons be-tween methods. Also, it is less systematic and more sub-jective in that preestablished criteria for selecting thestudies to include are not specified.14 Finally, a literaturereview does not include combining data or statisticallyanalyzing the data.

Although each level of the hierarchy may contribute tothe total body of knowledge, or a continuum of knowl-edge, ‘‘.not all levels are equally useful for making patientcare decisions.’’7 As you progress up the pyramid towardLevel 1, the number of studies and the amount of availableliterature decreases, while at the same time their relevanceto answering clinical questions increases. Recognizing thelevel of evidence that you have is important to evidence-based practice.

EVIDENCE-BASED DECISION-MAKINGPROCESS AND SKILLS

Evidence-based practice is not just a new term for anold concept, and as a result practitioners need (1)more efficient and effective online searching skills tofind relevant evidence, and (2) critical appraisal skillsto rapidly evaluate the evidence for its validity and ap-plicability. Skills related to the EBDM process are out-lined in Table 1.

Traditionally, the first step in the EBDM approachrequired writing a PICO question where the specificproblem (P), intervention (I), comparison (C), and out-come (O) were identified. This was necessary to narrowthe focus and establish key terms for conducting an ef-fective search. Finding relevant evidence quickly is oftendifficult for practitioners without formal evidence-basedtraining because searching the literature involves havingan understanding of the terminology and indexing(MeSH terms and tree structure) and functions for com-bining search strategies (Truncation, Boolean opera-tors—AND, OR, and NOT). The searches used asexamples in this article were conducted using the MED-LINE database with free access through PubMed(http://pubmed.gov) (Fig. 3).

Volume 9, Number 2

Fortunately, there is a shortcut to accessing relevantevidence through the PubMed ‘‘Clinical Queries’’ feature(Fig. 4). It uses evidenced-based filters to quickly andeffectively access relevant articles using structured algo-rithms that streamline the process of searching theliterature. For example, using the ‘‘Find Systematic Re-views’’ specialized Clinical Query, an individual with lim-ited computer-searching skills can find relevantevidence by typing in a main topic or specific interest,such as fluoride varnish, and automatically be given cita-tions of relevant systematic reviews to answer thequestion.

However, it may be necessary to use PubMed directlywhen citations are not retrieved using Clinical Queries,or for comprehensive searches involving multiple vari-ables and for those conducting individual research stud-ies or systematic reviews. In these cases, there are severalonline resources that provide instruction in how to doso, including the PubMed tutorial (see Figure 3). The fol-lowing case example will demonstrate a search step-by-step to show how questions can be answered using thePubMed Clinical Queries feature and then contrastingit with a traditional PubMed search.

CASE EXAMPLE

Mrs. Wyatt is a 52-year-old woman whose chief complaintis a persistent burning sensation, moderate pain on theanterior two thirds of her tongue, and a metallic taste.She reports continuous symptoms during the day and isbeginning to have trouble falling asleep at night. Thishas been going on for the past 8 months and she hasseen 2 other dentists and a physician who have notbeen able to solve her problem or relieve her symptoms.They even have gone as far as telling her that there’snothing wrong, it’s all in her head and she will have tolive with it, which is not an acceptable solution for her.Upon examination you note there are no lesions or obvi-ous possible causes. Based on her reported symptoms and

61

Figure 3. PubMed home page.

JOURNAL OF EVIDENCE-BASED DENTAL PRACTICE

clinical examination you determine she has burningmouth syndrome and discuss this with her. Not being fa-miliar with the most effective therapy regimens, you tellMrs. Wyatt you will get back to her within a day afteryou’ve had a chance to look up the most current scientificinformation.

PUBMED CLINICAL QUERIES

To address Mrs. Wyatt’s problem, the phrase ‘‘burningmouth syndrome’’ was searched using the ClinicalQueries ‘‘Find Systematic Reviews’’ feature, whichautomatically searches for systematic reviews, meta-analy-ses, reviews of clinical trials, evidence-based medicine,consensus development conferences, and guidelines.Six citations were identified (Figs. 5 and 6).

Fortunately, several SRs were identified; however, forsome clinical topics, there may not be a systematic re-view or other secondary research available. In thesecases, using the ‘‘Search by Clinical Study Category’’Clinical Query is most effective in identifying individualstudies. To use this search tool it is helpful to have anunderstanding of the type of question you are askingand the research method or study design that will pro-vide the highest level of evidence. These are outlined inTable 2.

62

As previously mentioned, had we not found any SRs forburning mouth syndrome, we then would have typedburning mouth syndrome in the ‘‘Search by ClinicalStudy Category’’ (Fig. 7) and searched under the ‘‘Cate-gory’’ of therapy and under the ‘‘Scope’’ of narrow,specific search.

Using this strategy, 15 citations were identified andtheir abstracts can be quickly reviewed to determine thelevel of evidence and to see if any appear applicable(Fig. 8).

If none of the 15 citations proved helpful, then runthe same search using the ‘‘Scope’’ of broad, sensitivesearch. In this case, 229 citations were identified(Fig. 9).

THE PUBMED LIMITS FEATURE

The ‘‘Limits’’ feature is helpful in filtering the citations sothey can be sorted by level of evidence. The ‘‘Limits’’ fea-ture is a folder tab located on the top, left side of the nav-igation bar (Fig. 10).

Click on this tab to be brought to several categories thatcan limit the search findings. For example, it is easy tolimit the 229 citations to those studies that are writtenin English, are conducted with humans (versus animalsor laboratory studies), and are RCTs (Level 1 evidence)

June 2009

Figure 4. PubMed Clinical Queries.

JOURNAL OF EVIDENCE-BASED DENTAL PRACTICE

by checking off the appropriate category: Language, Hu-mans or Animals, and Type of Article. Also, note that the‘‘Type of Article’’ category directly parallels the levels ofevidence (Fig. 11).

Volume 9, Number 2

Results of limiting the search that met these 3 criteriaidentified 11 studies (Fig. 12), which is a much more reason-able starting point to determine if you have found relevantinformation.

63

Figure 5. Search using the ‘‘Find Systematic Reviews’’ Clinical Queries feature.

JOURNAL OF EVIDENCE-BASED DENTAL PRACTICE

THE TRADITIONAL PUBMED SEARCH

When searching on PubMed without using the ClinicalQueries feature for evidence on interventions to treat

Figure 6. Results of search using the ‘‘Find Systematic Rev

64

burning mouth syndrome, first type in burning mouthsyndrome and click on GO. In this case, the 567 citationswere identified (Fig. 13).

iews’’ Clinical Queries feature.

June 2009

TABLE 2. Type of question related to levels of evidence and study methodology

Type of QuestionType of Study or

Methodology of Choice16 Question Focus17,18 Why Study?16-18 Example Questions

Therapy/Prevention

Meta-analysis or System-atic Review [SR] of RCTsSingle RandomizedControlled TrialSR of Cohort Studies

Study effect of therapy ortest on real patients;allows for comparisonbetween interventiongroup and controlgroups for a particularcondition. Largestvolume of EB literature.

To select treatments, if any,that do more good thanharm (improve function,avoid adverse events) thatare worth the effort andcost.

Do sealed permanent firstmolars need lessrestorative treatmentthan unsealedpermanent first molars?

Diagnosis

Meta-analysis or SR ofControlled Trials (Pro-spective cohort study)Single Controlled Trial(Prospective-comparetests with a reference or‘‘gold standard’’ test)

Measures reliability ofa particular diagnosticmeasure/test fora disease against the‘‘gold standard’’diagnostic measure forthe same disease.Sensitivity and specificityof the measures arecompared.

To select and interpretdiagnostic methods ortests. To determine thedegree to which a test isreliable and useful;establish the power of anintervention todifferentiate betweenthose with and withouta target condition ordisease.

How reliable is the ‘‘x’’saliva test as compared tocurrent caries activity forpredicting future cariesactivity?

Etiology, Causation, Harm

Meta-analysis or SR ofRCTsSingle RCTSR of Cohort StudiesSingle Cohort Study(Prospective data collec-tion with formal controlgroup)

Compares a group exposedto a particular agent withan unexposed group.Important forunderstandingprevention and controlof disease.

To identify causes ofa disease or conditionincluding iatrogenicforms. To determinerelationships betweenrisk factors, potentiallyharmful agents, andpossible causes ofa disease or condition.

Does smoking influencevertical alveolar boneloss?

Prognosis

Meta-analysis or SR ofInception Cohort Stud-ies (Follow patients fromwhen disease first be-comes clinically mani-fest)Cohort Study

Follows progression ofa group with particulardisease and compareswith a group without thedisease. Groups must beas similar as possible andmust have good follow-up>80% of each group.

To estimate clinical courseor progression ofa disease or conditionover time and anticipatelikely complications (andprevent them).

What patient and implantcharacteristics influencethe survival of dentalimplants?

JOU

RN

AL

OF

EV

IDE

NC

E-B

AS

ED

DE

NT

AL

PR

AC

TIC

E

Vo

lum

e9,

Nu

mb

er

265

Figure 7. PubMed Clinical Queries Clinical Study Category.

JOURNAL OF EVIDENCE-BASED DENTAL PRACTICE

Next, to identify which of the 567 articles were of the high-est levels ofevidence, additional steps using the ‘‘Limits’’ fea-ture were required. The results of limiting the citations canbe seen by clicking on the History tab. In this case, only 1MA, 11 RCTs, 27 clinical trials (which include the 11

Figure 8. Results of Search by Clinical Study Category, Th

66

RCTs), and 80 case reports were identified (Fig. 14). Theremaining 448 citations did not meet these levels of evi-dence.

Consequently, the one-step process using Clinical Queriesis much easier and faster, as demonstrated by the results of

erapy and Narrow, Specific Search.

June 2009

Figure 9. Search by Clinical Study Category, Therapy and Broad, Sensitive Search.

JOURNAL OF EVIDENCE-BASED DENTAL PRACTICE

finding 6 abstracts of SRs, in comparison with 567 abstracts,which then required additional steps to identify the highestlevels of evidence, the greater majority of which were nota high level of evidence. If you would like more informationabout PubMed and the Clinical Queries feature, access theirtutorial (Fig. 3) at: http://www.nlm.nih.gov/bsd/disted/pubmed.html

Figure 10. Location of the PubMed ‘‘Limits’’ feature on th

Volume 9, Number 2

ACCESSING FULL-TEXT ARTICLES

Some articles are provided as full-text for free throughPubMed. These can be easily downloaded for incorpora-tion into the clinical decision-making process. However,if you need a citation that is not full-text, it can be pur-chased online for a fee. For example, to purchase the

e main search page.

67

Figure 11. PubMed Limits feature.

JOURNAL OF EVIDENCE-BASED DENTAL PRACTICE

68 June 2009

Figure 12. Results of limiting the search by Language, Humans or Animals, and Type of Article.

JOURNAL OF EVIDENCE-BASED DENTAL PRACTICE

SR, ‘‘Interventions for the treatment of burning mouthsyndrome,’’ published by the Cochrane Collaboration,the fee is $25.00. For the article published in OralSurgery, Oral Medicine, Oral Pathology, Oral Radiology,and Endodontics, ‘‘Management of burning mouth syn-drome: systematic review and management recommen-dations,’’ the fee is $31.50 unless you subscribe to thatjournal, then it can be accessed online for free throughElsevier.

ONLINE EVIDENCE-BASED DECISIONMAKING RESOURCES

There are several online resources for those who wouldlike to learn more about EBDM and develop the 5EBDM skills: asking good questions (the PICO process),conducting computerized searches with maximum effi-ciency, critically appraising the evidence, applying theresults, and evaluating your performance of theEBDM process (Table 3). Several of these resourceshave been mentioned in this article and can be usedas staff activities. The following are suggested tutorials

Volume 9, Number 2

and Web sites to reinforce the skills used in theEBDM approach.

CONCLUSION

Recognizing that clinicians have time constraints and yetwant to provide the best possible care to their patients, anevidence-based approach offers a convenient method offinding current research to support clinical decisions, an-swer patient questions, and explore alternative treat-ments, procedures, or materials. It also requiresunderstanding new concepts and developing new skills.The EBDM process provides a strategy for improvingthe efficiency of integrating new evidence into patientcare. Being able to search electronically across hundredsof journals at the same time using PubMed and thePubMed Clinical Queries feature overcomes the chal-lenge to find relevant evidence when it is needed tomake a well-informed decision. By integrating good sci-ence with clinical judgment and patient preferences, cli-nicians maximize the potential for successful patientcare outcomes.

69

Figure 13. PubMed search without using Clinical Queries.

Figure 14. Results of using the Limits feature.

JOURNAL OF EVIDENCE-BASED DENTAL PRACTICE

70 June 2009

TABLE 3. Online EBDM Resources

Evidence-Based Tutorialsand Skill Development Resources

Introduction to EBM, Duke University/UNC http://www.hsl.unc.edu/services/tutorials/ebm/index.htmGuide to Research Methods, The EvidencePyramid, SUNY Downstate Medical Center,SUNY Medical Research Library of Brooklyn

http://library.downstate.edu/EBM2/contents.htmhttp://library.downstate.edu/EBM2/2100.htm

PubMed Tutorial, National Library of Medicine http://www.nlm.nih.gov/bsd/disted/pubmed.htmlEvidence-Based Practice in the Health

Sciences, University of Illinois Chicagohttp://ebp.lib.uic.edu

Forrest, Miller, Overman, & Newman. EBDM: ATranslational Guide for Dental Professionals,Lippincott, June 2009.

http://www.lww.com/product/?978-0-7817-6533-6ISBN-10: 0-7817-6533-1ISBN-13: 978-0-7817-6533-6

Forrest JL. ‘‘Evidence-Based Decision Making:Introduction and Formulating Good ClinicalQuestions.’’

http://www.dentalcare.com/soap/conteduc/index.htmOnline Continuing

Education Course, March 1, 2008.Forrest JL, Miller SA. ‘‘Evidence-Based Decision

Making in Action, Part 1: Finding the BestClinical Evidence.’’ J Contemp Dent Pract,2002;(3)3:010-026.

http://www.thejcdp.com/issue011/index.htm

Forrest JL, Miller SA. ‘‘Evidence-Based DecisionMaking in Action, Part 2: Evaluating andApplying the Clinical Evidence.’’ J ContempDent Pract 2003;(4)1:42-52.

http://www.thejcdp.com/issue013/index.htm

Evidence-Based Centers/Resources

Agency for Healthcare Research and Quality(AHRQ), Evidence-Based Practice Program

http://www.ahrq.govhttp://www.ahrq.gov/clinic/epcix.htm

Evidence-Based Dentistry Champions Conference2008, American Dental Association

http://www.ada.org/prof/resources/ebd/conferences_champion2008.asp

Centre for Evidence-based Dentistry http://www.cebd.orgCentre for Evidence-Based Medicine http://www.cebm.netCentres for Health Evidence http://www.cche.netThe Cochrane Collaboration Library http://www.cochrane.orgEvidence-Based Decision Making,National Center for Dental Hygiene Research

http://www.usc.edu/ebnethttp://www.usc.edu/hsc/ebnet/ebframe/ebbasics.htm

EB Databases and Publications

SUMSearch is a ‘‘meta-search’’ engine forevidence-based medicine resources.

http://SUMSearch.uthscsa.eduUniversity of Texas Health Science Center at San Antonio

Evidence-based Dentistry Journal http://www.nature.com/ebd/index.htmlJournal of Evidence-Based Dental Practice http://www.us.elsevierhealth.com/

product.jsp?isbn¼15323382ADA Guidelines http://www.ada.org/prof/resources/positions/

statements/index.aspThe Trip Database, searches over 61 sites of high-

quality medical information on the Web.http://www.tripdatabase.com

Netting the Evidence access to helpful organizationsand useful learning resources

http://www.shef.ac.uk/�scharr/ir/netting

Drug Databases- CoreyNahman.com updated Daily http://www.coreynahman.com/druginfopage.html- RxList http://www.rxlist.com- MEDLINEplus Health Information http://www.nlm.nih.gov/medlineplus/druginformation.html

JOURNAL OF EVIDENCE-BASED DENTAL PRACTICE

Volume 9, Number 2 71

JOURNAL OF EVIDENCE-BASED DENTAL PRACTICE

REFERENCES

1. Sackett D, Straus S, Richardson W. Evidence-based medicine: how to

practice & teach EBM. 2nd ed. London, England: Churchill Living-

stone; 2000.2. Bader JD, Shugars DA. Variation in dentists’ clinical decisions. J Pub-

lic Health Dent 1995;55(3):181-8.3. Committee on Quality of Health Care in America, IOM. Crossing the

quality chasm: a new health system for the 21st century. Washington

DC: The National Academy of Sciences; 2000.4. American Dental Association. ADA Positions and Statements, ADA

Policy on Evidence-Based Dentistry. Available at: http://www.ada.

org/prof/resources/positions/statements/evidencebased.asp. Ac-

cessed September 20, 2008.5. Evidence-based Medicine Working Group. Users’ guides to the med-

ical literature, a manual for EB clinical practice. Chicago: AMA; 2002.6. Long A, Harrison S. The balance of evidence. Evidence-based deci-

sion making. Health Serv J 1995;6:1-2. Glaxo Wellcome supplement.7. McKibbon A, Eady A, Marks S. PDQ, evidence-based principles and

practice. Hamilton, Ontario: B.C. Decker Inc.; 1999.8. SUNY Downstate Medical Center. Guide to Research Methods, the

Evidence-Based Pyramid. Available at: http://library.downstate.

edu/EBM2/2100.htm. Accessed September 20, 2008.9. Mulrow C. Rationale for systematic reviews. In: Chalmers I,

Altman DB, editors. Systematic reviews. London: BMJ Publishing

Group; 1995, pp 1-8.10. Mulrow C. Rationale for systematic reviews. BMJ 1994;

309(6954):597-9.

72

11. American Heart Association. Guidelines for the prevention of infectiveendocarditis. Circulation, Apr. 19, 2007. Available at: http://circ.ahajournals.org/cgi/reprint/CIRCULATIONAHA.106.183095. Ac-cessed September 20, 2008.

12. American Dental Association. Professionally applied topical fluo-ride: evidence-based clinical recommendations. J Am Dent Assoc2006;137:1151-9.

13. Asadoorian J. CDHA position paper on commercially available over-the-counter oral rinsing products. CJDH 2006;40(4):1-13. Availableat: http://cdha.ca/content/newsroom/position_statements_2006.asp. Accessed September 20, 2008.

14. Sackett D, Haynes R, Guyatt G, et al. Clinical epidemiology: a basicscience for clinical medicine. Boston: Brown & Company; 1999.

15. Rosenberg W, Donald A. Evidence based medicine: an approach toclinical problem-solving. BMJ 1995;310(6987):1122-6.

16. Phillips B, Ball C, Sackett D, Badenoch D, Straus S, Haynes RB,Dawes M. Levels of Evidence. Oxford Centre for Evidence-BasedMedicine. Originally published in 2001, Updated March 2009. Ac-cessed April 16, 2009. http://www.cebm.net/index.aspx?o=1025

17. Haynes R, Wilczynski N, McKibbon A, Walker C, Sinclair J. Develop-ing optimal search strategies for detecting clinically sound studies inMEDLINE. J Am Med Inform Assoc 1994;1(6):447-58.

18. Duke University Medical Center Library. Health Sciences LibraryUniversity of North Carolina at Chapel Hill. Introduction to Evi-dence-Based Medicine, The Well-build Clinical Question. Duke Uni-versity Medical Center Library and Health Sciences LibraryUniversity of North Carolina at Chapel Hill, 2004. Available at:http://www.hsl.unc.edu/Services/Tutorials/EBM/Question.htm.Accessed September 20, 2008.

June 2009