EBM commonly consists of five steps, which are often referred to as the "5 A's of EBM":
Construct an answerable question using the PICO framework or other clinically appropriate approach.
Convert your question to a search strategy and locate the best evidence.
Critically evaluate the evidence for reliability and validity.
Integrate the evidence with patient preferences and clinical experience.
Determine the effectiveness of the process for improving patient care and outcomes.
Precision = Number of relevant documents retrieved / Total number of documents retrieved.
Recall = Number of relevant documents retrieved / Total number of relevant documents in the corpus.
The PICO Framework helps clinicians design an answerable question by breaking the clinical problem into searchable concepts. A complete search of the literature is needed to determine the best evidence to answer a question.
P: patient, population, problem
I: intervention, exposure
C: comparison or alternative treatment
O: outcome being measured
Identify "synonymous" terms and phrases found in the literature
Combine like terms and phrases for a single concept with the Boolean OR, then combine concepts with AND
Once you have developed a complete search strategy then use the following database features to narrow results to higher levels of evidence relevant to your question.
Type of Question | Description | Template | Type of Study |
---|---|---|---|
Etiology/Exposure/Harm | Does this intervention/exposure cause this condition? | Are (patient) who have (intervention/exposure), compared with those without (comparison) at (more/less) risk for (outcome)? | RCT>cohort>case control>case series |
Diagnostic | How accurate is this diagnostic test? | In (patient) are/is (intervention) compared with (comparison) more accurate in diagnosing (outcome)? | Independent, blind comparison to gold standard (cross-sectional) |
Treatment | Is this intervention more effective than another? | In (patient and/or problem), how does (intervention) compared with (comparison intervention) affect (outcome)? | RCT |
Prognostic | What is the likely outcome, progression or survival time for this condition? | In (patient) how does (intervention/exposure) compared with (comparison) influence (outcome)? | Cohort>case control>case series |
Level of Evidence | Description | Example Resources to Search |
---|---|---|
Systems (Highest Level) |
Evidence combined with EHR patient data using standard data structure and terminology | Computerized Decision Support embedded in EHR using FHIR HL7 and SNOMED CT |
Summaries |
Guidelines Point of Care Tools rank levels of evidence for recommendations using standard GRADE |
ECRI Guidelines UpToDate, DynaMed |
Synopses of Syntheses | Umbrella Reviews | Health Evidence |
Syntheses |
Focus on a specific clinical question Meta-analyses of RCTs Systematic Reviews |
Cochrane Library |
Synopses of Studies | Critical Appraisal of single studies | ACP Journal Club |
Primary Studies | RCT > Cohort > Cross-sectional > Case-Control > Case series/Case report | PICO search of databases PubMed, EMBASE, APA PsycInfo, etc. |
Expert Opinion (Lowest Level) |
Provide a general overview of foundational science, systems, diseases and symptoms. Often the best source for an overview of a large topic and background information that does not change frequently. Not a good source for new diseases, new diagnostic tests, new treatments, or co-morbidities and confounders. References can point toward higher levels of evidence but have the potential to be "cherry picked." The search is not reproducible, not assessed for risk of bias, and levels of evidence for recommendations are not ranked using a standard system. | Textbooks, Step Prep, Review Articles |
Below the pyramid: Blogs, Wikipedia, Organization web sites, Trade journals, Discussion boards
Haynes R. B. Of studies, syntheses, synopses, summaries, and systems: the "5S" evolution of information services for evidence-based healthcare decisions. Evidence-based medicine,. 2006; 11(6), 162–164. https://doi.org/10.1136/ebm.11.6.162-a
Dicenso A, Bayley L, Haynes RB. Accessing pre-appraised evidence: fine-tuning the 5S model into a 6S model. Evid Based Nurs. 2009;12(4):99-101. https://doi.org/10.1136/ebn.12.4.99-b
For patients aged 65 years or older (P), how does the use of an influenza vaccine (I) influence the risk of developing pneumonia during the flu season (O)? [The comparison (C) is not receiving an influenza vaccine]
Identify synonymous terms and controlled vocabulary found in the literature
P: Aged OR elderly OR geriatric OR "older adult*"
I: "Influenza Vaccines"[Mesh] OR ((flu OR influenza)(vaccin* OR immunization*))
C:
O: "Pneumonia"[Mesh](risk OR prevent*)
Search Strategy: ("Aged"[Mesh] OR elderly OR geriatric OR "older adult*") AND ("Influenza Vaccines"[Mesh] OR ((flu OR influenza)(vaccin* OR immunization*))) AND ("Pneumonia"[Mesh](risk OR prevent*))
Limit to last 10 years, and higher level studies appropriate to the question