Skip to Main Content

Evidence Based Health Care (EBHC)

Objectives and Definitions

Learn to use databases and tools available to locate methodologically sound and clinically relevant journal articles. Practice using resources such as DynaMed, ACP Journal Club, and the Cochrane Library. Use the clinical query filters in PubMed to limit searches to the study designs that best answer your question. At least one month of searching experience is advised as a prerequisite to this class.
  1. Recognize the hierarchical nature of evidence, represented by the level of evidence pyramid.
  2. Search clinical information resources such as DynaMed, ACP Journal Club, and the Cochrane Library, noting level of evidence designations.
  3. Formulate a clinical question using the PICO format.
  4. Limit PubMed searches to the study design that best answers a specific type of question, using clinical query filters.   
What is evidence-based practice?
the conscientious, explicit, and judicious use of current best evidence in making decisions about the care of individual patients. --Sackett DL, et al. Evidence-based medicine: what it is and what it isn’t. BMJ 1996;312 (7023):71-72.  
“Evidence-based medicine (EBM) is the integration of best research evidence with clinical expertise and patient values.” --Guyatt G, et al. Evidence-Based Medicine working Group. Evidence-based medicine. A new approach to teaching the practice of medicine. JAMA 1992; 268: 2420-5.

Why is retrieving “current best evidence difficult?

  • Glut of published literature
  • Quality of literature
  • Novice searching skills

Formulating a Question

Formulating a Clinical Question
The well-built clinical question: a key to evidence-based decisions. Richardson WS, Wilson MC, Nishikawa J, Hayward RS. ACP J Club. 1995 Nov-Dec; 123(3): A12-3.


Example Clinical Scenarios:

Diagnosis: How to select a diagnostic test or how to interpret the results of a particular test.

  • In patients with suspected depression what is the accuracy of a two-question case-finding instrument for depression compared with six previously validated instruments?
  • In patients with no symptoms, which screening test is most accurate for diagnosing breast cancer?


Therapy: Which treatment is the most effective, or what is an effective treatment given a particular condition.

  • In children with colds, are zinc lozenges safe and effective for relief of cold symptoms?
  • In patients with acute bronchitis, do antibiotics reduce sputum production, cough or days off work?


Harm or Etiology: Are there harmful effects of a particular treatment, or how can these harmful effects be avoided.

  • What are the risks of treating asthma patients with beta-adrenergic agonists?
  • In women taking oral contraceptives, is there an association between their use and cardiovascular disease?


Prognosis: What is the patient's likely course of disease, or how to screen for or reduce risk.

  • In a healthy woman who has suffered a miscarriage, what is the normal grief process and are any factors associated with longer than normal grieving?
  • What are the risk factors for developing diabetes mellitus type II in adolescent males?

Define the clinical problem using the PICO components described above:



Now describe your own clinical scenario:


Create a PICO for the scenario –

P –



I –



C –



O –



Types of Evidence Based Resources

Finding the best evidence in the literature

Online tools and databases where you find evidence-based resources can be divided into three categories:

1. Those that do most of the work for you

2. Those that do some of the work for you

3. Those where you do most of the work

But even when most of the work is done for you you will still need to use your critical judgment!

Level of Evidence

Type 1


       1.   Cochrane Database of Systematic Reviews (Cochrane Reviews) rigorous, full-text reviews done by


Cochrane Review Groups with explicitly stated objectives and eligibility criteria, a comprehensive literature search, comparison of studies to criteria, justification of exclusions plus occasional updates


- Every Cochrane Systematic Review must have a standardized title including: Intervention / for a specific Problem / in a Patient Population (when appropriate)

- Every review goes through the following stages:


Title Review > Protocol Review > Protocol Publication > Review Development > Review Publication > Update > And sometimes, Withdraw


2.   Database of Abstracts of Reviews of Effectiveness (DARE or Other Reviews) - systematic reviews published in journals and other, non-Cochrane sources, with quality assessed structured abstracts, produced by the Centre for Reviews and Dissemination (CRD) at the University of York, UK.


3.   Cochrane Central Register of Controlled Trials (Clinical Trials) journal citations of randomized trials, just under half are from MEDLINE, the remainder are from the European medical index EMBASE, as well as other published and unpublished sources, including Cochrane Groups Specialized Registers.


4.   Cochrane Database of Methodology Reviews (Methods Reviews) full-text systematic reviews on research methodological studies


5.   Cochrane Methodology Register (Methods Studies) - citations on publications on how to conduct controlled trials and systematic reviews


6.  Health Technology Assessment Database (Technology Assessments) structured abstracts of studies on the medical, social, ethical and economic implications of health care interventions. Produced by the CRD.


7.   NHS Economic Evaluation Database (Economic Evaluations) structured abstracts of critically appraised economic evaluations of health care interventions, produced by the CRD.


8.   About the Cochrane Collaboration (Cochrane Groups) description of Cochrane Review Groups, Centres, etc.

 - Updated quarterly.

Individual subscription available  


Type 2


A1 might mean - a systematic review> where all included studies are RCTs with good follow up studying outcomes of clinical importance.

B2 might mean - a good quality non-random study that meets only one additional criteria – for example, the outcome is of clinical importance.

C3 might mean - consensus or other expert based information – which meets no criteria for study design.

Type 3

Tips for retrieving EBM literature in Ovid MEDLINE, PubMed, and EMBASE:

· Use Clinical Trial publication types (pt) to limit your MEDLINE search to articles about actual clinical trials and to automatically explode all 7 clinical trial terms (below).

Clinical (1991)
Clinical Trial, Phase (1993)
Clinical Trial, Phase (1993)
Clinical Trial, Phase (1993)
Clinical Trial, Phase (1993)
Controlled Clinical (1995)

Randomized Controlled (1991)


Clinical Trials, the medical subject heading, retrieves articles on how to conduct clinical trials.

· Dont limit your search to Review (pt). Reserve the Review limit for “quick ‘n dirty searching.  -Ann Intern Med 1999;131:947-951.

· Core Clinical Journals, also known as AIM (Abridged Index Medicus), limits your search to 120 clinical journals commonly found in hospital libraries. Use this limit for "quick ‘n dirty searching only, and not EBHC searches.