Online Guide to EBM: 
Solving a clinical problem related to Therapy.
Notes on using the lecture

1) Ask The Question
2) Finding The Evidence: choosing a source
3) Finding The Evidence: constructing a search strategy (OVID)
4) Reading the Study: Is it Valid?
5) What are the Results?
6) Will the results help me care for my patient?
Follow along with an example patient, search, and analysis
 

 
1) Ask The Question
Follow along with an example patient, search, and analysis

Start by framing a concise question.  This is essential to finding information.  The question can be divided into 4 parts, depending on your needs.

1) What is tthe disease or condition of interest?
2) What is the intervention (treatment) of interest?
3) What is the comparison intervention of interes (if any)?
4) What is the outcome of interest? 
 
2) Finding The Evidence: choosing a source
Link to Primary Sources PageSearch OvidSearch Grateful MedSearch PubMedLink to Secondary Sources pageTurning Research Into Progress: The TRIP DatabaseGlobal Emergency Medicine Archives: GEMANational Guideline ClearingHouse

Secondary information sources tend to be strong on articles concerning therapy.  As usual, TRIP is a good place to look for critically apraised materials.. Clinical guidelines frequently provide concise therapy recommendations.
3) Finding The Evidence: constructing a search strategy (OVID)
Searching 101MeSH BrowserOnline Statistical Text'sClinical Calculator linksFollow along with an example patient, search, and analysis

If you have not already done so, this is a good point  to stop and read more about literature searching.

We suggest the following initial search strategy:

  • Enter a search filter   for therapy 
        Randomized controlled trial.pt  or drug therapy.sh  or random.tw
  • Search the Disease
  • Search the intervention of interest
  • If needed, search the comparison intervention
  • Combine these (Boolean AND): 1 AND 2 AND 3  
If you didn't find what you want, try the following search tips 
4) Reading the study: is it Valid?
Searching 101MeSH BrowserOnline Statistical Text'sClinical Calculator linksFollow along with an example patient, search, and analysis
  • Was the assignment of patients to treatment randomized? Randomization (assuming adequate sample size) ensures that both known and unknown determinants of outcome are equally distributed between treatment and control groups. 
  • Were all patients entered into the trial properly accounted for and attributed at its conclusion? 
    • Was follow-up complete?  Validity of study is questionable if substantial numbers of patients are lost to follow-up. 
    • Were patients analyzed in the groups to which they were randomized (Intention-to-Treat Analysis)?
  • Were the groups similar at the start of the trial? All factors known to influence outcome should be equally distributed between the control and treatment groups.  Hopefully, randomization will equally distribute unknown prognostic factors. 
  • Were patients, their doctors, and study personnel “blind” to treatment? 
  • Were the groups treated equally, except for the intervention?  The control and treatment groups should receive identical care excepting the studied intervention. “Co-interventions” administered to one group but not the other in equal frequency can dramatically alter outcomes.

    •  
5) What are the results ?
Searching 101MeSH BrowserOnline Statistical Text'sClinical Calculator linksFollow along with an example patient, search, and analysis
  • How large was the treatment effect? Make sure you understand some important concepts
    • What was the  Absolute Risk Reduction (ARR) - difference between the proportion who have the outcome in the control group minus that proportion in the treatment group.
    • Relative Risk (RR) - risk of events among patients in the treatment group relative to the controls.
    • Relative risk reduction (RRR) - complement of relative risk expressed as a percent.  (1 - RR) x 100%
    • What was the number needed to treat (NNT). If not provided calculate this value
  • How precise was the estimate of treatment effect?   A study provides a “point estimate” of the true risk reductions provided by a therapy. The “true value” is somewhere in the neighborhood of the confidence interval (CI). 

    •  
6) Will the results help me care for my patients ?
Searching 101MeSH BrowserOnline Statistical Text'sClinical Calculator linksFollow along with an example patient, search, and analysis
  • Can the results be applied to my patient care? 
    • Would your patient have been eligible to be enrolled in the study? 
    • Is there a compelling reason the study results should not be applied to your patient. 
    • Does your patient fit into an analyzed subgroup?
  • Were all clinically important outcomes considered?
    • Were physiologic outcomes rather than outcomes important to the patient (death, hospitalization, quality of life) measured? 
    • Were all outcomes, including “costs” of  therapy measured?
  • Are the likely treatment benefits worth the potential harm and costs?  The critical assessment you have performed will help you make an educated determination for your patient. 
Back to the top