Skip to main content

Evidence Based Medicine Guide: Evaluate Search Results

Scan Titles and Abstracts

Rapid Triage: Once you have conducted your literature search, a quick scan of titles and abstracts will enable you to pick out the papers most likely to provide the best evidence to answer your question. 

  1. Look for abstracts about studies that are relevant and generalizable to your patient: those that come closest to addressing all your PICO elements, including the criteria you didn't incorporate as search terms. In particular, note whether the outcomes measured are clinically relevant.
     
  2. Look for indicators of validity: appropriate study designs that minimize bias. Note key methodological criteria such as the use of randomization and blinding, and the length of time participants were followed.
     
  3. Look for indications of a study's significance, such as measures risk reduction and confidence intervals. Larger studies that had more complete follow up are more likely to provide precise, statistically significant findings.

Ensure You Have the Current Evidence

Systematic reviews, meta-analyses and practice guidelines can go out of date quickly, so make sure that you are working with the most current evidence.

Check "Related Citations"

Click the "Related Citations" link for the record: frequently you'll find newer versions of reviews or guidelines very near the original article, at the top of the result list. Be aware that the Related Citations search is not filtered for quality and may not retrieve articles that report the "best evidence". 

The results  of a Related Citations search are ranked by similarity to the original article, not by date. To find the most current articles, use the Display Settings menu to change the sort order of your results. From the "Sort by" menu select "Pub Date" (Publication Date) and click the Apply button.  Not sure how to change the settings?  See the NLM's page on using Display Settings in their PubMed tutorial.

Check for recent primary studies

Even if you find a current systematic review that addresses your question, it is prudent to review the results of Clinical Queries' Clinical Study Categories filter for recent, high quality primary studies.

Remember: Avoid using the Article Type filters to identify the most recent clinical studies. Most Article Types filter out articles that have not yet been indexed for MEDLINE.

ACP's Selection Criteria for Articles

The American College of Physicians has established criteria for selecting studies for review in the ACP Journal Club. Their standards for the types of questions considered in this tutorial are listed below; the original is available here (1). 

We suggest using the ACP criteria as a guide when you review citations retrieved in a search for the best evidence. They can indicate whether a study is likely to offer evidence strong enough to inform a clinical decision — and whether it is worth going to the next step of the EBM process: reading and critically appraising the original article.

Question Type

ACP's Criteria for Primary Studies

Treatment & Prevention

  • Random allocation of participants to comparison groups
  • Follow-up (end-point assessment) of at least 80% of those entering the investigation
  • Outcome measure of known or probable clinical importance

Diagnosis

  • Inclusion of a spectrum of participants, some but not all of whom have the disorder or derangement of interest
  • Objective diagnostic ("gold") standard OR current clinical standard for diagnosis, preferably with documentation of reproducible criteria for subjectively interpreted diagnostic standard (i.e., report of statistically significant measure of agreement beyond chance among observers)
  • Each participant must receive both the new test and some form of the diagnostic standard
  • Interpretation of diagnostic standard without knowledge of test result
  • Interpretation of test without knowledge of diagnostic standard result

Prognosis
(Natural History)

  • Inception cohort of individuals, all initially free of the outcome of interest
  • Follow-up of at least 80% of patients until the occurrence of a major study end point or to the end of the study

Etiology or Harm

  • Exploration of the relationship between exposures and putative clinical outcomes
  • Prospective data collection with clearly identified comparison groups for those at risk of the outcome of interest (in descending order of preference from randomized controlled trial, quasi-randomized controlled trial, nonrandomized controlled trial, cohort studies with case-by-case matching or statistical adjustment to create comparable groups, to nested case-control studies)
  • Masking of observers of outcomes to exposures (criterion assumed to be met if outcome is objective, i.e., all-cause mortality, objective test)

PubMed Tip: Displaying Abstracts

To review titles and abstracts of a set of articles in PubMed, click the Display Settings menu (just under the search box), select Abstract and then click Apply.

Set Abstract View