NUR 67301: Health Policy Residency for DNPs: Levels of Evidence

Evidence Levels, or Hierarchies, are systems used to rank evidence according to certain criteria.

Tips:

• Especially in the health sciences, research done in vitro (in a petri dish) and in animals may not actually hold when repeated with people.
• Research with a control group (no intervention) and an experimental group is more reliable than research without. 
• Research that is blinded—where the researchers don't know who is in what group—is better able to prove an effect than unblinded studies.
• Science experiments should be repeatable. 
• Because statistical flukes can happen, meta-analyses and systematic reviews from respected groups like Cochrane, which take a structured look at the existing research about a topic, are best able to reach a conclusion on how complete or established a scientific position is.
• Responses matter. Scientific publishing is a conversation, papers cite each other to agree or disagree. Collecting articles that refer to each other gives you a better picture of your topic. 

• In many databases the reference list includes links. You may need to search the title for full text access. However the publishers page, which the DOI will lead to, will have the abstract so you can see what the article was about.

What's Not in the Pyramid of Evidence?

There are some types of sources you may encounter that aren't on the pyramid of evidence. These sources may be useful for your understanding, but should not be cited as scientific evidence. Examples of these include: Blogs, magazine or new articles, and popular non-academic sites such as Wikipedia.

• These types of sources do not reflect primary research, nor are they an expert review of primary research the way clinical practice guidelines or systematic reviews are.
• They may refer to or summarize research articles. Locate the research article rather than cite a journalist's summary of it. Wikipedia will include linked citations to the original sources. 

Content reproduced with permission from Grand Canyon University Library: https://libguides.gcu.edu/EvaluatingSources

JBI Levels of Evidence

Melnyk's Levels of Evidence

Level 1: Systematic review & meta-analysis of randomized controlled trials; clinical guidelines based on systematic reviews or meta-analyses

Level 2: One or more randomized controlled trials

Level 3: Controlled trial (no randomization)

Level 4: Case-control or cohort study

Level 5: Systematic review of descriptive & qualitative studies

Level 6: Single descriptive or qualitative study

Level 7: Expert opinion

Modified from: Melnyk, B.M. & Fineout-Overholt, E. (2015). "Box 1.3: Rating system for the hierarchy of evidence for intervention/treatment questions" in Evidence-based practice in nursing & healthcare: A guide to best practice (3rd ed.) (pp. 11). Philadelphia, PA: Wolters Kluwer Health.

Therapy/Intervention

Systematic Reviews or Meta-Analysis
Randomized Controlled Trials
Clinical Practice Guidelines
Nonrandomized Controlled Trials
Cohort Study or Case-Control Studies
Meta-Synthesis of Qualitative or Descriptive Studies
Qualitative or Descriptive Studies
Expert Opinion

Diagnosis, Prognosis/Prediction

Synthesis of Cohort Study or Case-Control Studies
Single Cohort Study or Case-Control Studies
Meta-Synthesis of Qualitative or Descriptive Studies
Single Qualitative or Descriptive Studies
Expert Opinion

Meaning

Meta-synthesis of Qualitative Studies
Cohort Studies
Single Qualitative studies
Synthesis of Descriptive Studies
Single Descriptive Studies
Expert Opinion

Prevention

Randomized Controlled Trials
Cohort Studies
Case Control
Case Series

Economic Analysis

Economic Analysis