Skip to Main Content



Evidence based practice is a way to help nurses make decisions in clinical care guided by the best possible evidence. There are five stages in evidence based practice:

Click through the tabs on the box below to find out more about each stage.

Evidence based practice stages

The first step of evidence based practice is to form a clear clinical question. This will help you find the information you need.

PICO is a framework that can help you turn your question into search terms that will give useful results. 

Not all questions will use every element, but it helps to be as specific as possible to help evaluate which results are relevant.

There are variations on PICO which you may need to use, depending on your question.

PICOT: Timeframe

PICOS: Study type

PICOC: Context

This stage is where you search for information to answer your question. You can use your PICO elements to plan how to search.

Levels of evidence

When answering clinical questions, you need to use the highest level of evidence possible. In the diagram below, evidence that is higher up the pyramid is of higher quality i.e., systematic reviews (secondary research).

Secondary research appraises the quality of original research studies, summarises the findings, and often makes recommendations for practice. However, secondary research is not always available, and searching for primary research such as randomised control trials may be required.

Secondary research Primary research
  • Review, evaluate, and discuss other peoples' original research and findings
  • The 'methods' or 'methodology' section describes how the authors searched databases and decided which studies to include in their review
  • The 'results' and 'discussion' sections may include a table of the characteristics of the studies they included, and describe the themes of the studies.
  • Reports on the authors' own original research that they have conducted themselves
  • The 'methods' or 'methodology' section describes how the research was carried out, such as recruitment of participants, experiments, tests, interviews, surveys, and other data collection.
  • The 'results' or 'findings' section describes what the researchers found out, and may include statistics, graphs, or tables to show their data.
Secondary study-type examples Primary study-type examples
  • literature review
  • systematic review
  • meta-analysis (quantitative) 
  • meta-synthesis (qualitative)
  • guidelines
  • evidence-based care sheets
  • expert opinions
  • editorials
  • surveys 
  • interviews
  • focus groups
  • experiments
  • randomised control trials
  • cohort studies
  • case studies
  • qualitative studies


Use the following checklists to help decide if an article is primary or secondary research.

In secondary research, you will probably see:

  • The word 'review', 'meta-analysis', or 'meta-synthesis' in the title
  • A description of how the researchers found articles e.g., search strategy, and databases searched
  • A flow chart showing how the articles were chosen
  • How many articles were reviewed
  • A description of their characteristics or themes

In primary research, you will probably see:

  • Methodology section describing:
    • the study participants, e.g., how many there were, and how they were chosen
    • What tests, experiments, surveys, or interviews were done
  • Results section:
    • Charts or tables of the data found
    • Quotes from participants

Systematic reviews

A systematic review is regarded as a high level of evidence and is secondary research. It starts with a specific and answerable clinical question, identifies all the research available and eliminates flawed and low-quality studies, then uses the results of the remaining high-quality studies to answer the clinical question and thus make recommendations on best practices.

Randomised controlled trials

Randomised controlled trials are primary research. Researchers assign participants randomly into an intervention group or a control group, then results are compared to see if the intervention had an effect.

Cohort studies

A cohort study is primary research. It looks at the health outcomes of a specified group, comparing people in the group who have been exposed to a potential risk factor with those who have not been exposed, and looks for links and correlations.

Case control studies

Case control studies looks at existing data to compare people who have a specific health outcome or condition, compared with those who don't, to find what factors differ between the groups.

Case series and case reports

Case series and case reports look at one or more individual cases and outcomes or responses to interventions. Regarded as a low level of evidence, the information might not apply to a wider population.

Editorials and expert opinions

Editorials and expert opinions can be based on a practitioner's individual experience. Generally regarded as one of the lowest levels of acceptable evidence, they do not necessarily account for uncontrolled external factors, differences between different groups, nor placebo effects.

Systematic reviews Randomised control trials Cohort studies Case control studies Case series and case reports Editorials and expert opinions


Hover over each level of the evidence pyramid for more detail


To learn more about levels of evidence, click on the links below:

JBI Levels of Evidence 

JBI Grades of Recommendation 

Use the arrows at the right and left of the box below to click through and see the characteristics of a number of different study types from the levels of evidence pyramid.

Plan your search

Use your PICO elements to create search terms. You might not include every element in your search, for example an outcome can sometimes be difficult to define.

The next step is to consider alternative terms that authors might use, for example synonyms, or medical terms. A table can be a useful way of recording these. 

Population Intervention Comparison Outcome
Adults over 60 30 minutes of daily exercise No exercise Reduce heart attack risk




Physical activity

Aerobic training



Cardiac arrest

Myocardial infarction

Coronary event

Now you can use Boolean operators to group the terms together, and use search tips such as phrase searching and truncation:

(elderly OR aged OR geriatric) AND ("physical activit*" OR "aerobic training" OR "exercise*") AND ("cardiac arrest*" OR "heart attack*" OR "myocardial infarction*")

Where to search

The library has many databases you can search in, depending on what level of evidence you need. Some of these databases allow you to filter results according to evidence type.

  • Secondary research:
    • Trip Database: Useful for finding evidence based material, including clinical guideline, systematic reviews, and evidence based synopses
    • The Cochrane Database of Systematic Reviews: Detailed, structured reviews of articles written by teams of experts who evaluate the literature and summarise the findings
    • Joanna Briggs EBP Database: Contains evidence summaries and systematic reviews, as well as best practice recommendations
  • Primary research:
    • CINAHL: Useful for finding primary, peer reviewed studies. There are extensive filters that allow you to select which study type you are interested in.
    • MEDLINE: Also useful for finding primary research, and using the Clinical Queries filter allows you to select from study categories such as therapy, diagnosis, etiology, and prognosis, and indicate whether you want a more sensitive or specific search

When you have found the articles and information, you will need to evaluate it for validity, and risk of bias. How you do this will depend on whether you are appraising primary or secondary research, and the study type, but there are common elements to consider. The CRAAP test is a good way to evaluate articles. A lot of what you need to appraise articles will be in the Methods section. Keep in mind evaluating articles is a skill that can take a while to learn, so if you need guidance, contact the library.

  • When was the article published? Generally in nursing, you should be using articles less than 7 years old.
  • When was the data collected? Is there a delay between the data collection and publication, and if so does the article explain why?
  • This depends on what you are trying to find out, but the article should be peer reviewed. Looking at the title and abstract should give you an indication of what the article is about.
  • What are the authors' qualifications and are they relevant to the topic? This should be listed somewhere in the article. If it is not, you can try and look them up.
  • Who do the authors' work for? It should state this as well. 
  • Is the authors' workplace likely to lead for bias? For example, if an author is testing a product made by the company they work for, they may be at risk of being biased and not accurately reporting results.
  • What is the study population size and characteristics? Is there a control group? How were they recruited? Keep in mind this is dependent on the study type, for example qualitative studies usually have smaller study populations than quantitative
  • Do the statistics make sense and match the authors' claims?
  • Who funded the study? Is it a company, university or research organisation? 
  • Is there any obvious bias where the authors or their employers are likely to benefit from the study recommendations? For example, if the authors work for a particular company and recommend the use of one of their products, this could indicate a risk of bias.
  • Does it state what the authors' were trying to find out?

If you are interested in finding out more, the following article is old, but has some great questions to ask yourself when reading primary research, and is easy to read.

Greenhalgh, T. (1997). How to read a paper: Assessing the methodological quality of published papers. BMJ315(7103), 305–349.

This is another older paper by the same author that describes in more detail how to appraise secondary research.

Greenhalgh, T. (1997). How to read a paper: Papers that summarise other papers (systematic reviews and meta-analyses). BMJ315(7109), 672–675.

The article below has been annotated using the CRAAP test from above. Scroll through and hover the mouse over the highlights to see the types of things to consider when evaluating an article.


There are a number of tools and checklists researchers use to appraise studies that you can look at for a more detailed way of evaluation.. 


Riegelman, R. K. (2005). Studying a study and testing a test: How to read the medical evidence (5th ed.).

Appraising qualitative research articles:

Côté, L. (2005). Appraising qualitative research articles in medicine and medical education.

Critical appraisal of Indigenous research:

Harfield, S., Pearson, O., Morey, K., Kite, E., Canuto, K., Glover, K., Gomersall, J. S., Carter, D., Davy, C., Aromataris, E., & Braunack-Mayer, A. (2020). Assessing the quality of health research from an Indigenous perspective: the Aboriginal and Torres Strait Islander quality appraisal tool. BMC Medical Research Methodology, 20(1), 1-9.

Lock, M. J., Walker, T., & Browne, J. (2021). Promoting cultural rigour through critical appraisal tools in First Nations peoples’ research. Australian and New Zealand Journal of Public Health, 45(3), 210–211.



In this stage, you decide if the evidence you have collected is suitable for your clinical situation.

You also need to consider the other two areas of evidence based practice: clinical expertise and patient values.

  • Clinical expertise: Is the intervention suitable for your setting? Will the benefits outweigh harm?
  • Patient values: What are the patient's circumstances, views, beliefs, and values? Are they concerned or worried about anything? What is their preference for treatment?

Sometimes you are not certain of the right thing to do even after looking at the evidence. To help you decide, you can:

  • Ask a more experienced colleague
  • Take a quick break to reflect
  • Consider the standards, codes of practice and legal requirements

The final step is to reflect on your practice and consider strengths, weakness and where you need to improve.

Consider how you applied the evidence, whether what you implemented was effective, and was the outcome satisfactory for the patient.

Evidence-Based Medicine Information Sites

Centre for Evidence-Based Medicine (Oxford University)

The Centre promotes evidence-based health care and provide support and resources to anyone who wants to make use of them. Includes the EBM Toolbox, an assortment of materials which are very useful for practitioners of EBM, and EBM Teaching Materials, including PowerPoint presentations.

Centre for Evidence-Based Medicine (Canada based)

Includes many resources for practicing and teaching EBM

Alberta Research Centre for Health Evidence

Contains links to recent publications published by the ARCHE team.

Introduction to Evidence-Based Medicine

From Duke University Medical Center Library and University of North Carolina at Chapel Hill Health Sciences Library.

Levels of evidence

Levels of evidence and evidence grading

Levels of evidence and evidence grading can be used to indicate the certainty of evidence - if the evidence is strong and can be used to support or change practice, or if the evidence is weak and should be used with caution.

Useful documents for grading evidence quality

Joanna Briggs Institute. (2013). JBI levels of evidence. JBI.

More explanation:

Joanna Briggs Institute Levels of Evidence and Grades of Recommendation Working Party. (2014). Supporting document for the Joanna Briggs Institute levels of evidence and grades of recommendation. JBI.

Nice tabular summary of the JBI levels of evidence:

JBI, & University of Adelaide. (2021). JBI EBP database guide. Wolters Kluwer.

National Health and Medical Research Council. (2019). NHMRC levels of evidence and grades for recommendations for developers of guidelines. NHMRC: Building a healthy Australia. 

National Institute of Clinical Studies. (2009). Appendix F: Levels of evidence and recommendation grading. In Emergency Department stroke and transient ischaemic attack care bundle: Information and implementation package. NHMRC.