Evidence-based practice

Evidence-based practice (EBP) integrates published research, professional expertise, and specific relevant circumstances to improve the quality and efficiency of professional practice

Advantages of EBP include:

• increased consistency of professional practice
• better quality decision making
• reduced costs
• improved efficiency
• enhanced end-user outcomes

A well-structured question can help you identify which databases to search, structure your search, and select the most reliable levels of evidence to answer your question.

Types of questions

Determining your question type will guide how you:

a) interpret and structure your question using a mnemonic

b) identify which level of evidence or study design(s) are most appropriate.

There are 4 common question types used in medicine and health (which may not be relevant to other fields). These are also referred to as clinical questions.

They are:

1. Therapy:  the most beneficial treatment or prevention for a condition (drug therapy, surgery, physical therapy, etc.).
2. Aetiology:  the relationship between an exposure or causal factor and a patient developing a condition.
3. Diagnosis:  the validity and reliability of a diagnostic test or screening method.
4. Prognosis: the likely disease progression for patients with or without treatment.

Mnemonic tools

Mnemonic tools can help you structure your question in a meaningful way. 

This mnemonic guide outlines a range of mnemonics designed for different question types and purposes. It provides frameworks, along with example questions.

PICO 

PICO (Population, Intervention, Comparator, Outcome) is a common mnemonic used in health and medicine that can also be applied to other disciplines. 

This video explains how to use PICO to break down a research question.

Study designs refer to the way to the way research is undertaken, including collecting evidence. 

“An evidence hierarchy classifies evidence according to its rigour. ‘Gold standard’ evidence sits at the top. Less rigorous forms of evidence sit below it, in descending order.” (Source: Centre for Education Statistics and Evaluation, NSW Government). 

Published research evidence is presented in 2 forms:

• primary study design presenting outcomes from formal original research methodologies

• secondary study design which synthesises the results of multiple primary studies

You can find more information about level of evidence according to type of research question in NHMRC Evidence Hierarchy.

Study designs refer to the way to the way research is undertaken, including collecting evidence. 

“An evidence hierarchy classifies evidence according to its rigour. ‘Gold standard’ evidence sits at the top. Less rigorous forms of evidence sit below it, in descending order.” (Source: Centre for Education Statistics and Evaluation, NSW Government). 

Published research evidence is presented in 2 forms:

• primary study design presenting outcomes from formal original research methodologies 
• secondary study design which synthesises the results of multiple primary studies

You can find more information about level of evidence according to type of research question in NHMRC Evidence Hierarchy

Study designs in medicine and health

When answering a clinical question or treating a patient, it’s important to find the best level of evidence. 

This video explains three different levels of evidence hierarchies and their uses:  

1. The Evidence-Based Medicine Pyramid
2. The NHMRC Evidence Hierarchy.
3. The Oxford Centre for Evidence-Based Medicine's Level of Evidence Table.

For most types of clinical questions, a well-done systematic review or meta-analysis of all available studies is the best level of evidence. A comprehensive systematic review will compare multiple randomised controlled trials that have studied the same treatment and aggregate their results.  

When a well-done systematic review or meta-analysis is not available, you may need to consult primary study designs. The list below shows which primary study types are the most appropriate to answer different types of clinicals. 

• Therapy: randomised controlled trials (RCTs).
• Diagnosis: independent studies > blind comparison to a gold standard.
• Prognosis: cohort studies > case control studies > case series. 
• Aetiology: RCTs > cohort studies > case control studies > case series.
• Prevention: RCTs > cohort studies > case control studies.
• Cost-effectiveness: economic analysess
• Quality of life: qualitative studies. 

For more information about study designs, see 'How to read a paper' by Trish Greenhalgh.

A search strategy is essential when searching databases for evidence-based literature to answer your question. Your search strategy should include both keywords and subject headings, as well as alternative search terms and advanced search syntax.

See this step-by-step guide for help developing your search strategy and searching databases.

Searching specific databases

Databases use different syntax and subject headings, so you'll need to adapt your strategy to fit each database you plan to search. 

To build comprehensive searches in various databases, refer to these quick search guides.

Appraising the research you’ve found will help identify the most reliable evidence.

Appraisal tools  are available for some disciplines. These tools are used to critically evaluate an article by checking for:

• strengths and weaknesses of the article’s study
• design and methodology
• relevance to your question
• importance of the research question being answered
• legitimacy of conclusions
• potential conflicts of interest