VETS5016 - Evidence-based veterinary medicine and systematic review: Get started

A systematic review is an overview of primary research (studies written by the originators) performed in a rigorous and methodical way. The approach aims to be transparent, reproducible and unbiased.

It may stand-alone to contribute to the body of knowledge; or it may be the first step in a research project and may subsequently help shape a research plan.

A well-conducted review will:

• determine what's already known about a topic
• appraise the quality of what's known
• summarise the evidence of the best studies
• identify and resolve conflicting evidence
• identify gaps or reveal new areas to investigate
• avoid unnecessary duplication of research

A meta-analysis is a statistical analysis of the results of two or more primary research studies, which address the same clinical question.  Mathematical values are given to the results so they can be statistically analysed.  This allows the effectiveness of a treatment to be quantified to make it easier to compare and distinguish differences between studies. 

Systematic reviews and meta-analyses can be the strongest and most reliable form of evidence. (Higgins JPT, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, Welch VA (editors). Cochrane Handbook for Systematic Reviews of Interventions version 6.3 (updated February 2022). Cochrane, 2022. Available from www.training.cochrane.org/handbook.)

 Contact the UQ Library team via

librarians@library.uq.edu.au

1.  Develop a systematic review protocol

• this is an outline of the approach to be taken
• it's done before the review process begins, in order to eliminate bias and promote transparency
• it should include:
  - the background, justification and purpose
  - detail the clinical question (in the PICO format)
  - literature search method, the databases, keywords and their relevance
  - the criteria for screening and inclusion of the articles
  - methods for data extraction, assessment of bias and analysis

2.  Define an answerable clinical question

• define your clinical question clearly in the PICO format
• examine an intervention:  this may be a treatment, therapy,  procedure, pharmaceutical, diagnostic test, preventative measure, or management plan.

3.  Search for evidence

• identify appropriate sources for potential studies, depending on topic
• perform an initial screening of the results to identify relevant papers
• critically examine the full-text of those relevant papers
• save the details of the search strategies and incorporate the details into a flow diagram
• reference management software such as EndNote is to be used with collecting and collating the references.

4.  Evaluate the evidence

• using a standardised form for each, pull out the key data including:
    - reference details, objective, study design, population, intervention protocol, statistical method, outcome, comments
• each trial should be critically evaluated using a critical appraisal checklist to assess for:
    - methodological quality - avoidance of systematic errors (bias)
    - precision - likelihood of random errors
    - external validity - transferability to the population in the PICO
• summarise the findings in tables.

5.  Interpret and discuss findings

• present the evidence as it translates to practice
• summarise the strengths and weaknesses of the studies
• make conclusions on the evidence
• identify gaps and other areas to investigate.

Before formulating the research question, you could do a quick search for other systematic reviews on your topic.

PubMed Clinical Queries is a tool in the PubMed database that filters a search to only bring back systematic reviews and clinical trials.

Enter a couple of keywords on your topic to see what aspects are already covered.

Note: This feature is only suitable as a quick, preliminary search. The filter can mistakenly eliminate suitable studies. You need to use the main database when doing your full search strategy.

The CEVM, based at the University of Nottingham has relevant information and links on all aspects of evidence-based medicine, including systematic reviews.