Structure of the Evidence Summary

Owned by Kylie Porritt
Last updated: Nov 21, 2022 by Kim Dell
5 min read

Title

Evidence Summaries aim to follow a strict naming/title convention, and that should state the population, condition, or topic of interest, followed by the intervention as shown in the examples below.

  • Adhesive Capsulitis: Physical Therapy

  • Aggression (Older Adults): Risk Factors

  • Appendicitis: Surgical interventions

  • Axilla Burns: Post-Surgical Management

  • Carpal Tunnel Syndrome: Surgical Interventions

Where this is not possible, the title of the Evidence Summary needs to be clear and explicit, e.g. ‘Improving Symptom-related Communication skills for Patients with Cancer’

Date

The date the search was conducted should be detailed.

Author Details

The name(s) of the person(s) developing/updating the evidence summary and their qualifications should be listed.

Clinical Bottom Line

A short paragraph that introduces the topic and states its clinical importance, relevance and/or prevalence should be included to provide some context. The included studies and papers should be used to reference this section.

Example:

  • The experience of pain is complex, multi-dimensional and individualized for each person involving both physiological and psychological components.1 Wound care is noted as an intervention that is likely to cause pain, with dressing removal and application, wound cleansing and debridement reported as major contributors to pain experienced during wound care.1,2 Unresolved pain is noted to have a negative effect on both wound healing and quality of life; it is important to undertake an adequate assessment of pain, including determining the cause of the pain, in order to establish an effective plan of care to manage the pain experienced.1,3

Following this paragraph are bullet points summarising the ‘clinical bottom line’ or key findings from the selected studies/papers. Each bullet point of the ‘clinical bottom line’ should concisely describe the objective and key findings of the study. It is important that text is paraphrased (written in your own words), not simply copied verbatim from the paper or abstract.

When reporting the key findings:

  1. Only report the findings that are relevant to the topic i.e. those that are explicitly related to the clinical question.

  2. The clinical relevance of the findings should be reported. There is no need to provide complex statistics; however, statistics may be provided if they add context and meaning to the evidence point.

  3. Include some information on the clinical relevance of the results e.g. the conclusions/implications to practice.

  4. Cite the reference and include the JBI Level of Evidence in brackets. Only the major levels are applied in Evidence Summaries e.g. Level 1, Level 2, etc. (sub-levels not required)

An example of how a bullet point in the ‘Clinical Bottom Line’ section should be constructed is as follows:

  • A systematic review investigated the effects of asthma education on health outcomes in children who presented to the emergency department (ED) for the treatment of asthma. Asthma education provided to children and/or parents resulted in a 27% lower risk of future ED presentation and 21% lower risk of hospital admission. However, the long-term effect of asthma education on quality of life, symptoms, and lung function is still unclear. Details of education content and method of delivery need to be further researched.1 (Level 1)

Please note:

  • Evidence Summaries only include the current best available evidence, so when updating an Evidence Summary with any older evidence that has been superseded by new evidence (unless it is still relevant and of high quality), it should be removed.

  • If a new systematic review is identified in the search, check if other existing references in the Evidence Summary are included studies in the systematic review. Primary studies included in a systematic review should not be included separately in the Evidence Summary.

  • Where systematic reviews or clinical guidelines have been updated, the older review/guideline should be replaced with the new version, and the reference updated.

Characteristics of the Evidence

Characteristics of the evidence should outline the types of evidence included in the summary with the evidence sources listed in dot point with a citation. Details on the study design, the number and type of studies included (for systematic reviews), and the number of participants involved should be described within this section.

Standardized text is used at the beginning of this section; "This evidence summary is based on a structured search of the literature and selected evidence-based healthcare databases. The evidence included in this summary is from:"

An example of the information to include in this section is as follows:

  • A systematic review that included 38 studies (randomized controlled trials [RCTs] and quasi RCTs) with a total of 7,843 children (aged 5 months to 20 years).1

  • A cross-sectional descriptive study involving 1,221 critical care and ward nurses from an urban pediatric hospital.2

  • Consensus-based guidelines and standards based on research evidence and expert opinion.3

Best Practice Recommendations

Best practice recommendations are directive statements that should operationalize or guide practice using the evidence in the clinical bottom line (these should be congruent). Each recommendation is rated using the JBI Grades of Recommendations. 

There are two levels of recommendations:

  • Grade A – a ‘strong’ recommendation

  • Grade B – a ‘weak’ recommendation.

If there is a scarcity of evidence, then clinical judgement is recommended.

An example of a Best Practice Recommendation is as follows:

  • Immunization against measles is recommended for all susceptible children and adults for whom measles vaccination is not contraindicated. Contraindications include high fever or other signs of serious disease, pregnancy, history of anaphylactic reaction to vaccine components, or a severely compromised immune system. (Grade A)

NOTE FOR UPDATING RESOURCES:

  • When updating an evidence summary, if new evidence is added and/or old evidence deleted, the ‘Best Practice Recommendations’ must be reviewed. Depending on the evidence, it may be necessary to:

    • add a new recommendation, or

    • remove/amend an old one, or

    • modify the Grade of an existing recommendation.

  • Having looked at the ‘Best Practice Recommendations’ section, decide if any new evidence added to the ‘Clinical Bottom Line’ section warrants an alteration to any of the recommendations or the addition of a new recommendation. The inclusion of new evidence does not necessarily mean that the recommendations need to change.

  • If adding a new recommendation, it should be worded as a specific recommendation that is actionable, with words like: “should”, “may”, “use”, “is recommended”, etc.

  • A JBI Grade of Recommendation is then assigned, Grade A for a ‘strong’ recommendation or Grade B for a ‘weak’ recommendation.

Referencing

References are to appear in numerical order and are detailed in the reference list using Vancouver format. For example:
John EA, Smith J. XXXXXXXXXX. J Nur. 2008;74(2):304-10.

Points to consider:

  • With Vancouver style referencing, articles are assigned a number in terms of when they first appear in the text.

  • When citing studies in the text, superscript all reference numbers.

  • If there are more than six authors, then list the first six followed by et al.

  • Use sentence case for the article title.

  • Abbreviate the journal name; you can find recommended abbreviations via https://www.ncbi.nlm.nih.gov/nlmcatalog/journals. 

  • The month/day of publication should not be included, only the year, volume number, issue number, and page range.

See the Vancouver referencing guide for further details.

Technical Development Report 

For each Evidence Summary, a Technical Development Report is completed.

This report should record and document:

  • The date range of the search (i.e. last five years)

  • The names of the databases searched

  • All search terms used

  • Appraisal results for new evidence (any new evidence found is assessed for methodological quality using a short, standardised checklist, except for Level 5 evidence which does not require appraisal)


Please also refer to our Frequently Asked Questions Page.