Template for Reporting Practice Guidelines in CMI

Template for reporting practice guidelines in CMI

Title page:

Title, authors, affiliations, ESCMID study group/s (if relevant)

Abstract (of no more than 400 words):

Scope

Methods

Questions addressed by the guideline and Recommendations

Full document (of no more than 4500 words;).

Scope

The background should define the focus of the guideline (in broad terms it can be a group of patients (e.g. solid organ transplant patients); an infection/ syndrome (e.g. community acquired pneumonia, febrile neutropenia); a pathogen (e.g. Staphylococcus aureus), an intervention (e.g. use of colistin); or a diagnostic procedure (e.g. use of PET-CT in diagnosis of infection). The goal of the guidelines should be explained here, the populations and sub-populations of interest as well as the intended users and the settings for which the guideline is relevant.

Context

The context should provide clinicians with an evidence-based up-to-date summary of the topic. Provide here the general background information of the topic that is not covered by the recommendations. Address the prevalence or incidence of the disease, the burden of disease and its forms (either severity or other variations); the mechanisms of the disease; disease epidemiology in different settings or locations, addressing (when relevant) differences in antibiotic resistance, diagnosis, management and outcomes. For pathogen-focused guidelines, spectrum of disease associated with the pathogen, resistance prevalence in different settings and diagnosis. For intervention guidelines describe the intervention, if an antibiotic its spectrum of activity, PK/PD parameters, availability and usage worldwide; and for diagnostic procedures the diagnostic modality, test performance characteristics, availability etc. For both intervention and diagnostic procedure guidelines, costs and alternatives should be provided.

The context should also be used to justify the decisions that were taken in formulating the foreground questions. For example, if the population of interest in the foreground question is divided into subpopulations, the background should make clear why this division is needed. The context should justify the classification of guidelines by settings, locale or other based on differences in epidemiology, resistance and feasibility of interventions/ diagnostic procedures. The hierarchy of outcomes considered when selecting recommendations should be highlighted and justified.

The context can be used to mention other guidelines that were published on the subject and why is the present guideline needed in light of the previous publications. If there is a special European aspect to the topic addressed by the guidelines, this is the place to mention it.

Questions addressed by the guideline

Formulate the foreground questions that are addressed by the guideline.The question should be in the PICO format, as relevant:defining population, intervention, comparator, outcome. The outcomes of choice are those that matter to patients.

Methods

Address the following topics:

• Evidence collection: describe the search strategy, the databases used to search for publications, date of the last search, filters that were used.
• Synthesis of data: describe the quantitative or qualitative (meta-analysis) methods that were used for synthesis of data.
• The process that was used to reach the recommendations and write the final document.
• Grading of evidence: CMI and ESCMID support the use of the GRADE system for quality appraisal. Describe any deviations from the GRADE criteria or specific considerations related to evidence GRADing. Define the risk-of-bias tool used per study design (tools available for randomized controlled trials, observational studies, diagnostic studies). Describe the process of GRADing (performed by whom, whether in duplicate, whether a GRADE expert was included in the panel or consulted).
• Review: describe the review process that was used before releasing the guidelines.
• Definition of terms and acronyms.

Recommendations

Address each foreground question under a separate subheading. Under each subheading address the following:

• A detailed question.
• Methods that were used to gather the evidence (search strategy, inclusion and exclusion criteria for studies, methods of data extraction).
• The evidence and the synthesis of evidence. Address potential benefits and harms using absolute and relative measures. Emphasize the main outcome on which recommendations are based (that should be the one that matters most to patients).
• Recommendation
• Grade the quality of the evidence and the strength of recommendation. See the uniform grading structure for ESCMID guidelines in Appendix A. In text the types of studies, quality of the evidence and strength of recommendation should be provided per recommendation.
• Patients' preferences: In instances in which weighting the pros and contras for an intervention or procedure depends on personal preferences, discuss the possible implications of personal preferences.
• Other implications: consider the consequences of implementing the guidelines in the different relevant settings; barriers for implementation and how they can be addressed, including cost considerations; temporal changes that might affect implementation and should be monitored (e.g. changes in resistance to antimicrobials); suggest criteria and process for measuring the impact of implementation.

Summarizing recommendations in tables is encouraged, detailing the question, relevant population/ subgroup, recommendations, its strength and quality of the evidence.

Description of the developing group

Describe the contribution of each member of the panel, and if relevant of ESCMID committees and sub-committees.

Conflict of interest

Describe conflicts of interest for each member of the panel using the items included in the International Committee of Medical Journal Editors Conflict of Interest Disclosure.

Funding sources

Acknowledgments

Including the services of a medical writer if used; external reviewers; etc.

Updating

Describe plans for updating the guidelines: who is the person in charge of updating; how often will upgrading be considered; which criteria will serve to decide whether a major, minor or no updating is necessary.

Appendix A: quality of evidence and type of recommendation

Quality (certainty) of evidence:

-High

-Moderate

-Low

-Very low

The quality of evidence per recommendation should be justified by a summary of finding table, provided per recommendation:

-Type/s of studies: if different study types contributed to the recommendation please specify all designs.

-Quality assessment, each item classified as not serious, serious or very serious risk.

• Risk of bias: addressing the internal risk of bias of the studies contributing to the recommendation. Preferably provide in an appendix the risk of bias assessment for the main studies used to devise the recommendations. Using the risk of bias scores, low risk of bias would translate to “not serious: unclear risk of bias would translate to “not serious” or “serious” as judged by the guidelines panel; and high risk of of bias would translate to “very serious” risk.
• Inconsistency: exists when the summary of evidence is heterogeneous and the guidelines panel fails to explain it (e.g. with a dose-response relationship). The guideline panel should use judgment in appraising statistical heterogeneity; for example evidence from studies showing heterogeneous magnitudes of effect all in the same direction can be judged as less inconsistent that studies pointing at different directions of effect.
• Indirectness: examines the directness of the evidence substantiating the recommendation and asks whether studies were conducted addressing the precise comparison, the relevant patient population, intervention or outcomes.
• Imprecision: is defined by the confidence intervals surrounding the effect estimate and is dependent on whether optimal information size was met when compiling the overall evidence of the recommendation.

-Other considerations

• Publication bias: will downgrade the evidence if strongly suspected
• Large effect: will upgrade the evidence if existent. Risk ratios of >2 or <0.5are suggested to denote a large effect and risk ratios of >5 or 0.2 a very large effects, if no plausible confounding exists.
• Plausible confounding: relevant only for observational studies not downgraded on other factors and can increase the level of evidenceif the influence of all plausible confounding would reduce a demonstrated effect or suggest a spurious effect when results show no effect
• Dose-response gradient: will upgrade the evidence if existent

Type of recommendations:

-Strong recommendation

-Conditional recommendation

-Conditional recommendation for either the intervention or the comparison

-Conditional recommendation against

-Strong recommendation against

Resources for authors:

-Alonso-Coello P1, Oxman AD2, Moberg J2, Brignardello-Petersen R3, Akl EA4, Davoli M5, Treweek S6, Mustafa RA7, Vandvik PO2, Meerpohl J8, Guyatt GH9, Schünemann HJ9; GRADE Working Group. GRADE Evidence to Decision (EtD) frameworks: a systematic and transparent approach to making well informed healthcare choices. 2: Clinical practice guidelines. BMJ. 2016 Jun 30;353:i2089. doi: 10.1136/bmj.i2089.

-GradePro: for the handbook and for the generation of summary of findings tables

-The interactive EtD (

-The interactive Summary of Findings (iSoF; org/)

-MAGIC (

1