2.1. WHO guidelines development process
The guidelines were developed according to the requirements described in the WHO handbook for guideline development (14) and a planning proposal approved by the WHO Guidelines Review Committee.
The development process included six main stages: (1) identification of the PICO (Population/Participants, Intervention, Comparator, Outcomes) question, an approach commonly used to formulate research questions; (2) the conduct of two systematic reviews for the retrieval of the evidence using a standardized methodology; (3) development of an inventory of national and regional IPC action plans and strategic documents; (4) assessment and synthesis of the evidence; (5) formulation of recommendations using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach; and (6) writing of the guidelines and planning for the dissemination and implementation strategies.
The development process included also the participation of four main groups that helped guide and greatly contributed to the overall process. The roles and functions are described herein.
WHO Guideline Steering Group
The WHO Guideline Steering Group was chaired by the coordinator of the WHO IPC Global Unit in the Department of Service Delivery and Safety. Participating members were also from the Antimicrobial Resistance Secretariat, the Water, Sanitation and Hygiene (WASH) Unit, and the IPC focal points at the WHO Regional Office for the Americas and the Regional Office for the Eastern Mediterranean.
The Steering Group contributed to the initial planning document for the development of the guidelines, identified the primary critical outcomes and topics and formulated the research questions. The Group identified systematic review teams, the guideline methodologist, the members of the GDG and the external peer reviewers. The GDG chair and the IPC Global Unit coordinator supervised the evidence retrieval, syntheses and analysis, organized the GDG meetings, prepared or reviewed the final guideline document, managed the external peer reviewers’ comments and the guideline publication and dissemination. The members of the WHO Steering Group are presented in the Acknowledgements section.
WHO Guidelines Development Group
The WHO Guideline Steering Group identified 24 external experts, country delegates and stakeholders from the six WHO regions to constitute the GDG (also referred to as “the panel”). This was a diverse group representing various professional and stakeholder groups, such as IPC, clinical microbiologists, epidemiologists, public health and infectious disease specialists and researchers. Geographical representation and gender balance were also considerations when selecting GDG members. Members of this group appraised the evidence that was used to inform the recommendations, advised on the interpretation of the evidence, formulated the final recommendations while taking into consideration the 2016 WHO Guidelines on core components of infection prevention and control programmes at the national and acute health care facility level (13), and reviewed and approved the final CRE-CRAB-CRPsA guideline document. The GDG members are presented in the Acknowledgements section.
External Peer Review Group
The Group was composed of five technical experts with high-level knowledge and experience in IPC, AMR, patient safety and health management, including field implementation, and a patient representative. The Group was geographically balanced to ensure views from both high- and low- and middle-income countries (LMICs); no member declared a conflict of interest. The primary focus was to review the final guideline document and identify any inaccuracies or errors and comment on technical content and evidence, clarity of language, contextual issues and implications for implementation. The Group ensured that the guideline decision-making processes incorporated values and preferences of end-users, including health care professionals and policymakers. Of note, it was not within the remit of this Group to change the recommendations formulated by the GDG. All reviewers agreed with each recommendation and some suggested selected editing changes. The members of the WHO External Review Peer Group are presented in the Acknowledgements section.
Research question/PICO
The specific PICO question was developed by the WHO secretariat based upon feedback and discussion by the GDG responsible for the 2016 WHO guidelines on core components of infection prevention and control programmes at the national and acute health care facility level (13). The main research question underlying this work was:
What is an effective approach to preventing and controlling the acquisition of and infection with CRE and/or CRAB and/or CRPsA among inpatients in health care facilities?
For each intervention, the PICO question was formulated as follows:
Population: patients of any age admitted to an inpatient health care facility including acute health care facilities, secondary or tertiary health care facilities, LTCFs and rehabilitation centres.
Intervention: any IPC measure (single measures or part of a multimodal strategy) implemented to contain CRE-CRAB-CRPsA transmission in the inpatient setting (for example, screening policies, contact precautions, hand hygiene interventions, environmental cleaning). We excluded studies exclusively dealing with bacterial isolate collection and identification, susceptibility testing, basic science or animal models, treatment, prophylaxis, stewardship, or duodenoscopes/endoscopes.
Comparator: regular care practices with no specific IPC intervention.
Outcome: CRE-CRAB-CRPsA transmission within the inpatient facility measured by the incidence or prevalence of acquisition of colonization and/or infection with these organisms
2.2. Evidence identification and retrieval
According to the guidelines development plan approved by the WHO Guidelines Review Committee, a literature systematic review and an inventory of national and regional IPC action plans and strategic documents were conducted.
Literature systematic review
The systematic review followed the guidelines development plan approved by the WHO Guidelines Review Committee as well as the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) statement (45).
Search strategy
The following databases were searched:
MEDLINE
Excerpta Medica Database
Cumulative Index to Nursing and Allied Health Literature
Global Index Medicus
Cochrane Library
Outbreak Database
Abstracts from the following international conference were also retrieved:
Interscience Conference on Antimicrobial Agents and Chemotherapy and the American Society of Microbiology Microbe;
European Congress of Clinical Microbiology and Infectious Diseases;
Infectious Diseases Society of America Annual Scientific Meeting (ID-Week);
International Conference on Prevention and Infection Control.
No time delimiters were used for the study selection, although conference abstract searches were restricted to the last five years (2012–2016). Languages included were English, French, German, Italian, Portuguese and Spanish. The search terms used were adapted to each database, but always based on a combination of three concepts:
carbapenemase/carbapenem resistance;
core IPC measures;
CRE and/or CRAB and/or CRPsA (CRE-CRAB-CRPsA) colonization and/or infection rates (that is, primary outcomes).
All steps of the systematic review concerning references retrieved from electronic databases were performed using the DistillerSR® systematic review software (Evidence partners, Ottawa, Canada). References retrieved from conferences were manually screened using the same inclusion/exclusion criteria. Removal of duplicates was performed before title and abstract screening using Endnote and the algorithm provided by the Distiller SR® software or removed manually.
Screening and data extraction
Using a standardized screening form, two reviewers reviewed all titles and abstracts retrieved from electronic databases and conference abstract sites. Disagreements between reviewers were resolved by discussion and a third reviewer as necessary. Studies were not considered when the title and abstract clearly indicated that the study did not meet the inclusion criteria (see PICO question above). If a study passed the title screening, but an abstract was not available, it was passed to the full-text screening level. If a report of the same study was duplicated in several records, the most recent peer-reviewed publication was included.
Full-text eligibility of all studies was independently conducted by two reviewers with reasons for exclusion annotated and tracked in Distiller (for example, “not about CRE-CRAB-CRPsA”). The primary reason for excluding studies was if the article did not meet the defined eligibility criteria (see PICO question above). Conflicts and uncertainties about whether to include or exclude a reference were discussed with another investigator of the team until consensus was reached.
Two reviewers independently performed data extraction of all included studies using a standardized data extraction form and any uncertainties about extracted data were discussed with the team.
Risk of bias assessment and evaluation of the evidence
All included studies were assessed against design-specific Effective Practice and Organization of Care (EPOC) entry and quality criteria (http://epoc.cochrane.org/epoc-specific-resources-review-authors). Eligible EPOC study designs included randomized controlled trials, nonrandomized controlled trials, controlled before-after and interrupted time series (ITS) studies with sufficient data to statistically assess before-after trends. When studies appeared to be potentially of EPOC standard, but there were insufficient published data to be certain, study authors were contacted to seek additional relevant information. Based on both the published data and the authors’ responses to the data request (as well as additional data analyses as needed), these studies were again assessed as to whether they were of EPOC or non-EPOC standard.
For potential ITS studies, the following four entry criteria were deemed necessary for a study to be classified as EPOC.
Clearly defined time points when the intervention(s) occurred.
At least three data points before the main intervention and three after.
Objective measurement of performance/provider behaviour of health/patient outcome(s) in a clinical situation (for example, CRE-CRAB-CRPsA detection in clinical cultures and/or screening swabs).
Relevant and interpretable data presented or obtainable. Preferred results included change in slope (that is, the trend in pre- compared to post-intervention periods) and level (that is, the immediate change after intervention implementation) in outcome prevalence or incidence.
The risk of bias among ITS studies was assessed using the ITS-specific Cochrane checklist including:
intervention independent of other changes;
shape of the intervention effect pre-specified;
intervention unlikely to affect data collection;
knowledge of the allocated interventions adequately blinded during the study;
incomplete outcome data adequately addressed;
study free from selective outcome reporting;
study free from other risks of bias.
For EPOC-compatible studies, the Grading of Recommendations Assessment, Development and Evaluation (GRADE) (http://www.gradeworkinggroup.org/#pub) evidence profiles were created assessing:
study limitations;
inconsistency of results;
indirectness of evidence;
likelihood of publication bias;
number of participants;
quality.
Risk of bias assessments using the EPOC framework were conducted by two reviewers. Disagreements were resolved by consensus or consultation with the project’s senior author and/or methodologist if no agreement could be reached. Studies not meeting the EPOC study design criteria (“non-EPOC studies”) were not formally assessed and their quality was considered very low.
An evaluation of the overall body of the evidence was conducted using the GRADE system and according to specific outcomes, including a synthesis of results and quality of evidence assessment. Evidence was synthesized descriptively. It was not possible to perform a meta-analyses due to the wide range of intervention packages and outcomes assessed and a large degree of heterogeneity in study designs and methods used in the included studies. Quality of evidence was assessed in terms of study limitations, consistency and precision of results, the directness or applicability of summary estimates, and the risk of publication bias (46, 47). It was classified as high, moderate, low or very low according to these factors (15).
Inventory of national and regional IPC action plans and strategic documents
A methodology and data capture approach was developed for the inventory to identify, record and analyse regional and national documents addressing guidelines related to the management of CRE-CRAB-CRPsA infections and/or colonization, including a web search and expert consultation. The approach covered all six WHO regions (African Region, Region of the Americas, Eastern Mediterranean Region, European Region, South-East Asia Region and the Western Pacific Region). WHO regional focal points and GDG members were requested to provide input on existing documents from countries and regional offices. For documents with no existing translation in English, French, Spanish, Germany, Italian or Portuguese, contacted experts were asked to summarize the key points presented in the documents.
A summary of the inventory’s findings is reported in Appendix 2.
Evidence appraisal and development of recommendations by the GDG
The results of the systematic review and regional inventory were presented at a GDG meeting held on 1–2 March 2017. The standardized methodology including the PICO question, GRADE framework and EPOC criteria were described. The GDG also evaluated the relevance of the 2016 WHO Guidelines on core components of infection prevention and control programmes at the national and acute health care facility level (13) as a foundation for the development of these recommendations for the prevention and control of CRE-CRAB-CRPsA.
Recommendations were then formulated by the GDG based on the quality of the evidence, the balance between benefits and harms, values and preferences (for example, those of patients, health care workers, and policy-makers), resource implications (for example, at the national and facility level) and acceptability and feasibility. These were assessed through discussion among members of the GDG. The strength of recommendations was rated as either “strong” (the panel was confident that the benefits of the intervention outweighed the risks) or “conditional” (the panel considered that the benefits of the intervention probably outweighed the risks). For some recommendations, the GDG decided that the strength of the recommendation should be strong despite limited available evidence and its very low to low quality. These decisions were based on consideration of the magnitude of effects reported in the included studies and expert opinion consensus. The methodologist provided guidance to the GDG in formulating the wording and strength of the recommendations. Full consensus (100% agreement) was achieved for the text and strength of each recommendation.
The draft chapters of the guidelines containing the details of the recommendations were then prepared by the IPC Global Unit team and circulated to the GDG members for final approval and/or comments. All relevant suggested changes and edits were incorporated in a second draft. The second draft was then edited and circulated to external peer reviewers and the draft document was revised to address all relevant comments. Based on the reviewers’ comments, the discussion was also expanded in some cases after the main GDG meeting via email or teleconferences. When this was necessary, feedback and final approval was gathered from all GDG members. Additionally, a review of the guidelines was conducted by the WHO Public Health Ethics Consultation Group. The group recommended greater discussion of potential harms (for example, psychological suffering among patients identified as colonized or infected), unintended consequences (for example, discrimination) with ethical implications and potential mitigation measures. To address this feedback, suggestions from the reviewers were added. Furthermore, key principles and lessons from guidance on ethical considerations for other infectious diseases were assessed and incorporated accordingly.