We welcome your questions about the 3e Initiative in Rheumatology as well
as your feedback and / or questions related to the educational content provided
through this Web site.
About 3e Methodology
The 3e methodology was in line with that recommended and used by EULAR 1. Our aim was to meet all the criteria demanded by the AGREE method 2 in order to guarantee the quality of our recommendations.
In line with best practices, all participants were asked to ratify:
- The target area of discussion
- The choice of themes and questions within each theme
- The methodology for the systematic analysis of the literature (evidence-based medicine)
- The modalities for the collection of experts’ opinions
- The presentation of the recommendations (decision trees or short sentences, appropriate vocabulary depending on the force of the arguments), dissemination and evaluation of the recommendations
- The process for implementation and updating the recommendations
Participants were asked to consider the questions and develop recommendations. All participants reconvened to decide on the questions for this programme through a voting exercise. By the end of this process, recommendations were reached on how to investigate and follow-up undifferentiated peripheral inflammatory arthritis.
- Dougados M, Betteridge N, Burmester GR, et al. EULAR standardized operating procedures for the elaboration,
evaluation, dissemination, and implementation of recommendations endorsed by the EULAR standing committees. Ann Rheum Dis 2004; 63: 1172-6.
First Multi-national Meeting
The first meeting of 3e brought together the international scientific committee and the members of the national scientific committees of the 17 countries involved in the project. The outcome of this first meeting was the selection of the themes and development of 10 questions per country.
- This process started with a review of approximately 170 questions from all committees.
- The steering committee formed a shortlist of proposed questions by removing any duplication.
- The Delphi method, a consensual, voting process, was used for this selection. Round by round, participants were asked to vote on their preferred questions to continually narrow the list. Several rounds of voting were necessary to make the final question selection.
- At each round the following rules were applied: Questions with 66% or more of the votes will be automatically chosen, while those with less than 25% of the votes were dropped. Only the questions selected by between 25 and 66% of the participants were put forward for the following round.
- Finally, 10 questions were retained to be answered.
Systematic Literature Research
The final list of questions were given to a bibliographic group to perform a systematic literature research (SLR) concerning the topic. The bibliographic fellows were tutored by mentors. The main stages of this SLR were:
- What is the exact phrasing of the question?
- How have I understood this question in order to anticipate the adequate analysis?
- Based on this anticipated analysis, which data will I have to collect in each selected manuscript?
- How have I planned the literature research?
- Selection of manuscripts
- Quality scoring of manuscripts
The evidence from the literature research for the 10 questions was presented twice: The first involved an SLR progress check with the mentors only and the second involved an SLR summary with one or two delegates from each country. All the national scientific committee members received detailed background to be able to present the results during their National Meetings.
The following flow chart demonstrates sample search strategies for diagnostic papers (Figure 1). Sources included databases such as: Medline, Embase, and Cochrane Central. Studies were excluded if they were reviews, did not deal with the population of UIPA, or did not specifically study the role of history or physical exam features. Non-English articles were reviewed and kept for analysis if felt to be relevant. Additional articles were retrieved using the “find similar articles” function. Duplicate studies between databases were noted, as well as the number of diagnostic studies found. Similar parameters were used for each Recommendation question under investigation by the scientific committee and bibliographic fellows.
Search strategies and methodology employed to narrow in on potential questions to be answered and made into a Recommendation:
Figure 1: Search Strategy for Diagnostic Papers
In each country, the evidence from the SLR was presented and discussed in discrete
workshop groups led by a designated moderator from the Scientific Committee. During the discussions, the review of the literature concerning the selected question was presented and then the participants discussed the question, adding their practical expertise to formulate their answer. These answers are the basis of the national recommendations, which are then finally ratified in a vote-session (Delphi method).
Multi-national Final Meeting
The 17 Scientific Committees met again to present their National Recommendations and assess the recommendations based on their knowledge and experience and the best clinical evidence available (Figure 2).
Figure 2: Evidence-based Practice
Both the quality and levels of evidence followed the standardized grading system, as per the Oxford Centre for Evidence-based Medicine Levels of Evidence (http://www.cebm.net/) This included Levels 1a,b,c, 2a,b,c, 3a,b, 4, and 5 as outlined in Table 1 (below):
Data quality and levels of evidence assessed via a standardized grading system:
Table 1: Assessing Strength of Evidence
Publication of Recommendations:
Consistencies and discrepancies between the different national proposals were discussed and we aimed to determine a multi-national report of these recommendations for submission to EULAR. The 10 Recommendations developed were, thus, evidence-based and agreed upon by expert consensus. Additionally, a publication and dissemination plan was agreed upon for the multi-national recommendations. Figure 3 summarizes the process and Table 2 summarizes the roles of the various committee participants.
The following overview diagram and table summarize the major steps that were employed to initiate, develop, and generate each of the 3e Recommendations for UPIA.
Summary Figure – Step-wise events leading to the development of the 3e Recommendations:
Figure 3: Development of the 3e Recommendations
Summary Table – Step-wise process leading to the development of the 3e Recommendations:
Table 2: Overview Process