Conference planning began with an implementation plan, which outlined the organizing committee, preconference activities, conference process, and postconference requirements. The actual conference involved four stages:
To begin, Dr. Mullins convened the conference and welcomed the participants. Dr. Lorenzi presented an overview of the process developed for the conference. During the presentation process, a keynote presentation to frame the issues was delivered. Presentations were made about UMLS (Unified Medical Language System) as a classification that could potentially link to all coding systems. Three coding systems panels, each consisting of three people, presented the strengths and weakness of the coding scheme they were representing.
Before the conference, each consensus panel member was asked to select one area in which to be the "primary reviewer." Four people selected the three coding schemes plus UMLS; the fifth was asked to be the primary person to look at the overview framing issue that the conference was addressing.
During the presentations, one of the facilitators developed a draft model of the consensus panel report. The draft consisted of the process, an overview framing issue, vision, critical elements, and where are we today.
With the draft report as a focus, the consensus panel members discussed the preamble, framing issue, vision, and critical elements. These required the most discussion time. Each person was the primary reviewer for one of the coding schemes; once everyone agreed on the vision and critical elements, outlining the strengths and weaknesses went quickly. The conference organizers included a transcription service as part of the process. Thus, once panel agreement was reached, it was easy to create the final consensus panel report that the total conference would discuss. Transcriptionists captured each member's statement of strengths and weaknesses as it was read. A disk of the report was then taken to a quick-copy service, which produced a report for all conference members.
The discussion process started early on the second day. Based on the number of attendees and the appropriate group size for discussion and participation, seven small groups were created. First, one discussion leader/facilitator and one reporter were chosen for each group. Each reporter had to have a laptop computer or access to a laptop computer to capture the information. After the discussion leader/facilitator and reporter were selected and placed in an appropriate numbered group, attendees were randomly assigned to groups.
After assembling in the appropriate group, participants were asked to introduce themselves to all the people present and note the names and addresses of the participants. In addition to outlining any group issues that they considered important, participants were asked to address the following five discussion issues, created by the facilitator.
All groups followed the five discussion issues, but each seemed to focus more extensively on one or more of the items.
The goal of the conference was to incorporate all of the discussion issues into the consensus panel report in order to develop the total conference report. Therefore, the discussion lasted for several hours.
After the discussion process, the participants convened as a total group to begin the consensus-building process. It was decided early in the conference that anything attendees said would be transcribed and projected onto the screen for everyone to see.
Each of the seven groups was asked, in turn, for comments and a summary of its discussion. Projecting the comments onto the screen had both a positive and negative effect. The positive effect was to build trust in the conference outcome, as all participants could see their precise comments projected for later incorporation into the final report. The negative effect was a sensory overload, since people were both participating in the discussion and also trying to read all the comments. In this conference, disagreement among participants was handled by agreeing to trust a small group to help resolve the issues. It was agreed that representatives from each of the seven groups would see the first draft of the total conference report and make any necessary changes after an appropriate electronic discussion.
After the meeting ended in New Orleans, a four-person edit group reviewed all of the documents (the consensus panel report, the reports of each of the seven discussion groups, and the printed verbatim transcription of all discussions on the second day) and extracted what "happened" during the meeting. This report was made available to all attendees by direct E-mail. It was also posted on the World Wide Web for review. Reviewers could give approval or recommend changes (along with giving the rationale for those changes). Most of the suggestions for change were incorporated in this final document, which has now been approved by all participants.
Return to Primary Care
Research Findings
AHRQ Home Page
Department of Health and Human
Services