Anthropology - Part 1: Organizing Semantic Modeling Meetings

Anthropology—Part 1: Organizing Semantic Modeling Meetings

To date we have used only face-to-face meetings. Teleconferenced meetings or collaborative, asynchronous approaches may be fruitful, but we have found the face-to-face setting so powerful that we have not experimented with the less interpersonal approaches.

Defining the Sessions

A semantic modeling session should be between 2 and 3 hours long. Few groups can stay on task at the level of involvement needed longer than that. A single moderator should not plan more than one session in a day; we have done as many as four in a day, and they are not nearly as productive (it requires a certain energy to conduct these well, and it requires time between sessions to digest and document what you've learned).

You will need a scribe who is also a semantic modeler. The action will be moving fast and furious, and an untrained scribe will not be taking down the appropriate information.

To keep the sessions to the 2-to 3-hour limit, as well as to figure out who should be invited, you will want to divide your scope into a set of "topics." You needn't worry too much about the size of the topic; the discussion will expand or contract as it needs to. Topics for a health care system might include "diagnosis and disease," "patient scheduling," "physician scheduling," "coding a visit for financial remuneration" "patient insurance."

Who Should Be Invited?

Once you have the topics you must determine who should be invited. You want at least 5 people (in addition to the moderator and scribe). You do not want more than 12. You want as many valid and varying viewpoints as you can get on a given topic. For example, on "patient scheduling" you might invite two physicians, preferably with differing opinions about how patients should be scheduled. You will want an administrator who is involved with routine patient scheduling, as well as someone who deals with emergencies and exceptions. You will want people affected by scheduling, such as office managers and nurses. Ideally, you would like a patient or two. You will want someone who deals with insurance (you'd be surprised at the insurance issues that come up with scheduling). You should include one or two people familiar with the current systems; you want to avoid having them dominate the conversation, but at the same time there will be many misperceptions about how the current system works that they will be able to clear up.

Setting

It's best to have everyone around a conference table, but auditorium seating can work. You'll need a lot of white board space. In the absence of white board, or as an adjunct to white board if you only have one board, you can paper the walls with flip charts.

Facilitator's Preparation

You will want to do some background reading before you go in, especially if you are not familiar with the vocabulary and concepts, but avoid trying to design the model on your own. If you do, you won't be listening to the dialog.

You will want to have some provocative questions ready in case things slow down, especially in the early part of the dialog. When we did "diagnosis and disease," the facilitator for that session opened with the question, "So what is the difference between diagnosis and disease?" The answers started as glib definition, but after returning to this question several times, a rich vein of meaning was opened.

Finally, you will want to cultivate what the Zen masters call "beginner's mind," meaning you want to set aside what you know long enough to hear what is being said, as if you were hearing it for the first time. (If you are asking leading questions and feel like saying "That's not the answer I was looking for," you're not in beginner's mind.) I have been involved with project management for years, and was certified as a Project Management Professional (PMP) from the Project Management Institute (PMI). I once started a session with architects, engineers, and project managers by asking the question, "So what is a project anyway?" Apparently I was sincere enough that I got a blank stare that seemed to say, "Who is this guy who purports to be a project management expert and doesn't know what a project is?" After the awkward silence, we had a great conversation.