Information Technology Reference
In-Depth Information
Documentation flow in Praxis is quite different than in traditional charting.
It normally does not begin with the chief complaint - the patient's own typically
vague statement of their problem - as physicians are trained to do. This is because
the chief complaint will not provide sufficient specificity for the concept processor
to find the best prior case. Physicians using Praxis are trained to first input the most
clinically specific “input” (a “clinical finding”) they can. This allows the concept
processor to more accurately find the identical or best match to the patient being
seen. Physicians can group subsets of an overall problem, such as acute pharyngitis
(sore throat), according to whatever clinical divisions make sense to them. This
further assists in getting to the best possible matching case and its associated clini-
cal concepts. Physicians can post their own personal approach to clinical concepts
so, for example, a family physician can import the approach used by an expert neu-
rologist to evaluate headache.
Praxis is available as a hosted service (the company recommends an iPad as the
end user device) or as a licensed client-server software program. The company says
it has around 3,000 clients with its largest user community being family physicians.
Visualizing the Data:
Today, computing is essentially ubiquitous in all fields of
endeavor. Across virtually all of those domains our ability to collect and aggregate
data is well ahead of our capability to analyze and visualize it effectively. In fact,
visualization is now an accepted branch of computer science. It hasn't received
enough attention in health informatics which is unfortunate and, I think, likely to
change.
Some of the visualization issues in health informatics can be illustrated with the
story of a group of my students who entered a challenge to visualize the data in a
CCD. Their visualizations were intended to support a busy physician on call for his
group who is faced with an unfamiliar patient.
14
Our three teams each decided that
a particular aspect of visualization was most important to a busy clinician.
One team emphasized rapid access to any part of the data from any other part.
They created a “dial menu” very much like a clock face except that each section was
part of the chart - medications, lab results, problem list and so on. The clinician
could just point to any section and the information stored in it would pop-up.
Clicking on any section provided more detail about the data stored in it.
A second team emphasized configurability of the data presentation. The clinician
could control what sections of the data would appear and where they would
be located on the screen. It was very easy to move them around, as needed.
This approach was very similar to the user configurable “My Whatever” pages so
common on the web.
The third team emphasized clinical relationships. The clinician could point to
any problem on the patient's list and only the data relevant to that problem would
appear. This team won the competition but I was allowed to watch the judges, many
14
The CCD was the one posted on the Internet by Dr. John Halamka that we referred to in the last
chapter. [23]
