Hi,
I would like to discuss my current thoughts on clinical
messaging systems, and encourage others do so here as well. What
follows is a brief documentation of the design of a clinical messaging
system that I have working in MirrorMed now. The MirrorMed parent
project, ClearHealth, is planning on doing clinical messaging in a
different way, and my not accept my patches. But when it is stable I
will be submitting it back, as a module. My purpose in emailing this
list is to both learn better ideas, but also to formally publish what
I am doing in a way that can be referenced as prior art, for software
patents. I would encourage others to use this list for the same
purpose.
My clinical information system is designed to work with an
interface similar to email. Messages can be "about a patient", if they
are, that patient is notified with a message that his providers are
discussing his case. To choose to send someone a message, you first
find the patient you want to discuss, then you move to the "send
message" interface. Here you either choose that person from a
drop-down, auto-suggest or simple typing an email. Then the user
chooses a "priority" for the message, and decides if the message is a
"todo" item. Depending on the priority of the message, user can
receive a standard email notification of a message. To read the
message, the user must either be logged in via a PHR or the EHR.
Patients can be messaged through the system. Patients can also
initiate messages through the PHR, but only to those providers who are
part of that patients "provider list".
In email, when you forward a message, that can be done with,
or without the original senders knowledge. However, in a clinical
messaging system, it is vital that all participants in a "virtual
consultation" know the entire context of the message. Thus, when a new
clinical user is introduced to a "thread", the system ensures that
that user can see all of the previous messages in that thread.
Generally, the system tracks which users have read which messages.
When the user logs in to view messages, already-read messages are
"rolled up" and only display the subject line. If a Message has not
been read, the system displays the entire message.
The priority of a message determines whether or not an email
is sent informing a user that a message exists, and when a todo item
is sent, users can mark items as "done" or "not done". So that the
messaging system can hand out simple tasks that might not be
encompassed in an "orders" system.
Features that I have not implemented yet that are pretty obvious:
1. Allowing intermediate users to approve and forward messages, so
that staff members can protect clinicians from frivolous messages from
patients.
2. Allowing "group" email targets, like "all nurses" or "all doctors"
or "all users at the main st location".
3. Allowing attachments.
4. Allowing links to different elements in the patients record.
I want to know what I am missing? What else makes a good clinical
messaging system? What else is available in open source that I should
be emulating. Remember publishing your ideas here is a big step to
prevent bad patents from being made!
-FT
--
Fred Trotter
http://www.fredtrotter.com
