RE: alternatives to Instant medical history

December 29, 2007

> This is very interesting. I would hope it could remain in the simple

> file format to encourage mere mortals to contribute to it. I think if

> it goes to a database backend, people will get intimidated, and you

> will be less likely to get contributions from the users.

Actually I think putting into a database would make it easier ... I

was considering using sqlite, an embedded database program used

locally, or, to store the questions on a web server so that anyone

making an addition would automatically contribute to the whole

database for everyone else to use. The many hands make light work

paradigm .. though I fear that it might turn out to be too many cooks

spoil the broth instead!

>

> I also programmed a decision making tool called " MedMapper " in

> conjunction with Dr. Pepper of the UCSF, Fresno Family Practice

> program. has developed numerous (over 700) decision making

> " Maps " , as he calls them. They are basically a sort of branching

This is more of the clinical decision support type of program. Which

has to be updated continuously as new knowledge and guidelines appear.

Lots of work

Either way, storing branching trees in a text / xml file and allowing

users to see the logic easily so that they can make changes is a huge

challenge in itself.

--

Graham Chiu

http://www.synapsedirect.com

Synapse-EMR - innovative electronic medical records system

December 30, 2007

Lynn is focused on the IMH model, so the patient fills it out thus

saving time. Tkfp has the ability to allow patients editing access to

their own records, but it doesn't have a structured Q & A format like

IMH, sherlock.exe, or MedMapper (recognizing the MedMapper does more

than just history). I think that the sort of stuff in the PMH/ROS

will not be subject to change as fast as the clinical decision making

part, and it certainly makes sense to have the patient do more work,

....er, participate in their own care to the extent possible.

Maybe it would be a worthy project to work on a patient history form

utilizing the some of the principles of the above programs that is

intended for the patient to fill out, but is easier to read, and can

be fairly easily incorporated into any reasonable EMR.

Jerry

December 30, 2007

Graham,

I agree using an SQLite database is a better idea. The user can be

given a GUI front end to use,

doesn't have to see the SQL database or worry about it. We are also

using SQLite in the past year or so

with tkFP as SQLite can also be used with the Tcl/Tk programming

language.

We made a little Rolodex database with SQLite with a Tcl/Tk GUI

front end.

It sounds like SQLite also works with the REBOL language that you are

using in Synapse. That might make some

interesting cooperative things possible where we could cooperate

on using a common database for some type of project.

It would be interesting if a program could be made that would

record clinician's decision making and learn from it, applying that

knowlege to new clinical situations. I agree it would be a monumental

piece of work to make decision making algorithms for all the problems

we encounter in medicine with the type of tools like MedMapper.

That's why all we have on the MedMapper web site is the basic

framework and a few demo Maps. We got tired of doing them, especially

as the response from people was sort of underwhelming. It would be

cool if an EMR could just sort of learn about and record clinicians'

decision making automatically as they use their EMRs,

and automatically generate decision making algorithms or other help

from that. I guess you'd have to rate certain clinician's decision

making higher than others by various criteria that somebody would

have to decide, also a monumental task. On the other hand, the list

of things we do in Medicine that truly have been PROVEN to help,

such as immunizations and a few other things is actually pretty

short, so it is not too difficult to have the EMR record some of

those quality measures automatically, as opposed to diagnostic

decision making, which is a lot more complicated.

> > This is very interesting. I would hope it could remain in the

simple

> > file format to encourage mere mortals to contribute to it. I

think if

> > it goes to a database backend, people will get intimidated, and

you

> > will be less likely to get contributions from the users.

>

> Actually I think putting into a database would make it easier ... I

> was considering using sqlite, an embedded database program used

> locally, or, to store the questions on a web server so that anyone

> making an addition would automatically contribute to the whole

> database for everyone else to use. The many hands make light work

> paradigm .. though I fear that it might turn out to be too many

cooks

> spoil the broth instead!

>

> >

> > I also programmed a decision making tool called " MedMapper " in

> > conjunction with Dr. Pepper of the UCSF, Fresno Family

Practice

> > program. has developed numerous (over 700) decision making

> > " Maps " , as he calls them. They are basically a sort of branching

>

> This is more of the clinical decision support type of program.

Which

> has to be updated continuously as new knowledge and guidelines

appear.

> Lots of work

>

> Either way, storing branching trees in a text / xml file and

allowing

> users to see the logic easily so that they can make changes is a

huge

> challenge in itself.

>

> --

> Graham Chiu

> http://www.synapsedirect.com

> Synapse-EMR - innovative electronic medical records system

>

December 30, 2007

I have vaguely thought about this in the past. The advantage of

writing one's own EMR is that one can control the data capture. So,

my idea was that one would collect as much de-identified information

as possible, include the final diagnosis ( probably ICD9 since that's

the one most commonly used ), and submit this to a server. Gradually

we build up a picture of what diagnosis constitutes, and using Bayes

theorem, can then produce diagnostic possibilities based on whatever

limited information the patient presents with.

This approach means that the clinician does nothing except allow the

emr to submit de-identified data, and the power of statistics does the

work.

Collaborative data collection like this means perhaps that early

epidemics might be picked up etc.

Well, that was my little dream

> It would be interesting if a program could be made that would

> record clinician's decision making and learn from it, applying that

> knowlege to new clinical situations. I agree it would be a monumental

--

Graham Chiu

http://www.synapsedirect.com

Synapse-EMR - innovative electronic medical records system

December 30, 2007

Jerry- this would be great to see! All of my notes would be in a a 3 x 4,5,8,12, etc matrix (except for the 1-3 liner which I type in) which would amuse me to no end but which might really confuse the chart reviewers. But it's definitely an idea that could only emanate from the halls of math geekdom (and I don't mean this in a negative way ;-) ).LynnTo: From: parkjerryw@...Date: Sat, 29 Dec 2007 17:31:17 +0000Subject: Re: alternatives to Instant medical history

Instead of reading repeatedly, He denied, He denied, etc., maybe

output from Formedic's MHQ or IMH would be easier to read if was just

placed into a table. The rows would be the symptom category being

addressed (e.g., Cough Duration, etc,) the three columns would be

"Admitted", "Denied" and "Other" (like refused to comment, equivocal,

or whatever).

True, this would no longer look like a nicely scripted medical note,

but as we have discussed, but in this context the literate medical

note is dying anyway. I think it would be a lot easier to read than

irregular columns of "She admitted", "She admitted", etc.

I don't think placing a medical history into a table would matter from

the standpoint of reimbursement.

Jerry

Get the power of Windows + Web with the new Windows Live. Get it now!

December 30, 2007

- I really DO like the idea of open source. In my next practice I will seriously look at what is out there a year or two before I make the move to consider switching. For reference: can you guys list (or is there already a comparison list) of features that each project provides or does not provide the user? On the other hand, I also love the cheap EMR I am using (functionality is great, it is easy to use, it continues to improve) and I hardly am going bankrupt on its account- recurring costs: essentially $500/year upgrades, $50/month IMH; Updox $200? per year or is that a one time fee? I don't know, Shortkeys $20 1x, Shorthand for Windows $70 one time - that is not a lot. Directly attributed EMR costs somewhere around $120 a month? This is NOT a lot, an extra 2 -3 patients per month. Consider how much you have to pay for malpractice- now which is the better deal?Other costs which are not recurring- upgrade for billing software - $300 this year due to NPI; And recurring costs which you'd have to pay free EMR or not: on line appointment maker $16/month, clearing house costs $60/month, website hoster $90/2 years, I suppose you can count Windows XP costs if most open source runs on Linux... hardware replacement costs you have to pay for no matter what you use. Many other pieces- 'free' - example paperport came bundled with all in one.I will look at formedic and see if I feel like tolerating the drug ads and see if I can do what I need to do with the pdf versions. Probably not (but if I had to - sure), I will post it when I get to it.If the pain of paying the above was more than pain of migrating my EMR data and starting out on the programming learning curve, I would switch. But clearly it's not so I don't.Did I miss any big points? LynnTo: From: alcald3000@...Date: Sat, 29 Dec 2007 20:04:50 +0000Subject: Re: alternatives to Instant medical history

The really important point to remember about open source is that once

there is even one open source EMR that really works and does

everything we need whether it's Synapse, tkFP, GnuMed, CottageMed

or ?? , we will all be forever free of vendor lock. Once a program like

that is out there and the code is freely available and can reproduce

itself on the Internet and be maintained and modified forever by any

interested programmers and groups of users, we will have close to

EMR Nirvana. Open source software licences specifically prohibit

making the code secret. Part of the license agreement requires that

you share it with others. It is the total antithesis of the business models

we have in the proprietary software world. That's not to say you can't

make money with open source software. But you do it by providing

support and services related to installation, maintenance, upgrading,

hardware, rather than licensing the code itself.

Caldwell.

> >

> > > >

> >

> > > > Are there any alternatives to instant medical history?. Paying

$50

> >

> > for

> >

> > >

> >

> > > I was challenged to see if I could write a literate medical history

> >

> > > program, and my attempt is called sherlock.exe

> >

> > > http://compkarori.no-ip.biz:8010/Unrelated_Software_Resources

> >

> > >

> >

> > > You can edit the questions.r file to create your own branching

> >

> > questions.

> >

> > >

> >

> > > It would take a helluva lot of work to create a comprehensive

system

> >

> > > and I was just toying with the idea of creating the framework.

> >

> > >

> >

> > > I didn't like IMH and their relations because of the stilted output.

> >

> > >

> >

> > > Graham

> >

> > > SynapseDirect.com

> >

> > >

> >

__________________________________________________________

______

> > The best games are on Xbox 360. Click here for a special offer on

> an Xbox 360 Console.

> > http://www.xbox.com/en-US/hardware/wheretobuy/

> >

>

i’m is proud to present Cause Effect, a series about real people making a difference. Learn more

December 30, 2007