I am wondering about what sort of content would be best for test/dev and demonstration databases.
I am thinking that a database/demo appropriate for different specialties might also be nice to have, as some folks(physicians and other decision makers) simply DO NOT have enough imagination to extrapolate, and have little tolerance for the term “configurable”.
I don’t believe there is currently a way in OEMR to assign a patient to a facility, so I can’t make a demo where if you sign in as “Cardiologist” you just get cardiology patients. I am pretty sure that we can support multiple sites on one server instance however, and each site can have a variety of configurations (Globals etc…). Anyone have any input on the best way to approach this?
Here are some off-the-cuff examples, but I would like to get some input as to what would be most useful to both developers and err….salesmen…pitching Openemr.
Scheme 1.
A general database such as a PCP clinic might use, with one facility.
Scheme 2.
Multiple facilities as billing locations for some sort of home care nursing practice with Office, Home, and Nursing Home facilities.
Scheme 3.
A Cardiology Practice that integrates a lot of imaging, as well as device monitoring. Facilities would be Office, remote monitoring, and Hospital (for proceedures).
Scheme 4.
Highly specialized database demonstrator:
Multiple facilities as billing locations for mental health counseling including home, office, and school, with modified searchable demographics data including commonly tracked variables such as counselor assignment. I also wish to integrate a fee sheet mod that will allow non-billing provider data to be included on the fee sheet for supervised contract provider payments.
Scheme 5
Multiple facilities with each provider being restricted to their facility. If you log in as the Intern assigned to make rounds at the hospital for a practice, then you only get the Hospital as the billing location. The Phleb tech only gets the office as a billing location etc…
Adequate for most, yes. As it is, I am making a little app to generate random names, vitals, and demographics out of lists to create random patients, as it would take about the same amount of time to do a single list of 25…and I want multiple versions and like 100 patients in each so that search features can be investigated more effectively. I will next look at creating notes randomly, but I am hoping to get some “denatured” notes from some people in various fields to use as templates. If anyone has got some progress notes or better yet proceedure reports that I can use, please send-em to me or copy and paste them here.
I thought the biggest pain would be the calender, but apparently, using repetitive appointments is going to make that easy.
Just posted the 4.0.0 demo (with sample data). This can give you an idea of some data elements to provide. Note this is done via the poor man method of manual entry(and added the icd9 codes), so not much there (obviously would be nice to have a lot more sample data): http://www.openmedsoftware.org/wiki/OpenEMR_Version_4.0.0_Demo
For Testing purposes, its very important to have these five NIST Patient Records in the 4.X Demo/Testing system, each consisting of one of the five “Patient Summary - Record Data Sets” starting on page 11 of the following PDF
Hi John,
OK, some of that I don’t get, like SNOMED numbers and the like. Apparently, we are going to need some decent ICD’s/CPT’s/Vcodes and HCPCS in the test database as well.
John, you are apparently somewhat familliar with this stuff. I would like to get what I can entered, and have you check it over before we toss out a database/site to cover this (probably #1 priority) to cover this need. Are you OK with that?
-Art
P.S. My expected date of completion for my first sets of data is Wednesday-ish.
*Victory Dance*
Just had to share. Finally got my damn app to write a valid SQL command text file to create a long list of patients (as many as you like). You just have to run the script and it makes the patients.
Demographics are out of the way at least….>one< little freaking boo-boo in the file writing code and everything went to crap. I hate those sorts of things, but it is indeed such a rush to finally track them down!
-Now to take the “DEV” label out of our EOMR installation…another great day (err…night anyway)!
Next thing, what insurance companies should we use for a test DB? Fake ones, or perhaps straight medicaid/Medicare? I think this may somehow be important.
I would put fictitious companies with a pointer in their name to the fact that they are Insurance companies to guide users e.g “Great Insurers Limited” etc.
How about creating a new project in git with your SQL script and a mysqldump of your output. You could then just forward us a link to the git repository.
That’s great for the database, but perhaps I need to include the uploaded documents in the GIT? I have not done so well with GIT so far…I gotta get that straightened out. Also, I have not just created an SQL script, I have written a program that uses lists in .ini files that creates the text for an sql script (not being an sql guru). It is a windows app (I hear groans in the background)…
Today is med appointment day, and I have to work on some forms before clients arrive….hope i can get this together tonight sometime. It would be easier if the documents were stored in a blob.
I would load the database using your script, then use mysqldump to save the database as an sql file. Then create a branch off the master in git that contains the sql dump (in the sql dir). Then … put the document dir tree in that branch as well. I can help with the logistics of this if you need me too.
The other option is to put both in a ZIP file and post them on the wiki or as a “patch” on tracker, for now.
It’s me again, Art. Did you ever do a zip of the database you created, or is it a database entry generator? Where can it be found? Inquiring minds want to know. Thanks, Jack
Oh Your God……
I friggin forgot my whole rear end here.
I was having trouble with my incremental/random date-times looking believable. I also need to add new elements so that the calender mod can be represented. I then got sucked into 3 simultaneous “emergencies” requiring all my time. I totally flaked out on this one.
When I get home, I should have a decent data base that was output. I will “box it up” and then see about compiling the database generator.
Thanks for the prod Jack!
K,
I got off my rear.
I made a version that I *Hope* works for 4.1’s current dev database. My old one was clunky and was tuned to my version which has a lot more insurance and x12 config data in the tables.
You can get it here: http://www.oemr.org/phpBB3/viewtopic.php?f=5&t=20
It is an .exe, but jack tested an early version on debian/Wine
You can make several million unique patient records, Thousands of Insurance companies, and Hundreds of x12 partners all in one go. It populates primary, secondary, and tertiary insurance data for patients.
If you are merely creating a few hundred patients, you can fire it up and have your .sql script about 3 seconds later. I hope this is helpful (if it passes muster on the quality of the sql script it writes once someone with a clean install tests it) for demos as well as to load up large db tables for sql query testing.
Note: I am not a db guy, and I have no limits set on the size of the scripts it creates…I don’t know what limits there may be. Start with a reasonable size output for tests I suppose.