Has there been any thought given to supporting access to clinical data by multiple independent practices/organizations? What I envision is a system perhaps running on AWS where users from all of the country log in. The organization that they belong to represents the context with which they access the system. Patient demographics and notes are visible to all users with safeguards to prevent a member of one organization from breaking the revenue cycle of another organization. This doesn’t seem difficult.
What I meant was that multiple organizations/practices accessing the same system. Each organization has it’s own taxid and complement of staff. There are trust issues when you allow multiple organizations to access the same system. Just adding facilities and users to an instance is not going to address that. No member of one organization should be able to break the revenue cycle of another by entering incorrect demographics on a patient cared for by both.
You could setup multiple sites on the same instance, I believe each site would operate independently of any others. @brady.miller Can you confirm?
Shared source code but different databases and sites directories.
1 Like
So the solution to this particular question seems to be setting up a site for each independent clinic
It is very well to set up a site for each clinic, but I think the question is whether you can share the clinical history of the patients. There are many patients who visit different clinics.
Multi-sites is actually pretty clever. You can have several sites/clinic/practices that share the same source tree however are autonomous with all data. I’ve seen one large practice group that is running three clinics on an AWS/Docker instance. It took a little tuning to balance the t2 machine with the right resources but runs well.
Multi-sites really shine when you need to do any maintenance or upgrades as the application is shared across all the clinics.
We have been promise interoperatibilty for many years. I feel like we are no closer to seeing true interoperatibility than we were 5 years ago. A system that can scale to manage all 300 million Americans and 1-2 million providers would help get us to true interoperatability. Implementing a universal patient chart with contributors from various organizations is fairly straight forward. Sharing facesheets (demographics) across organizations would be tricky. A universal chart would be helpful in caring for patients. The ability to share facesheets would be helpful to clinic staff. Registering patient is a fairly time consuming process. The ability to use a facesheet created by a member of another organization would be helpful.
Who promised this to you?
HITECH and to a lesser degree HIPAA. To be honest I have already implement a system to do what I propose but wouldn’t mind seeing openemr do the same to keep me on my toes. I am a physician and developer.
Sounds great. So much to do and resources are always at issue. It’d be more than welcomed if you wanted to share some of your insights with us. I know it was probably a labor of love for ya and i’ve found in many cases, physicians make great engineers.
Hey, we even welcome code contributions 
Labor of love is aprapos. I have spent way too much time on this given my hectic schedule as a physician. I don’t want to give away all my IP as I hope to market this but I would say that key to implementing a system like this a versioning system for facesheets and robust role/permission system. The sharing of clinical notes is easy after that. What motivated me to start this thread is the suspicion that I am missing some fundamental reason why no one else has done this already. I took me only a couple of months and 26k lines of code to implement it.
Difficult for me to way in as to the whys because, I don’t have the clinical experience. This is why physicians and clinicians in general, tend to write good software for the industry. I started out as an electrical engineer who got tired of waiting for coders to write the firmware to run my designs. So I took up programming which helped greatly with my end product.
I bet though there are others viewing this thread that could far better speak to you about the whys. If you want to try and integrate anything to OpenEMR then, i’m your man.
Great speaking with you and thanks for sharing.
1 Like
Firmware is fun. Thinking about real time execution is interesting and all the issues to be addressed is fun. In medical school I wrote software to tune the ECU on my car. If I get some time I will try to review the openemr schema and do a gap analysis on what it would take to to support multi org access. Unfortunately that is a big if.