The Open Source Software Dilemma in Healthcare IT

I have nothing against open source software, as a matter of a fact I am a fan of it, and I use many applications developed under this paradigm for my daily activities, I even enjoy them as weekend projects (e.g. ClearCanvas, PatientOS, Mirth Project, OpenDICOM, etcetera), but I just don't see them fit for the demanding and critical healthcare IT domain.


VistA is not an example of the typical open source application so let's not bring it to the table for this discussion.


CCHIT's step to support the FOSS realm is indeed laudable, Mark Leavitt has proven to be a nice guy, this is a genuine gesture which gives you (the open source guys) the opportunity to certify your Electronic Health Record (EHR) products. But this isn't the gateway to success since you may still have a long stretch to go.


Let me explain why:


Healthcare IT is extremely demanding and it requires discipline a discipline which can only be delivered by highly organized and structured entities.


Companies such as: GE Healthcare, HP, McKesson, MEDITECH, Philips, SIEMENS, and similar others have for many years devoted themselves to the development of highly reliable clinical applications and medical devices. Most of these aforementioned companies have strict product development methodologies and strategies that they have had to implement in order to meet with the ever increasing and strict requirements of the FDA for the verification and validation (V and V) of medical devices and related software.


Many may argue that for the software product development process they don't use the same guidelines as they do for the manufacturing of their medical devices, but knowing how these companies operate, and I have worked closely for and with them for many years, I am absolutely sure that they do.


Maybe the adventurism into the healthcare IT domain taking place by Google, Novell, and Sun Microsystems (soon to be part of Oracle), among many, and albeit this latter one at one point in time built very sophisticated medical image viewing workstations based on their SPARC technology, are probably giving you an over optimistic perception.


In my opinion these companies currently just don't have the culture to fit into this rigorous process. Undoubtedly, they are extremely successful organizations in their activities but just not in this one. You see, this is a different beast.


And maybe you do a great job at creating software, perfect software. But this is not the only piece of the puzzle since there is the rigid documentation process required for clinical product development. Only the documentation process will overwhelm your budget if not drain it completely.


The very laid back and lax nature of open source development is the antithesis of what is required to develop critical clinical applications.


The FDA has formed a working group to determine whether or not Electronic Health Records should be regulated and to what extent if such is the case. Let's say that they do decide to regulate them, and then this would be your demise in healthcare IT. The costs involved in meeting FDA requirements go beyond those that the most daring venture capitalists are willing to risk. Only companies with an infrastructure and logistics similar to those of GE Healthcare and SIEMENS can actually survive and strive in a scenario described here.


My recommendation to you, open source developers, is that along with the FOSS community members, you continue pouring your energies into your projects but don't expect to take the industry by storm. It just won't happen, sorry.


Stick with trying to overcome Microsoft Office some day in the remote future, maybe by 2021, or by creating a user friendly operating system that is not tailored "only for geeks", like Linux is. Did you know that almost 90% of the workstations currently used by clinicians are Microsoft technologies based? You see, open source is "only for geeks" and not for the real world, the real users.


And I am not referring to geeks in a derogatory way since I consider myself one as well. But I do have a professional maturity level to understand that I have to create solutions for real world users and not savvy information technology experts.


Once you break out of your silicon shell then let's sit down and talk seriously.


Thanks for reading!


The EHR Guy