My previous job was in the healthcare industry at one of the top two patient information companies. I left because we had spent the last 8 years working on a pr

My previous job was in the healthcare industry at one of the top two patient information companies. I left because we had spent the last 8 years working on a protocol that was extensible enough to handle all the different kinds of data available. At first it was interesting, but once we passed 100+ individual .xsd files, it got to be a bit unwieldy. Accessing patient data is tough because it's tough to model in a database. It's tough to model in a database because every single vendor exposes data in a slightly different way. HL7, the protocol that is used as an industry standard for this sort of data, has several flavors and everyone implements it their own way. Also, the industry as a whole has no incentive to move forward. Healthcare providers (aka hospital systems) don't really want your data to be portable because it makes it easier to keep you as a customer. On the other end of the spectrum, FDA regulation has scared the big players into spending 5 hours on documentation for every 1 hour of software written, meaning it costs $20-50,000 (depending on the salesperson you talk to) for a hospital to buy a server that handles a few hundred beds' worth of patient data. Needless to say, I'm glad I left.