If I'm understanding your question correctly, the single biggest obstacle in healthcare informatics today is interoperability. The industry grew up in a pretty ad hoc way and the result is that closed-off silos of data are pretty much the norm; every vendor has their own ideas about how data should be handled, and getting different systems talking to each other is a full-time job (my job, coincidentally).
HITECH and the ACA have forced the industry to start making meaningful steps toward real interop, but I'd say we're at least a decade out from it being less of a nightmare than it is now.
I agree. Integration is exceedingly complex. Data is stored and interpreted differently in every system, even if they share the same EMR, due to configuration differences in workflow, data setup, software versions, etc.
You can't exchange data without thoroughly understanding the clinic workflows that generated it or will be using it. It's all time-consuming and hard.
I work on a patient portal consuming data from the EMR, and even that is tremendously complicated to present medical data safely and correctly to a person.
--
Sibling comment mentioned wanting to work in Health IT. The big market problem in healthcare is small companies doing good innovation (usable patient-side workflows, modern clinical tools, shiny things) running into the consolidated, massive EMR systems. The first question when they approach a healthcare system will be "Are you integrated with Epic/Cerner/whatever?" and if not, they will be sent away. Or be ready to embark on a very long, slow process and integrate deeply into workflows, data APIs, etc.
When a system consolidates their EMR (driven by real needs but also Meaningful Use incentives), it forces standardization and special one-offs become much harder. Getting a doctor interested in using a new device or software means working within the whole EMR - the staff doesn't have the time or leeway to go use tools that don't integrate, just because a doctor really wants it. That doctor needs to align large groups, get agreement, and it's going to take a lot of time and money.
It all comes down to interoperability/integration. Building cool stuff in healthcare is really easy since most of the tech in use is outdated and slow moving. But interoperability - required to sell into healthcare systems - is really, really tough.
The irony being that those tightly inegrated solutions tend to be some of the worst offenders in terms of being a nightmare for interop and walled-off silos. But they're popular because the pieces they do offer generally work.
And totally agree that the opportunity is there for highly targeted applications that cater to specific healthcare niches because the downside of the huge top-down systems is the fact that they're more generalized. But you have to be able to integrate them into that larger EMR environment for them to be realistically useful.
That said, the facility I'm at now (midsize, ~400 beds) took a best-of-breed approach and... well, there's a reason I say current interop is bad. It's appealing on the clinical side because groups like surgery or the ED or even endo get to run software designed to cater specifically to their needs, but the backend integration ends up being a huge exercise every time anything changes.
I thought it was more than data interchange. Like more than inefficient commputer programs/interfaces, costs, lack of competition from entry barrier due to high amount of legislation/regulation.
I'd love to work in Health IT, what's your company's name ?
HITECH and the ACA have forced the industry to start making meaningful steps toward real interop, but I'd say we're at least a decade out from it being less of a nightmare than it is now.