undefined | Better HN

0 pointskendallpark7y ago0 comments

> Often the larger EMRs aren't even interoperable with themselves, each deployment is an island unto itself.

This is what I'm talking about. We're so far away from actual interoperability it's not even funny.

> The time it takes to request and receive records on a patient is measured in days not minutes.

Yup.

> While interoperability is something to strive for it's not the thing holding the system back.

Not sure I follow you. The siloed data seems to be a pain point for most docs. Patient is admitted for evaluation of X, Y, Z concerning symptoms. Patient was admitted to a different hospital a week ago for the same symptoms. Different hospital is on a different system, so I can't get the patient's history from that visit. Guess I need to have the records faxed over tomorrow morning. In the meantime, let me go order a bunch of tests that are probably totally redundant to what was run last week, but I can't get that data right now, so gotta run them again. Huh, that's a interesting x-ray. Wish I had the prior.

0 comments

pg_bot7y ago

The problem is that the fax takes until tomorrow morning, not that it comes as a fax. You could get the information over the phone, carrier pigeon, or smoke signal it doesn't matter the mechanism. (the latter two are likely not HIPAA compliant) The systems could be completely interoperable, but you would still have the same bureaucracy that slows everything down.

Just because the systems can talk to each other, it doesn't mean that I can just automatically grab whatever patient data whenever I want. There are rules and procedures you need to follow when dealing with PII. If you are cavalier with these systems you will be fined or sued into oblivion. Time until transfer of data is the hard part, which is independent of interoperability.

There are bigger fish to be fried in this space, so if we are triaging healthcare technology I would put this low on my list of things to take care of.

kendallparkOP7y ago

> Just because the systems can talk to each other, it doesn't mean that I can just automatically grab whatever patient data whenever I want. There are rules and procedures you need to follow when dealing with PII.

Yes.

> Time until transfer of data is the hard part, which is independent of interoperability.

This depends if you consider dealing with the human and/or legal factors as part of interoperability or not. If the technology exists but the regulations or human factors don't allow for timely communication between systems, the systems are not interoperable. When we talk about "implementation" of a system in healthcare, we're not just talking about the setting up the technology, but changing hospital procedure, training, getting it approved by various committees, navigating legal issues, etc.

When we talk about 'disrupting' healthcare, most people don't realize that making a sexy piece of technology is the easiest part. FDA approval is effort-consuming and expensive, but plenty of health tech start ups and company initiatives have also done what Apple has done. And hence why I don't consider Apple's ability to get de novo 510k on a cool wearable 'disruptive' or a 'tipping point.' It could be influential on the consumer side of things, building a bridge between consumer health and healthcare. But how much does it touch the massive, unwieldy machine that is the healthcare system?

> There are bigger fish to be fried in this space, so if we are triaging healthcare technology I would put this low on my list of things to take care of.

No one is saying interoperability is the biggest fish. But it's a pretty damn big one. Perceived fish-size will likely vary from specialty to specialty, big urban hospital to rural clinic to private practice. What would you consider the bigger fish?

j / k navigate · click thread line to collapse