3 ms·
The magnitude of the data systems migration is usually discounted as well. Insurance companies generally have atrocious IT (it's not their core competency) and
by ethbro 8y ago
The magnitude of the data systems migration is usually discounted as well.
Insurance companies generally have atrocious IT (it's not their core competency) and consequently inflexible IT stacks.
Elimination the layers and layers of bureaucracy also means change / rewritting all the places that's embedded in code.
... and the real kicker? Those changes have to happen on both sides of any party / counterparty transaction. While supporting a normal volume of work.
Small example? ICD-10 code support https://www.cms.gov/Medicare/Coding/ICD10/index.html https://www.cms.gov/Medicare/Coding/ICD10/index.html
- darawk 8y agoThat's a capital cost though, not a marginal cost. Amortize it over the essentially infinite savings on a long enough time horizon and it's basically free.
- ethbro 8y agoAssuming success. Which has a pretty atrocious probability from what I saw (see above reply).
- bunderbunder 8y agoI think, though, that they would find a way to clean this all up toute de suite, if only they were given a proper incentive to do so. For example, if there were real competition in the marketplace (say, by moving to a system where most consumers go to actually shop around and choose their own health insurance company), I expect you'd find insurance companies start to streamline their own processes, both internal and external, more-or-less immediately.
- ethbro 8y agoI think the issue there is risk. Specifically, what is the likelihood of success if I attempt to migrate / change / remove X system? I can't understate the complexity and intricacies. Imagine legally defining all possible medical transactions, then writing software thats compatible with those legal agreements, then running it, changing the agreements, and adding features over 50 years. While the systems being interacted with also mutate. And all of this being done haphazardly and uniquely because there's no one in control of the entire process (Medic-are/aid aside). I'm hard pressed to think of a messier computing system in aggregate. I'd guess even the financial system is more rigidly defined, has fewer players, and doesn't have to map with as high fidelity to real life. ACA provided a fair number of incentives and penalties, re: actual vs industry standard overhead. And it caused change, but there were a lot of failures.
- mmt 8y agoIndeed, imagine a system where some providers are cash-only. They wouldn't even care about that ICD-10 stuff, let alone pay even the cost of entry for the proprietary coding system. Of course, I realize, that were an insured patient to use such a provider, that would shift the insurance claims burden onto the patient (much as is often the case PPO out-of-network claims today), but this could be a point of competition for would-be insurers under your proposed system.
- ethbro 8y agoIf a patient is cash-only, how would their electronic records be stored, if not ICD-* compatible?
- mmt 8y agoUsing any technology that predates ICD-*, including paper. Current regulations may preclude using anything but a compatible EMR, but this implementatino detail is hardly fundamental to the practice of medicine.
- ethbro 8y agoIronically, when I asked an old timer about the IDC field format, I was told they were that way "because that's how they were on the previous paper forms." Point being, using proprietary data formats works up until you need to share data (e.g. labs, consults). And then the forms are the API. So either you're writting a translator or you aren't communicating with the other system. But this is prolly a bit of a tangent. Standardized forms and coding are actually a huge benefit to efficiency. The biggest loss in productivity (I saw) is the back-and-forth nature of the system. Provider submits claim. Insurance rejects it. Phone call is made. Claim is updated. Insurance rejects it for a different reason. Claim is updated. Claim goes through. Note: rejections are usually for technical / procedural reasons, not necessarily not wanting to pay the claim. And that process is different for each insurer. If you really wanted to eliminate hours spent in billing, enriching the "negotiation" phase somehow would be best. But also difficult because you're essentially coding the legal contract between insurer and provider, then attempting to apply it to an actual situation. And have to play nice with the existing systems on both sides.
- LyndsySimon 8y ago> Insurance companies generally have atrocious IT (it's not their core competency) and consequently inflexible IT stacks. I've been working at startups in the healthcare industry for a few years now, and agree. It's not just insurance companies, either - hospitals are just as bad if not worse. It's even worse for hospitals that are affiliated with a University.
- dragonwriter 8y ago> Insurance companies generally have atrocious IT And mandatory IT standards (e.g., those adopted under HIPAA) in the healthcare field are written almost entirely by insurance industry IT.
- ethbro 8y agoEveryone gripes about HIPAA, but from where I sat, it was a pretty sane standard. Admittedly I wasn't in implementation. But it read a lot like an IETF RFC. "Covered entities shall implement network protection." "Covered entities shall not disclose PHI." Outcome-specified, implementation-agnostic. And then set penalties based on whether the breach was accidental / willful / limited / or ongoing.
- dragonwriter 8y ago> Everyone gripes about HIPAA I am not griping about HIPAA. The quality of stone of the specific transaction standards developed by industry groups and adopted as mandatory in regulations implementing HIPAA, OTOH, are another story.