Hacker News Reader

Comment by to11mtm | Hacker News Reader

Also, I've never heard a good thing about Epic...

It's very du joure to talk bad about Epic (and certainly not without reason re: HHS healthcare interoperability proposed rules) but they have been a huge success from a business perspective. The reason why they provide EHR support for the majority of large healthcare system hospitals is that they provide a lot of flexibility to implimentation.

If you want the ability to support some very unique functionality, they will find a way to help support it. In a business that is almost nothing but edge cases (almost all of healthcare is complex, which is why you need 8+ years of training), you need lots of custom software behavior to fit these needs. The UCSF switch from GE healthcare to Epic about a decade ago is the prime example of what value Epic. [1, 2].

[1]. https://www.wired.com/2015/03/how-technology-led-a-hospital-... [2]. https://www.forbes.com/sites/zinamoukheiber/2012/04/18/epic-...

They're in a very different market from Practice Fusion though. There aren't many small/independent practices rolling out Epic deployments...

Now you're right in that traditionally Epic has served larger healthcare systems and tertiary hospitals, (hell they have even been known to turn down customers), but I think there's potentially a bit of role reversal on the horizon. Some Epic deployments have been creeping into community groups through licensing agreements [1, 2]. Probably makes more sense if you want your clinic pop to have greater access to tertiary services. It's something I would keep my eye on within the EHR business space.

Also, my previous comment was in response to no one liking Epic though, so certainly on topic.

[1]. https://health.ucsd.edu/news/releases/Pages/2015-05-06-uc-sa...

[2]. https://ucsdhn.org/providers/physician-network/advantages/te...

Small independent practices are a dying market anyway. Provider organizations are all consolidating in order to achieve economies of scale (especially for EHRs), provide comprehensive services to patients under one roof, and achieve greater pricing power when negotiating with payers. Plus most physicians prefer being employees; they just want to practice medicine without the additional administrative hassle of running a small business.

Its true that running independently is getting more difficult, but that's not "on the market" but on the insidious regulations around insurance, malpractice law, etc.

A doctor can't open up an office and tend to patients with pen and paper anymore, and they have to pay thousands of dollars a month to systems that are truly terrible to be compliant.

*If they want to accept Medicare dollars (and usually private insurance just follows Medicare). I remembered some South Western (I want to say Arizona) specialty clinic that basically didn't accept insurance and had pen and paper charts still.

Medicine is also a team sport now. Caring for patients with complex conditions requires coordination with multiple other practitioners. Exchanging paper records with them just isn't practical or safe. Plus if the doctor has to write a prescription, doing it on paper is so error prone as to border on malpractice.

I will be willing to say the opposite as well. Exchanging records is terrible. If you get records from say, Epic, you might get 300 pages.

Which doctor is going to read 300 pages of documentation for one if his 16 30-minute appointments a day?

When small practices get bought or acquired by hospitals they typically switch to epic.

Epic is far, far away from what us MDs imagine when someone mentions the benefits of EHRs.

For starters, if one could make an EHR that just only attempts to also help practitioners in the trenches instead of just the billing dept. that would be stellar!

So 'the value EPIC brings' is huge... for anyone but care providers, thank you very much.

Sounds like you have a billing department problem.

I think one big issue is that the US subsidized EHRs for a while. So non-US non-billing-focussed EHRs never really developed.

Now those US EHRs are getting dumped on the rest of world, with all of that carefully developed billing tech disabled out-of-the-box.

Sounds like I have a problem with anything making my job more painful.

The actual problem has nothing to do with nation-focused issues. With healthcare liberalization came competition, and therefore search for the cheapest solution. Problem is, in healthcare we have slaves (called residents) and what's cheaper than an educated slave? So actually, even european EHRs follow the 'for-billing' model because it's far easier to just dump the work on the residents.

EHRs are the whip with which we 'encourage' human healthcare providers.

Billing department pays everyone's salary. The billing system in the US is so complex that it eats up a huge part of administering care.

They've been a huge success for the wrong reasons. Its very much a rent-seeking driven business: vendor lock-in, big contracts with government money, etc. What they do would be considered anti-trust if it weren't that government paid them to do it.

You can build an ehr with google docs if it were legal. But it isn't.

Worked for Epic after the dot-com crash, as they were finishing up their Verona headquarters. I will tell you one good thing about it: underground parking. That matters in Wisconsin.

All of the good things I could tell you that I remember about Epic have zero relationship with the software they sell. And none of them could stop me from leaving, because the software was the job. It was all essentially gold spray paint on a turd.

It's made in Visual Basic, and it literally cannot interact with its own versions. Hospitals have different versions of the same system in the same building and cant exchange information.

All EMR's suck though. (note: I made one)

”It's made in Visual Basic”

We shouldn’t feed the illusion that old stacks are inherently bad and that new stacks will magically solve all problems.

Nobody makes this complaint about C++ (which is older) or Java (which is almost as old.)

Sure people do.

I think it's a matter of having the right tool for the job.

If someone wanted to write a modern website in C++, I'd probably gripe about that.

But Java, with either J2EE or Spring? No complaints here.

VB just isn't appropriate anywhere anymore.

Every app has to start somewhere and will after a while be outdated. Whatever problems EPIC has are most likely not because of VB but because of bad management. I work in medical and I see almost every day management decisions that will lead to future software problems but nobody in leadership understands or wants to understand. Have a problem? Just throw the cheapest offshore team you can find at it. Repeat that a few times and you have a completely incoherent architecture. You will get this result with any stack.

I certain wouldn’t advocate using VB for new code. All I am saying is that the real problem is not the stack.

This 'app' charges in the range of 6 digits yearly per provider.

I’ve always heard it was programmed in MUMPS

MUMPS in the back. Front is c# and typescript replacing the much maligned VB.

It's written in MUMPS.

This is only my individual anecdote, but when I switched between two EPIC-using care providers (Kaiser->Anthem/Sutter), once I provided consent they pulled all my records over and 5 minutes later I was talking to a new doctor who had all the info from my previous doctor.

All EMRs suck though.

Yes, but that only works epic-to-epic. If you go to a non-epic place, epic will send you a 100 page PDF.

Some of a competing EMR is PowerBuilder. Hopefully not the one you made ;-)

The front-end is in Visual Basic.

The backend is most certainly not in anything close to Visual Basic.. (MUMPS - using Intersystems Cache as the object store implementation)

That’s because their customer is the institution, not the provider. The experience of the end user is probably not even in their top 3 priorities (similar to other enterprise software like Atlassian).

Reply on news.ycombinator.com