Healthcare IT: Nevermind, risk is too expensive

Politics and religion
1

So while the govt wants to decrease healthcare by trying to jumpstart EMR & EHR records, as IT should decrease a lot of the paperwork, they’ve inserted new penalties in the ARRA/Section 13410 whereby any disclosure of these EMR / EHR records will now have penalites from $500 to $50,000 per occurrance and open the way for civil penalties (the above penalties are levied by govt)

See here and search for “Tiered Increase in Amount”

The funniest part of this is the lowest penalty is if the person had no knowledge they were stolen, so if you’re hacked and records show up, it’s almost a free pass. (my layman’s intepretation)

If however you leave a laptop with data on it, you could face the maximum.

So I’m left wondering how willing institutions will be to go to EMR/EHR records if they’re now going to have to buy insurance to cover fines if any of this data is breached.

On one hand, I’m all for some sort of penalty, as the various breaches over the past few years, while they recieved press, didn’t really amount to any real penalty (AFAIK). OTOH, opening it up for litigation through civil penalties will just make this more expensive.

I think this could have easily been solved by not focusing on data breech, but by focusing on why people are concerned about data breeches:

  1. loss of employment / employment opportunities if employer knows of a condition
  2. loss of insurance b/c of pre-existing condition

If they were to write a new provision / law that said employers and insurers can’t deny employment/insurance because of known conditions, it would have made this a lot easier

2

While I think they are actually looking to decrease costs, I think this is what is going to happen. This is going to require much more employee training, and just slow things down in offices, as every triple covers their asses. Hopefully this is just one more of the death of a thousand cuts that gets rids of the current private insurance industry. It is a huge money sink.

3

Speaking of risk, I’m curious to see how the liability issue is going to be handled in regards to transcription errors on major EMR/EHR pushes.

4

Doctor’s fault, he’s supposed to review it? You sue whoever has the money.

I see a lot of records in a hospital’s EMR and they say “NOT REVIEWED”

5

You’ve met doctors, right? They hate and revile all things they didn’t think of. In general, anyway. They’re among the worst consumer groups to ever do computer stuff for. I would think that their malpractice insurance would include gross idiocy in the handling of digital records, though - I’m surprised they have to get new coverage.

6

How is a doctor in the loop on legacy records? Talking about conversion, not entering new data into an existing system.

7

Not in conversion, but physicians are supposed to review the final output. After all, how else can you trust whether it’s right or not?

I don’t think doctor’s are any more techno phobic than other groups. It’s an age issue. At this law firm the secretary printed out the boss’s emails everyday. He would handwrite responses and then give her a big stack on his way out for her to type back in.

Related: Walmart, Dell and some vendor to sell EMR software at discount prices. 25k first physician, 10k each additional.

They didn’t mention how much each patch will cost. IMHO one big problem with medical software companies is that they are little. Say they have 1000 clients. They then charge 10k for windows 3.1 software and 2k yearly for a simple stupid patch. Pretty ridiculous. A GUI that doesn’t remind me of Wordstar would be nice. Hate to say it, I would love Microsoft to get in the business.

8

Consider your wish granted, then. OK, it’s not an actual EMR but you have to start somewhere.

9

What percent of medical expenditure in the US is on insurance? I understand it to be crippling.

10

They should, in theory, but I see an awful lot of Dictated:Not Read notes from other physicians and in the hospital records.

11

I don’t know how it applies to this conversation but the Heath industry as a whole is one of the lowest paying IT gigs around. I looked into it a few years back and what they were playing for a senior system administrator was laughable. They will pay 10 million from a contractor to design and build a system and higher a system administrator, with no traning for $35k to run it.

12

Like I said before, it’s one of the areas that Just Doesn’t Get It, in part because the doctors, who are ostensibly the ones making decisions, generally try to stay removed from technology and they want to destroy anything that interferes with their schedule and routine. Some are different, and I love those doctors, but a lot of them are (kind of justifiably) arrogant, resist change, and think they can get by doing things like they’ve always done them. As a result, the IT departments tend to get short-changed in the long list of things to pay for.

At least, this is what I hear from my friend of a friend who is from a family of doctors and does IT. He could be lying to me.

13

I work for a healthcare EMR company, and I can guarantee you that not all EMRs suck.

Also, I don’t think that this will cut into people switching to EMRs as it is a law that medical facilities have to be on an EMR system by 2012 (or some such year)

14

There’s gonna be a riot if that happens. I’d say 2020. Not a very good track record so far.

The HIPAA/EDI provision was scheduled to take effect from October 16, 2003 with a one-year extension for certain “small plans;” however, due to widespread confusion and difficulty in implementing the rule, CMS granted a one-year extension to all parties. As of October 16, 2004, full implementation was not achieved and CMS began an open-ended “contingency period.” Penalties for non-compliance were not levied; however, all parties are expected to make a “good-faith effort” to come into compliance.

Take the billing requirements for HIPAA - still not fully implemented 5 years from the final deadline. Today you can still send paper claims to any company you want. In fact, a buncha small payers still have problems with their electronic systems (handling complex names which get autorejected, for example.)

Another was the establishment of an NPI (an id number for health entities) and mandatory use. Today I still need to use the legacy numbers to access Medicare. Except of course, they don’t call it a legacy ID numbers cause that would be illegal. They call it PTIN.

but I see an awful lot of Dictated:Not Read notes

That’s it! Thanks.

15

My hospital is switching to EPIC, a purportedly fancy and professionally crafted EMR, this May. I’ve attended about a dozen hours of formal training and spent about 4 hours on my own in the last month trying to learn about it.

It’s better than the in-house developed system that we’re abandoning, but as far as 2009 applications go, it’s very meh. I consider it like upgrading from 1997 Ford to a 2001 Ford.

There are Warcraft addons that are more sophisticated, easy to configure, and professional. An exaggeration on my part, but not much of one.

This new EMR is going to replace a lot of paper charting, particulary order writing, and I’m very skeptical that it is going to save me any time at all.

One example, I can write with a pen and paper admission orders in two minutes for a simple admission. With the new system, it takes me about 10 minutes. Sure, I’m still new at it, but all the clicking around is very time consuming. There are 4 drop down windows for each medicine. I can write “Tylenol 650 mg po q4 prn” in 5 seconds. The computer has couple dozen Tylenol formulations, multiple different dose, multiple routes of administration, and different dosing schdules. That’s for Tylenol, which is about as simple as you can get.

What it should do is present the most common Tylenol order, followed by the second most common. It should do that for at least a hundred of the most commonly prescribed medications. It should store preferences for individual physician. I’ve been around EMRs for 10 years now, and none of them do this.
.

On most days, I’ll admit anywhere from 1-5 patients. I figure on average, this new EMR is going to cost me at least 30 minutes. Perhaps there will be time savings elsewhere.

Looking at the quality of this product and other professionally developed software, things just aren’t in the same league. There is a goldmine for someone that can develop an intuitive EMR interface.

A paper chart is organized in a particular way for a reason, and so far every EMR that I’ve used (six so far, including the VA) has completely abandoned that organization.

No EMR has allowed me to customize my own charts display, which I also find absurd. I sat in my training course next to a very sweet pediatrician. I’m an oncologist, and we want very different things in our patient’s charts. This EMR was completely incapable to fill that need.

16

Epic, heh. You’re in for some fun. Are they still using the MUMPS-based model?

17

I spent 5 years working for Meditech as a developer (I quit last summer). I spent the last 2 of those years on their new Oncology product. They’re one of the leading EMR vendors in the country, and they stubbornly cling to their antiquated technology (The MUMPS based model, it’s called Magic now).

You’re right about all other EMRs being equally shitty, which is the only reason Meditech is still in business I reckon. If Microsoft, for instance, made an EMR built on real technologies… like an Object Oriented language for instance, and event driven, with a robust browser interface, they’d dominate the field.

Every time I go to a doctor and I see a nurse or doc working with the Meditech system they are constantly frowning, complaining, etc about how terrible it is.

18

I’m betting EMR software is arcane and hard to use for the same reason that a lot of other niche business products are. The requirements for the project as encapsulated in the RFI/RFQ that went to the software devs were quite likely cooked up by some committee combination of system analysis, business analysts, and possibly legal.

They probably wrote some multi hundred page specification for how each little feature would work, and at no point were GUI designers or usability experts consulted. Hence some sprawling confusing system that does hundreds of little thing, all exactly as the requirements specify of course, but overall is a hardly usable mess.

19

Well at MT the problem was coding was expected to start before specs were finalized due to absurd deadlines. That never makes for a good product, as the specs would continuously change during development. It was exacerbated when working on projects with other applications.

The new development division pres had recently tried to curb this by making changing specs a huge PITA once they were approved, but never addressed the problem of ridiculous deadlines and terrible understanding of the needs of physicans causing the specs to constantly change.

Again, I can only assume the competition is even more ridiculously managed because Meditech makes oodles of money (and pays its employees far below market value, which causes all the top talent to leave, which is another cause for the awful products).

20

Not sure what the MUMPS-based model is. I’ll presume it’s an acronym that is shorthand for a software design philosophy created by a committee of at least a dozen people, and is accordingly a marker for only the very highest quality.

If Microsoft, for instance, made an EMR built on real technologies… like an Object Oriented language for instance, and event driven, with a robust browser interface, they’d dominate the field.

I know Microsoft isn’t the most popular company in the world, but I’d give two month’s pay for an EMR product made by them. Same for other people who know how to make something that works in the real world. If it made my life even half as simple as some of their other Office products, I’d be thrilled.