Slashdot Mirror
Switching Hospital Systems to Linux
jcatcw writes "Health care software vendor McKesson Provider Technologies is focusing on ways to cut IT costs for customers, including hospitals and medical offices. The cure is moving many of McKesson's medical software applications to Linux, which can then be used on less expensive commodity hardware instead of expensive mainframes. A deal with Red Hat allows McKesson to offer its software in a top-to-bottom package for mission-critical hospital IT systems."
If this catches on, health care will become a little more affordable. 60% of IT costs is quite a bit of money for hospitals to save.
Seek and ye shall find.
If you use McKesson's software every day like I do, you would be amazed at its expense, sluggishness, and irritability. Lab systems that insist on running on Internet Explorer 6 and resizing to fit your whole screen aren't a big surprise - however mediocre. But mission critical systems that routinely crash with Java errors, can't run reliably remotely, require large IT departments to maintain, are slower and more difficult to use than the tty-based systems they replaced, can't trend labs, can't reliably wildcard search patient names, and die miserably if the wind blows more than 5 miles an hour or the moon is waxing - this is truly sad.
I wish our hospital system could dig its way out of it. I don't think running on top of Linux will help much. See if you can get a screenshot of their software on their website - I can't - they don't promote this stuff to the physicians and nurses who use it - it gets sold to the suits. There's a goldmine out their awaiting some entrepreneur who could really take pride in writing good software of this sort, and though I love Linux, I don't really care what it runs on top of.
Two possibilities: in the process of porting, they have to rewrite all of the bits that call grody Windows bits, such as IE, and therefore many problem bits get fixed . . . or they just write bad code all over again, Linux gets the blame, and hospitals revert at great cost.
RedHat may help though - they might insist on some level of quality / provide some assistance in the creation of software that does not suck quite so much. They have a reputation to maintain, as well as sufficient company-ness to explain to suits that when things go wrong, it is *not* their fault. So, I'll be optimistic about this.
"There are four boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order." - Ed Howdershelt
Well, here I sit waiting for McKesson to refund $1300 of my money after 15 months of hassle. Our oldest daughter takes HGH supplied by McKesson. In 2+ years we've had multiple missed shipments, chronic overbilling, "lost" packages. Last night they sent the wrong vials - 3x the dosage - glad we caught that before injecting our daughter. We have dealt with a lot of medical issues and many hospitals and medical companies. McKesson is the most disorganized company we've ever had to deal with.
Would also be supremely good for the Open Source community in general because of the massive amount of publicity we could gain from this (especially if the code is stronger, better and faster).
My $0.02 AU, ignore at will
Me failed English...
FreeBSD over Linux. If my comments seem odd, this may explain...
from the and-you-care-because-why? dept... a post-LinuxPPC story.
A year or so after the dust had settled following my departure from and the subsequent implosion of LinuxPPC, I got a call from some southeast Wisconsin consulting group. The woman who called admitted she didn't know what she was talking about, but here it was: "Linux development on an embedded PowerPC processor." Apparently one of the people there, whose name I recognized at the time, had said "find Jason Haas! we need to find Jason Haas! He'd be perfect!" I laughed and told the nice woman that I honestly had no clue how to do what she was talking about, but I knew just the person for it... and I gave them Jeff Carr's phone number. She thanked me and we hung up.
Time passes...
A few months later I get a call from jcarr. "Heeeeeeeeeeeeeeeeeey!!!" (this will sound very familiar to anyone who knows jcarr!) It took a few minutes for me to be able to tell what he was saying, as he was rather excited.
Turns out I'd gotten him a job subcontracting for GE Medical Systems. What I like to refer to the military-media-medical industrial complex.
That was 2001 or 2002... and now jcarr's out in the Valley, hooked up with old school Mac developers like Chuck Boich,
and Linux folks too. We'll see about working on The Next Big thing some time soon. (RSN!)
-- haaz.
A while ago I was hacking at parts of that great mess commonly known as McKessons "top-to-bottom package" for almost 5 years. As far as I can tell the "package" is actually hodge-podge collection of applications accumulated over time by acquiring various software vendors which barely talk to each other. In a lot of cases the people who wrote the apps and knew how to maintain those cached out and jumped the ship years ago. Last thing I heard of it this summer entire locations were shut down and routine maintenance moved to Bangalore. As a twist those left to the end had to do some time there training replacements. Personally I am extremely skeptical about their ability to maintain what's there, much less move this all stuff to radially different platform such as Linux.
We're in beta testing with actual patients now and my boss is bankrolling us into starting a company to sell the software and other medical-related IT solutions to local doctors (many of whom have horribly inefficient offices and don't fully realize it). I'm hoping we can expand beyond just local doctors, because it is a huge market and the best anyone else seems to be doing (around southern Ontario at least) is holding seminars to talk about how technology could be used to enhance medical practice someday.
No, the spending wouldn't decrease at all. They would just reapportion the funds to equipment such as the latest digital X-Ray machines. Or the newest CAT scanner. Or the latest robotic surgical nurse.
You get the picture. In fact, in most of health care, that's just what happens already. They spend as little as possible on IT and reapportion the cost to areas of service that will directly benefit their ability to attract doctors and customers and therefore generate greater revenues.
Those reading this might think I'm kidding, but let me tell you this: I once replaced a token ring network with an ethernet network connecting Pentium IIs and IIIs. In 2005.
-- A former healthcare IT worker.
If Murphy's Law can go wrong, it will.
Posting as AC for obvious reasons... Unfortunately for me I work in healthcare IT. But at least I can pass the bad news on to you.
We can argue about how much of healthcare costs are sucked up by IT. But whatever percentage you come up with is likely to be not insignificant. And one of the biggest costs of healthcare IT is the amount of money paid to so-called "IT Consultants".
My understanding is that pure healthcare people don't understand much about IT and since they figure IT is the next biggest thing, they are willing to give money hand over fist to people who have decent resumes in this field who present themselves as IT "experts".
They are throwing their money away. It's really awful.
If you don't believe me, look at some of the so-called IT "standards" documents coming out of the healthcare IT community.
Sure, HL7 V3 is a good, robust yet flexible standard definition. But look at some of the abysmal crap that is being built off of it.
I mean, seriously, read some of these "standards" documents coming from non-HL7 sources. Not only are they inconsistent with reality. They have massive internal contradictions, logical inconsistencies and even simple syntax errors. And this is stuff from organizations that have been around for A DECADE.
Believe me, IT consulting has nothing to do with helping the healthcare industry actually make the best use of modern technology and everything to do with lining the pockets of a few contractors who would be thrown out of any other domain for sheer incompetence.
I work in the Medical Imaging field, a MRI Field Engineer for Siemens, users in hospital want something that works, they are less tolerant of reboots and system hangs.
In the past Unix (SunOS) was the preferred platform, there are actually many MRI systems running on a 100Mhz Sparc processors today, which still do and excellent job.
We've moved to Windows, it's a common interface for users who can learn it quickly. Windows requires CPU's in the 3Ghz range and higher to be effective. Windows also has major issues with Service Patches and hotfixes in the Medical imaging world, all updates have to be QA'd so there is a delay of months before they get applied. Medical Imaging will probably continue to move away from Windows and it's patches if can make an interface easy for the average user who walks upto a system and start using it.
Recently at Siemens Medical http://www.siemensmedical.com/ the MRI systems moved from Windows to Linux (Suse) for the image reconstruction computers (Not at the user console). During MRI imaging data is coming in from the scanner at 10MSamples/Sec at 24bit accuracy up to 32 separate channels, that's a significant amount of data to be processed, having a mouse pointer and a GUI interface is just not needed, Linux just more efficient.
I'm a doctor in a hospital that just deployed an electronic health record system that is slower than the system it replaced - which was slower than the TTY system it replaced - that refuses to search patient names if you can't provide a first initial.
Pay attention here, IT freaks. Notice that the user here (possibly your doctor) says nothing about the OS. This is simply abysmal design and implementation. Unix/Linux/Windows/OSX/Oracle/Postgres/MySQL/MSSQL....ALL could end up thusly. Or all could end up not too bad. Design it right, and build it right. Think about what your user is actually trying to accomplish.
I saw some comments upthread about RedHat this and Linux that...Bullshit. The user interface is (most of) the key. If you screw that, the backend matters little.
Yes, if you start from a stable base, it is easier. But no matter what the base is, if you fuck up the actual program and interface the that user, in this case a doctor or nurse, uses....everything else is irrelevant. They will hate it. And still not care what the base OS is.
The original Computerworld article cited is confusing, but it refers to UNIX mainframes. The most likely educated guess is they're talking about high-end UNIX servers from Sun, Hewlett-Packard, and/or IBM, not what we would generally think of as true mainframes, notably IBM's System z.
Yes, among System z's five popular operating systems z/OS contains a complete and certified UNIX(TM) implementation (called z/OS UNIX System Services). And yes, System z runs 100% GPL open source kernel.org Linux. And yes, OpenSolaris on z will be z's OS #6 before too long, and that's clearly UNIX(TM) too. But I doubt the article is talking about any of these technologies, based on the context of the article. There are not 2,500 U.S. hospital IBM mainframes (the number of McKesson hospital customers cited), for example. Maybe there should be.
Computerworld's editors seem to be on vacation, unfortunately, so their usually good copy editing is suffering, resulting in some gibberish articles. This week they also reported that Steve Jobs and The Woz approached Commodore in 1982 to talk about the latter company selling the Apple II, pointing out that Apple's two founders didn't have enough money to launch the product, worked out of a basement, and the safety and stability of cashing out for a couple hundred Gs was better than the alternative. Unfortunately for Computerworld they got the date wrong: by 1982 Apple was doing just fine, and The Woz was doing Nissan commercials.
Usually what you see is people going from mainframe to windows servers(and expecting the same stability(ha)), so with that in mind, I am happy with the choice of Linux.
Some years ago, when I entered the greek army to do my military duty and after boot camp, I ended up in a military hospital. To make the story short, I worked there for a couple months as an office assistant.
The resident Captain Psychiatrist called me to his office one day to request assistance for some Microsoft Access database he was building. I told him I didn't knew anything about it, but I was going to find out. I had time to spare.
I went through the built in help files and solved his problem and that excited him probably thinking he found some computer genius. So, he showed me his project, which was an extremely basic database in Access for his patients and asked me if I could take over. I am not a programmer (having only typed a few lines in C64 Basic and Amiga AMOS) and not the programming type (I even failed math at high school),but I am comfortable with computers and in the greek army people that know how to use a keyboard and click with a mouse are a tiny minority. Plus, I definitely had lots of time to spare!
So I bought a book on the subject, borrowed his Psycho-bible and sat down and learned about MS Access, databases, interface design and psychology while programming this thing. It became a complicated beast with all kind of diagnostic entries and references and pushed Access and myself to the limits. After about a month, Captair Doctor was jumping around with joy as the project seemed to have a good starting point and lots of potential. He told me that there was nothing good in the market, especially the greek one and that we should develop and market it commercially. We also became friends and even had dinner with his family, a very rare thing to happen, considering I was a drafted private and he a professional officer. I had to abandon the project when my time to leave the hospital came but I found out that it was picked up by another, just like me, drafted soldier who took my position in the hospital office.
Now, isn't that more or less the situation with Open Source? Imagine if I didn't have to use a developing base that sucked (both OS and tool), actually was skilled with coding, had more time and a whole community to take the project from me, instead of letting it die in some dusty box...
I work with McKesson Software in Pharmacy and it is probably the worst I have used since computers were introduced to Pharmacy. The server will make no difference if the software and its support suck.McKesson itself is terrible on support with pharses such as "we know" or "yessss we will get to it and all the while you have crashes, lockups, lost data, corrupted data, lost connections. They do updates in the middle of the day and take down their servers at the time. So do not tout someone solely on the O/S which they MIGHT use. They are Windows Server 2003 and MS SQL Server 2005 still