I want to take a moment to make sure we are all on the same page here with the business of health care reform. This is inanely simple. When it comes to health care, keep doing things the same way. It’s a proven business model. Here are a few specific pointers.1) Don’t Involve ConsumersThis is really critical. Do *not* ask consumers what they want. Whatever you do, don’t ask consumers to define “meaningful use.” These kinds of rhetorical debates are best left to academics and bureaucrats inside the beltway. Every time a consumer mentions anything resembling meaningful use or a “personal” health record, change the subject immediately.2) Act Like Privacy Issues are InsurmountableThe possibilities here are endless. The more you can distract consumers with potential privacy issues, the less they will pay attention to the ways in which they would benefit from having true ownership of their health care data.
3) Don’t Learn from Other IndustriesDon’t bother reading that book by Clay Christenson. He has spent a decade studying the inefficiencies of the health care system. Inefficient by whose standards? Let the academics put their two cents in when it comes to meaningful use, but don’t listen to any of that Harvard B-school innovation nonsense.4) Act Like Open Source Doesn’t ExistFortunately, most people have long forgotten that once upon a time, software was free and/or inexpensive. They continue to blindly support proprietary software, even during a prolonged recession. They even purchase new computers to run this bulky, expensive software!This ties into the next point. 5) Think Short TermThe time to think through any major conceptual problems is not now. Come up with brilliant, yet strangely expensive health care solutions (remember, they must be proprietary). Don’t worry about long term sustainability or stupid things like sharing your source code. Having proprietary solutions is exactly the leverage you need to maintain your involvement in perpetuating, I mean solving, the problem. This is advice you can (both literally and figuratively) take to the bank.Oh, yeah, speaking of the bank, by the time tax payers realize what you’ve done, you will have already deposited your bonus check and had a fabulous spa treatment.
Cindy Throop is a University of Michigan-trained social science researcher specializing in social policy and evaluation. She is one of the few social workers who can program in SAS, SPSS, SQL, VBA, and Perl. She provides research, data, and project management expertise to projects on various topics, including social welfare, education, and health. www.cindythroop.com
You are a bad, bad cat owner. And a bad, bad prognosticator.
Thus endeth the lesson.
Come on, ePatientDave. The cat owns *me*. How many cats do you know who have their own web sites called “I own peeps dot com”?
Great post yesterday: Forrester’s five phases of open-source success/open source adoption – http://bit.ly/2liz5 (I would love a framed copy of the graphic, hint, hint.)
In the whole discussion about EMR and their “meaningful use”, I feel that EMR use is seen as an end and not as a process parameter.
The question should not be: what constitutes “meaningful use”, but rather: what specific improvements can be made (with the help of EMR)?
Also, the question who the consumer is in HC is unclear in this heavily ironic post … I assume Ms. Throop means the patients. I would argue this is an ethically an economically untenable premise. Health care MUST be patient centered (that’s its purpose), but not in the sense of a “the customer is always right” attitude.
Cindy: Right on. great article. I love cats.
RE: Open Source. Often VA VistA is touted as THE open source solution. While it is true that VistA can be downloaded from SourceFordge, there are no instructions and there is really no open source community around it. Meanwhile companies like Medsphere go around marketing as an “open source” solution because they are using VistA as a base, while they are really selling very expensive proprietary software annd services. Nothing against proprietary software companies, but please don’t try and get free press for being something that you are not. VistA is also written in MUMPS, an arcane language from the 1960s. MUMPS is so obscure that an open source community will not develop around VistA. This is why VistA is so expensive to use/maintain despite it being “open”. (I put open in quotes for a reason.) Most developers have never even heard of MUMPS. No modern books on the topic..books no college classes….even COBOL would be preferable. And this is a way forward?
Politicians (Rockerfeller D-WV), policy makers, and journalists are usually not “in the weeds” enough to understand these distinctions. Try and download Vista and install it…yea exactly! Its not an open source project by any Linux geek’s standards. Apache and Linux are real open projects because there is plenty of instructions, and you can have the system fully up and running in just a few minutes. We need an EMR like that…at least for free clinics.
Lastly, I’d argue that it is more important that the formats (i.e. standards and code tables such as HL7, CPT, SNOMED, CCR, etc) be readily accessible by the everyone and in the public domain than it is for everyone to use open source EMRs. This is America and there is nothing wrong with making money on software, but the artificial monopoly created by the death grip on formats maintains the status quo, quells competition, and keeps the prices of health IT sky high. Patients, providers, payers, taxpayers, friends, and countrymen….now is the time for action! We need to elevate this debate. -Alan
Cindy: I think the cats are into this bag, big time. Lots of loot out there, if you know how to litter the floor (the House of Representatives floor, that is.)
Great post. Should be quoted often.
Regards, DCK
Don’t forget the tactical level: Hire experienced IT people then pay them dirt, refuse to make any capital investments, hire inexperienced assistants and overwork them with 24X7 zero-tolerance operations, 365 day-a-year on-call cycles. no merit pay, and no training. Extra points for screaming at a technician because your Blackberry doesn’t work in your exclusive, $3M cave.
Ahem. My point? Open Source is nice, so long as you hire and properly lead the staff you need to bootstrap and maintain the system according to your own policies and workflow. That includes both geeks and Clinical Information people. And you have to LISTEN to the advice and EMPOWER your experts.
Oops. that’s only if you DON’T want to fail…
Hahahaaaa; I love this post! I hate to be sexist, but women are sooo good at being direct and sarcastic! We have a female sports columnist at our local paper who is just the same; she’s the only one who can see that the owner, not the players or coaches, has been the 15 year problem with the Washington Redskins. Go, women!! ((:
Happy Memorial Day, everyone, including men!
Yes, Rob knows what I’m talkin’ about.
And I’m putting the cat in charge of identifying the wolves in sheeps clothing. Good point, Alan!
If you want to debate this in person and you plan to be anywhere near DC next week, meet Cindy, @modulist (aka Claudio Luís Vera), and I at the Nerds and Wonks in NW Tweetup next Wed. May 27, 6-8:30pm at Vapiano.
See: http://www.eventbrite.com/event/346826367
There is nothing new in statement that consumer does not know what they want looking forward. But in the present they do.
The question is not that for useful use of the money. The question is whether a business knows what is needed. And is it capable of making a good and useful product based on that understanding.
If you look around, you will find that anwer is more often not! After years of HIT development, the use of it is limited. People blame it on the user. While there may be some truth to that, but a product is measured not by how incompetent the users are but by how good the product is in helping every type of user.
So now, we are giving these companies money to do what they have not been able to do. What I surmise is lacking is not necessarily the idea and the IT completence. It is the competence to take an idea and create a useful product where the focus need to be. Barring that, we will be having same discussion few years and few billion dollars later.
Before anyone gets personal, let us take Microsoft as example. It is said that they hire the best..they have been in the window business for decades and still most of us find it crashing alot….
rgds
ravi
blogs.biproinc.com/healthcare
http://www.biproinc.com
PS: On an unrelated note, I am creating a database of ideas (actionable and specific) that could be used for transforming healthcare. I would invite those with thought to add it to the blog address given above. Thanks. rgds Dr. R.K. Pandey
This whole era of healthcare modernization includes consumer involvment. It’s probably one of the major aspects of it. This has been recognized by all of the current bodies in healthcare IT (e.g. HIMSS, CCHIT, AMA, HL7, IHE, HITSP, etc.) There are many that it creates a balance. It’s not just handled by a few bureaucrats and academics.
Consumer involment has been taken into account in many senses. The progression of EMR to EHR (we should take HIMSS definition of these terms into account) will include the progression towards the PHR which is an instrument that is managed by the patient.
Yes, the privacy aspect of sharing patient information is a big concern. We have the laws and technology in place. What has to be done is determine the best approaches. There will be many lessons to learn in this endeavor and most of us are aware of this.
The HL7 Guy
With respect to open source. There are very few options in healthcare.
I agree with Alan Viars comment with regard to the VA VistA and Medsphere.
What guarantee is there with open source. We have just witnessed that if there is a succesful open source project it will just be bought out by a corporation. Look at Sun purchasing MySQL and now Oracle has both. What’s the future of MySQL?
Open source has been just a tricky way (dishonest in some respect) of leveraging free software developers from around the globe and then once having a succesful product obtaining a lucrative profit from this free work. To me it sounds more like modern slavery.
Cindy, your opinion is taken into consideration but I don’t think you have enough experience in our industry to make the statements you have. But they do give us an idea of what many may be thinking about this.
Thanks,
The HL7 Guy (a.k.a. The EHR Guy)
I am not suggesting that open source is some kind of panacea for health care reform. The concept, however, does embody the type of transparency that is needed to use HIT money effectively. I may not be an “expert” in “your industry,” but I once witnessed many hundreds of thousands of dollars wasted on an “HIT solution” that was conceptually flawed. Okay, that was only *one* experience, but it was a heck of a lot of taxpayer money flushed down the toilet.
Let’s not waste a boatload of money. Let’s give open source a fair chance. And, yes, let’s make sure that open source programmers don’t get screwed by laissez-faire capitalism.
Open Source is just a red herring. For some reason it equates in some people’s minds to “free stuff”. It is not. It never has been.
Sure there are good open source tools and platforms that a technology company can use for free, in order to build consumer goods that are sold for a price. Open Source does not translate into free stuff for consumers.
I think you should just ask for a free EHR, period. Maybe all those techie slaves that HL7 Guy is mentioning will oblige. Otherwise somebody is going to pay for creating this Open Source EHR which will probably turn out to be something like Vista and every physician will need to pay to technology companies for custom built modules that will probably come at the same price as non open source EHRs, not to mention support and maintenance of all these customizations.
Looks to me like a fire to frying pan solution.
Cindy,
You are correct when you say that many HIT projects fail.
Most of us in healthcare information technology use the same technologies that other industries use (e.g. financial, insurance, etc.)
Many industries have learned from healthcare interoperability techniques. We have the most advanced integration tools you can find. And yes, we utilize modern technologies as well such as: SOA, ESB, Web services, etc.
We have to deal with very complex data that is not found in other verticals. And our data has to be extremely accurate because we have lives in our hands.
Healthcare has the most diverse and comprehensive, albeit extremely complex, set of interoperability and coding standards of any industry.
There are many reasons for the failure of these implementations that you mention and which have very little to do with the solution being proprietary or open source.
Unfortunately and statistically, open source and in-house solutions have failed the most.
Big vendors such as MEDITECH, McKesson, SIEMENS have had success in large institutions and integrated delivery networks (IDNs).
One of the great reasons of failure of HIT projects is because of physician apathy.
Until we overcome this behavior very little progress will be made.
Some physicians claim valid reasons to not adopt technology and others simply don’t want it for no valid reason.
Most healthcare IT related organizations know this and this is why we are trying, as an industry, to use the carrot-and-stick approach to get them on board with technology.
We have been patiently working with hospitals and physician practices for decades. This is the best thought out plan that we have, as a collective, come up with.
Even though you may not have experience in “our industry” you have certainly a talent as a blogger to initiate a debate with your opinion.
The HL7 Guy
Concur with HL7 guy. “One of the great reasons of failure of HIT projects is because of physician apathy. Until we overcome this behavior very little progress will be made.”
Regrettably this is accurate and one of the reasons why the good-intentioned allocation of $30 billion (more or less of funds) will fail. Zero incentive for “users” primarily physicians to dedicate the time and effort that are essential to finding an appropriate system, then learning how to use it. See my comment to Dr. Kibbe’s posting.
The interoperability issue – along with so many other issues raised by entrenched economic interests in the healthcare industry – is a red-herring. Interoperability has been an issue since the inception of digital technology, not to mention since the origins of language use by humans. Anyone ever heard of Tower of Babel, whose construction was halted due to confusion of tongues (aka problem with interoperability) of those doing the building? That issue has been resolved.
Exchanging data from one computer system to another is not that big a deal. Yes, the interpretation of data fields can be a problem, but protocols such as HL7 are designed to correct for that.
Again, I am not suggesting that open source is “the” solution, nor is it a free solution. The UK has just begun to implement open source EHRs, so it remains to be seen whether the open source folks can do it any better and/or cheaper.
There will be thousands of fragmented solutions to the complex problems facing the health care delivery system. The real question is what will we be doing differently in the future? Humans have an incredible tendency to repeat the same behaviors over and over again. What is different *today*? It has taken a prolonged recession and acute health care crisis to get some decent conversations going between all of the stakeholders.
Judging from the above conversation, we have a little way to go before *all* of the stakeholders can engage in constructive conversation. What I know for sure is that consumers/patients/people/tax payers will be playing a greater role in deciding what things like “meaningful use” mean. If you want to belittle this group of people and/or alienate people from other industries and perspectives, don’t let me get in your way.
If HL7 (or proprietary software) is a piece of the solution, I’d like to hear about it, preferably from people who use their real names in public.
6. Never communicate with people on the workfloor (with truckloads of hands-on experience)
7. create ambiguity
8. something that works for centuries: divide et impera
consolation: same conclusions in Europe
HL7 is just a standard for exchanging data. It stands out because it has been in existence for many years and it is used successfully by almost every HIT application. It is not the only standard. X12 and NCPDP are in wide use as well, and of course XML is everywhere.
There is no shortage of standards. They are accepted and used every day. Presuming that there are none, or that they need to be reinvented, at tax payers expense, by some committee is a fallacy. Let alone being certified by self serving, self appointed bodies.
Whether the software itself is proprietary or open source is irrelevant if everybody communicates using these standards and agreed upon terminologies, which are out there as well.
As to physician apathy. I don’t think it’s just that. Physicians in private practice are small business owners. They have a process in place for conducting business that allows them to stay afloat – paper. When we ask them to adopt a new process, we must show how this new process benefits their business, or at least how it’s not going to bankrupt them altogether. Granted the public good is important, but we usually do not require segments of the economy to commit mass suicide for the public good.
Yes, patients/consumers/tax payers need to benefit as well, but ignoring physician needs is a non-starter.
Patients, or consumers as we are called now, are definitely at the center of healthcare, but you cannot have much care without doctors either.
Of course FOSS is the solution. The F is free in both the no-cost sense since there is no licensing cost associated with the use of the software and free in the sense of free of copyright restrictions other than to insure than anyone who uses the source code as a base is obligated to offer any enhancements license-free to anyone else who wants to use the enhanced version.
Three existing more or less usable open source EMR products are Indivo (which is available in a PHP and a Java version, although presumably it needs significant enhancement to have relevant applicability), PatientOS and openEMR. There are likely others available outside the USA.
PatientOS seems to me to have great potential and to my knowledge is primarily the product of one individual. With some attention and some modest funding applied to its development (as I note elsewhere both for software development and equally important for development of user manuals) it could easily be a fully usable product for hospitals and physician practices alike. It is almost at that stage now.
Alternatively the federal government could “strong-arm” under the guise of furthering public welfare a company like eClinicalWorks to release some version of its product in open-source. I suspect the principals in that business would be open to such a suggestion if it came from the right source, e.g. someone with the authority of Dr. Blumenthal for example.
Cindy,
If you go to my blog at http://www.hl7guy.com you will discover my “real” name.
There is nothing wrong with using a handle on the internet. By no means am I anonymous.
Thanks,
The HL7 Guy (a.k.a. Michael Planchart)
Thank you, Michael. When I read your first comment, I went to the link and did not see your name anywhere. It gets kind of creepy talking to a person without a name
My name appears on the articles. Read the insights. I started using the handle on twitter and other social sites and most people like the hl7 guy more than “just mike”
The HL7 Guy
Wendell, what exactly do you mean by Free? Is it free to the doctor, or is it free to the technology company that uses it as a base for further development?
Who is paying for training, implementation, data migration/extract, faxing, customization, upgrades, maintenance, support, hosting (if necessary), documentation, interfaces, clinical content licenses (drug database, patient handouts, CPTs, etc.)?
Basically, what is the cost to a physician to deploy and use one of those solutions to his satisfaction?
I concur with Margalit that HL7 stands out because it has been in the healthcare IT domain for many years, approximately 20.
We also have to recognize that the role of the HL7 organization has been a commendable one.
Many years ago their was very little need of interoperability. Technology wasn’t mature enough to be extended throughout the entire healthcare realm and applications were very specialized.
It is also a fact that hospitals have silos of data not caused by technology but more because of their own cultural behavior.
And believe me when I say that hospitals are one of the most political institutions in the world. Even more than congress. I swear.
Hospitals would implement “best-of-breed” solutions in the different departments (e.g. Radiology, Lab, Pharmacy). They weren’t concerned about interconnectivity. They would handle the data exchange between departments by manual methods and most still do!
HL7 created an open standard that would help alleviate the interchange of data. But since there were so many different vendors and applications it had to include many optionalities so that it would be adopted. This flexibility made it also very complex and large.
But we don’t have to worry about this anymore since IHE and HITSP have come along to help use the different standards through harmonization. They achieve this by creating profiles that represent common use cases.
Everyone single of these organizations are open. HL7 has a membership fee that is too high. I think this revolution in healthcare IT will force them to rethink their membership model.
There is no such a thing as open source standards or proprietary standards. Standards should be open and period.
Now, let’s keep an eye on Microsoft since they have created an open community for interoperability. They have also realeased the HealthVault platform for anyone to leverage in an open way. I have just begun reviewing this and once I have a better understanding of it I will post a blog of it.
Thanks,
The HL7 Guy (a.k.a. The EHR Guy)
The cost to the user, presumably the medical service provider, but also a patient, is time and commitment.
That is a big cost, but it is an implicit cost as is any training/learning cost. But it is a cost that is incurred whether the software used is commercial or FOSS. Two traditional weaknesses of FOSS product are (1) too great a focus on technical software issues at the expense of the issues having to do with the ultimate purpose of the software and (2) a general lack of high-quality documentation. In both areas that is where some public funding could do wonders.
The FOSS model has worked forever in software development and FOSS software happily co-exists with commercial software. HL7 guy will be familiar with the open-source HL7 interface engine Mirth:
Mirth can listen to and send HL7 messages and connect to a variety of protocols:
* TCP/MLLP
* Database (MYSQL, Postgres, Oracle, MS SQL, ODBC)
* File (local file system and network shares)/PDF
* JMS
* FTP/SFTP
* SOAP (over HTTP)
An open architecture allows for the easy addition of custom and legacy interfaces. (from Wikipedia).
An excellent commercial interface engine and messaging toolkit are produced by a company called iNTERFACEWARE. The two products happily co-exist. There are many examples.
There is no question that the $30 billion in government funds will flow through to commercial software vendors for licensing costs and to commercial installers to implement those commercial systems without sufficient commitment on the part of “users” to effectively (i.e. productivity-enhancingly) use the software installed.
There is an opportunity now given the deferral on disbursement of funds for two years to create a widely publicized and well-supported (but at minimal cost development cost nonetheless) FOSS application or range of applications depending on need within that time frame.
All commercial vendors know that they overcharge for their software and they all know that once a customer is committed to the software that they will do everything possible to charge for any imaginable add-on to the software, e.g. force a physicians’ practice for example to go through the vendor’s computers to relay and received HL7 messages from outside parties or try to force practices to use the vendor’s “client portal”. All activities that can be accomplished cheaply, quickly and well through alternative means.
Wow. I have learned a lot in one day. I would love to sum up what I’ve learned today (the other days don’t matter so much, ha ha) and find a way to explain it in a way that non-techies and policy wonks can understand.
Perhaps I am delusional in my optimism, but I would like to think that there is still time to provide some refined feedback to the folks inside the beltway.
To that end, today’s post was partly in preparation for the (strange?) Tweetup next week, although I officially expand the invitation to include proprietary software peeps as long as they promise to play nicely and/or have a good sense of humor.
I am so tired I can’t see straight, so you’ll have to look up in Susannah Fox’s comments (up, higher up) to find the DC Tweetup information. I promise to provide updates for non-DC folks who are interested in hearing about how it goes/went.
Cindy: You rock and you are right. I don’t know your background or experience level in this market but based on your analysis I think you are a thought leader in this area. New thinkers like you should be welcome. I think @Modulist said it best when he said the old guard attitude/response is FUD.
HL7 has is NOT an open standard. You have to pay a lot of money for it and join the club. Can we really say HL7, as a standard, is a success? Really?!? There is very little we can say about the American health care system, at this point, that is a success. Health IT is generally not interoperable and is wildly expensive. If that is success, then what is failure? Let’s call a spade a spade. Health IT needs to take a lesson from other industries.
While well intentioned I’m sure, no one can say HL7 has achieved any noteworthy level of interoperability. Some would argue that it is spaghetti code. I have to temper my comments and point out that health care is a complex domain. Its not the format that concerns me as much as the fact it is closed from public view.
Are we denying patients access to their medical information through the closed standards? Its like saying, “Here you go patient Dave. Here is your record in PDF format, but gaining access to Adobe Reader costs $5,000+ and you have to join a club too.” So the patient is left with a binary format that when they try to open they find only garble. Is what American citizens get for $30B? Is this patient empowerment and health transparency? Does HL7 and other closed formats violate ARRA/HITECH?
RE MySQL: MySQL is still free and you can still get the source code. My understanding is a free “community” version of MySQL will remain…just like may open source projects do.
Sure there is a cost with open source software….a “total cost of ownership” as they say. I don’t think anyone is arguing that, but generally FOSS can save an organization money. Case in point, using MySQL over Oracle can save a ton of money and in most cases MySQL is more than sufficient. The great thing about MySQL (and Redhat and other projects) is that you can choose to buy commercial support if you want it or need it. It really depends on your staff and other parameters.
-Alan
Twitter: @aviars
Wendell, I am very familiar with both iNTERFACEWARE and Mirth as well as Cloverleaf and other engines.
I can see very well how a software engineer can happily download the open source EHR and interface engine along with all sorts of other open source tools, run them all on Linux, enhance and tweak the code and even write new modules on top of the existing ones, with enough time and commitment.
What I cannot see is how a physician does all that with no financial expenditures. I’m sure that there are much more expensive solutions on the market, but I’m equally certain that there are equally priced ones and even some that are cheaper than an open source solution, many of which are better quality and more user friendly.
Bottom line, I don’t see why the government should invest money directly into software development, open source or otherwise. If they want to provide incentives to physicians to buy software, let the doctor choose what he wants to spend the money on – FOSS or commercial – based on value, preference, or whatever else drives free markets.
My understanding is that providers can choose whatever they want, whether it be open source or proprietary. The only stipulation I am aware of is that it must be “certified,” which apparently means it gets a stamp of approval from CCHIT. Both open source and proprietary solutions can be CCHIT certified.
Alan, using HL7 does not require you to join any clubs. There is no entrance price and no royalties to pay.
As to its success, how about every single lab order and result sent electronically in this country, every day? Millions of transactions. How about every ADT (demographics) sent every day between thousands of disparate applications in hospitals and private practices? Many, many millions of those transactions. And that is only some of what HL7 is being used for.
HL7 is not spaghetti code. It is not code at all. It is a specification on how to arrange data elements so the other side can read them.
HL7 has nothing to do with providing data to patients. It is meant for communications between applications.
My SQL is a lovely little database. Maybe nice for some and totally inadequate for others.
FOSS can reduce the cost of software development for technical organizations that have options to buy or not buy support. It only minimally impacts the price that consumers pay, unless the physician has a degree in engineering and buys raw FOSS products and puts them all together in his “spare time”.
There are many things that are wrong with our helthcare system, none of which can be blamed on HL7 and none of which can be alleviated by FOSS.
All versions of HL7 messaging standards are freely available except for the latest.
But most likely you will not need the latest version of HL7 for a product you are developing in the present moment.
As an individual it costs $600 to join HL7 and as an organization it is $1100 if the annual revenue is under 1 million dollars and it increments further as the annual revenue is larger.
HL7 is a succesful messaging standard. No other vertical domain has a messaging standard that can be considered worthy. I have several bank accounts from different banks and none are interoperable. I have insurance policies from different companies and none are interoperable. What are other industry are you comparing healthcare with? What interoperability are you talking about? What standards are you mentioning?
The only industry that has ever created standards for interoperability is HEALTHCARE!
Don’t compare apples with mangoes!
The HL7 Guy
As Margalit indicates HL7 is not code. HL7 is not a programming language! HL7 is a messaging standard that was invented way before XML was ever thought about!
HL7 has made things easier in order to exchange patient and medical information.
Alan, give me an example of interoperability from an industry that you know of.
The HL7 Guy
Cindy,
You should start a blog about “Meaningful use”. You seem to stir passions in healthcare IT folks!
The HL7 Guy (sorry, Michael Planchart)
Alan,
Your opinion seems biased
http://www.eventbrite.com/event/346826367
Thanks,
The HL7 Guy
If HL7 is indeed open as you say then please please provide a link here. I’ve been searching to no avail. All I find is tools thats say:
“…These tools carry a license that restricts their use to activities that support the development and implementation of standards by members of HL7 and of the HL7 International Affiliate organizations.”
I misquoted the price. My bad. Even $1 is too much. If the latest is not open/free then I guess its closed. Its a membership, not a club, and in general patients aren’t members.
I understand HL7 is not a programming language, but many complain that the “schema” itself is not well defined but this is improving in recent iterations. Its still a little more complex than it needs to be IMHO. CCR, although for a different purpose (PHR), it is XML and is cleaner in my opinion,
I also understand HL7 (except the CCD) is not meant for patients, but the public has a right to all formats related to health care. Does that mean that patient’s don’t have a right to it? No formats should not be sanctioned by the government unless they are in the public domain. I wonder if this could fall under FOIA? Probably not, but I think the public has a right to know. Let’s all work together to create a transparent health system.
In the end this will help providers payers government and patients alike.
Why isn’t there a download button on the front page of the HL7 site? Please prove me wrong. Post the link. I will eat my words.
For the record I will support MUMPS. HL7, Microsoft, what have you for any customer who asks. These are just my opinions, but the customer is always right even when I personally disagree.
I must admit I am biased. I am biased for being a hawk for patients’ rights, biased for the public’s right to access to health information, and biased towards fixing our very broken health care system. I hope we can all agree that patients do have the right to their medical information in ALL forms and formats.
Do you think the public/patients should not have access to these formats? If so, why?
Our health care system as a whole is failing and this is just one piece of the puzzle. I just want to fix it mor my generation and for future generations.
There are a growing number of people out there who share these opinions and we are starting to ban together to fix what is broken, even if it is painful and displaces some.
America: Write your senators and congressmen!
-Alan
You IT people can talk about interoperability all you want, but when my elderly relative had 2 hospital admissions in hospital A in two weeks, and then was taken as an emergency to hospital B last week; his records from hospital A were not available to the doctors or staff of hospital B. This is interoperability as we in the trenches understand it. Who cares about HL-7 or all the other stuff? Just give me the $#@&&!!! records!
Is this what you mean by physician apathy?
Alan,
You can download all versions, except the latest, of the HL7 standard here: http://www.hl7.org/Library/standards.cfm
Anyways, I don’t think people care too much about the latest. It hasn’t been significantly adopted and many experts are hessitant to implement it since it is overly complex. It was bornt while technology was changing at a rapid pace. 12 years ago it followed the XML format but it couldn’t evolve as fast since it was very specific. Plain vanilla XML was given much more opportunity since it had a broader audience that helped tailor it much more.
There are many sources for HL7 in the world and one is the Australian chapter site: http://www.hl7.org.au
And the standards and technology have absolutely nothing to do with the broken healthcare system.
The system is broken from fundamentals and not from processes.
Blame the insurance companies for the broken system, the people (current industry leaders) who benefit economically from keeping it the way it is, the fraud and corruption that surrounds Medicare and Medicaid.
With Information Technology we can set in place rigorous processes but we can not change the fundamental bases.
Therefore, whether it is open source or proprietary nothing will be fixed unless we build solid bases.
What was your point?
The HL7 Guy
bev M.D. – You’ve cut through to one of the really critical points here. There is a huge disconnect between “interoperability” and “meaningful use.”
This is one of the things that should be figured out *before* the money has entered the Health Care Bermuda Triangle.
Last comment on this, but federal government sponsorship of a FOSS or several FOSS projects is not even worth debating.
The reason for sponsorship by the government (which theoretically at least is supposed to represent the collective will of the people) is to facilitate the development of a public good at low cost that potentially offers great benefit to the public.
Any investment of public funds would and could be minimal, vanishingly so with respect to normal disbursement of funds now for sundry projects, not to mention the $30 billion allocation in the ARRA. If there is little “uptake”, i.e. usage, there is little “waste” of public funds. The upside is enormous however.
In the case of a FOSS EMR, there would certainly be an out-of-pocket cost to a “user”, e.g. physicians’ practice, for software installation (potentially minimal), for training on software use, probably for some equipment installation, primarily for setting up a network and for the inevitable, occasional trouble-shooting. That cost could be significant, but does not have to be. The big cost – to repeat – is the implicit time and effort that users of all types have to commit to learn how to best use any application’s features.
Commercial vendors would continue to exist and presumably thrive as commercial products thrive alongside FOSS product in all existing applications.
An immediate benefit to the public as a whole is that a well-supported FOSS product would greatly increase price competition among commercial vendors, a characteristic almost completely lacking now, as it is in most aspects of medical services in general. It might increase interest on the part of physicians, many of whom seem to refuse to consider EMR system adoption on the basis of price – a more or less legitimate concern in my opinion. In regard to pricing I note a comment made by one of the founders of eClinicalWorks a few years back that the economics of its business were such that the product could be offered at something like $35 (as opposed to $10,000 for an initial provider and $5,000 per additional provider) and the company would still make money.
There is of course nothing about a FOSS product that necessarily requires any particular software engineering expertise.
The comment by bev M.D. is the overriding issue in any case. Whether electronic messages use HL7 as the standard or use XML is immaterial. They both can be used effectively. The important thing is to routinely, accurately and universally digitize data so clinical records are available at a moment’s notice anywhere, so the situation she cites is the exception, not the norm.
On your link I read, “Note: Many documents and presentations referred to in this section are available to HL7 members only or for sale through the HL7 Bookstore. To get more information about HL7 Membership, click here.”
I would emphatically disagree that the standards and technology are not are part of the problem. I, and many others, contend that while it may not be the whole problem, it is a component.
Agreed there are external disincentives for information exchange and transparency. What incentive does any EMR vendor have to be interoperable? Interoperability exposes the vendor to competition and can decrease revenue. Competition and free markets is the American way! There is nothing wrong w/ commercial, proprietary software, but the formats should be open. I don’t want to single out HL7 either. Good people have spent many years on those formats. This is a systemic problem in health care.
Memberships are fine too. Charge me (the developer/corporation) for a seat at the table, but the work product (the formats) should be open…all versions.
Kudos to the Indivo Document Model,used by Dossia and others, for getting this right. In answer to, “Where is the public XSD and documentation?”, the answer is right here:
http://bit.ly/2lJ3NG
I have no relationship whatsoever with this group..as an open source and patient advocate I appreciate their licensing model.
The standards will only reach harmonization/interoperability when they are in the light of day.
Agreed sometimes complexity is introduced and sometimes for complexity’s sake. Many large companies have discovered that adding complexity is a great revenue model, because it means lots of developers, lots of services, etc. This is especially true for SOA/SOAP style systems. A good resource on the topic is “Big Web Services” in the book “REST-ful Web Services”.
Some people might think a person who goes by “The HL7 Guy” and makes a living at this might be a little biased?
C’mon America, lets push the reset button on Health IT.
HL7 Guy: Do you think patients should have access to these formats?
Bev MD: I hear you, and I agree…trying to help. This is just one angle.
-Alan
Alan,
You have been twisting the blog around and for reasons I have no idea of.
Cindy mentioned several points that I gave my opinion on several of them.
The first one has to do with consumer involvment and I answered that this an end goal for the healthcare interoperability domain. Efforts from Google, Microsoft, Dossia, and many others are in this direction. HIMSS for many years has been promoting the PHR. HL7 an SDO organization has also been working on specific standards for this purpose.
Folks in healthcare interoperability have been tough critics of monolithic technology companies taking over hospital organizations and keeping the hospitals silo-locked.
Second she mentions the privacy issue. The privacy issue has been a big factor in keeping interoperability between hospitals difficult. This is a hurdle that has to be dealt with. We have the laws and technology but we still have a long way to go to get it right. Even banks and other industries can’t deal with privacy issues very well. What about theft identity? What domain are you comparing health IT with? Financial? Seems like this is a bad example, huh?
Third, what industry should we learn from? The financial? I just mentioned in the above paragraph how good example they are, good for nothing. The insurance industry? They are part to blame for the big mess healthcare is in. Oh, the fast food industry, oh yeah! Healthcare should follow McDonald’s and Wendy’s model. Ludicrous! No, you mean WalMart! Let’s wait and see.
Fourth, with regard to open source, healthcare interoperability has nothing to do whether the software is open source or not. The only cost that open source saves on is the entry cost. After implementation, open source is more costly or totally fails.
I am a strong supporter of 2 serious open source projects in the healthcare IT domain, one is Mirth Project and the other ClearCanvas.
You are saying that HL7 is responsible for the interoperability mess. HL7 came about to give the healthcare industry a common language because it used to be a “Tower of Babel”.
You have been questioning whether HL7 is open or not. *YES, HL7 IS OPEN AND FREE!* HL7 can be freely used by anyone. There is no licensing to use it. If you want to become a member of the HL7 organization then you have to pay membership fees. This is the only source that the HL7 has in order to finance itself. This organization is formed by thousands of industry experts and many who are from the open source world. Take a look at the list of members for yourself.
Fifth, I think Microsoft is taking the long term vision of this whole issue. They could be accused of being proprietary, but sorry, they are opening their interoperability capabilities to the community. Now what?
And responding your question of the patient having access to the data exchange formats: What the heck are they going to do with it? Do bank consumers have access to XML for a logical purpose? No.
What industry should we copy?
And yes, I do make a living at making healthcare interoperable and I have accomplished succesful outcomes of some of the biggest interoperability challenges in the country.
The HL7 Guy
Alan,
By the way, if you visit my blog site you can see for yourself that it is an OPEN and FREE service to the healthcare interoperability community.
You can also see how I clearly support projects, both open source and proprietary, that lead to the common goals we are all seeking for.
I also make a clear distinction of open source and proprietary endeavors that are only relevant to OPEN interoperability.
Thanks,
The HL7 Guy
From my perspective, the conversation is coming full circle. The earlier comparison Michael made between the banking industry and the health care industry makes sense on one level, except the problem is that the health care industry is *supposed* to be, as a whole, a functional system.
It is not surprising or unethical that banks choose not to share data with each other. This is the norm for a capitalist system. A capitalist approach to health care, as has been demonstrated, is simply not sustainable. Regulations, standards, checks and balances, or whatever you want to call them need to be in place to strike a balance between capitalism, pragmatism, profit, and concern for the common good.
Clay Christenson’s book is a must-read.
Unless someone can come up with a better example, it looks like the Data Documentation Initiative (DDI) http://www.ddialliance.org/ might provide some clues about how to make data interoperable and usable to those interested (a.k.a. transparency).
I know this is a social science-based initiative. And I know the data isn’t extremely sensitive in terms of life-and-death. But I think it’s interesting how far this initiative has come without breaking tax-payer bank accounts. It’s been in use for years.
If you want to conduct publicly-funded social science or behavioral research, chances are you will be required to meet these standards (and demonstrate how you plan to meet the standards) before you collect a penny of government money, let alone a single piece of data.
Frankly one of the better running commentaries on here ever. I frequent other sites where article might be interesting but the commentary is real what drives the value of the site.
Few points:
- Disco died in Comiskey Park in 1979 yet MUMPS is alive and well in most U.S. hospitals and a ton of other healthcare providers. We essentially are going to pay billions of dollars over the next 7 years for state of the art technology circa 1975. Alright.
- Consumer Empowerment is a nice and warm fuzzy term that has been getting bantered about alot the past year or two. Problem is that that is so vague for most of the time and has so much attached to it that it essentially has become meaningless. It needs much better granularity particularly in terms of what processes it is trying to address and what healthcare IT applications this covers. While I agree with some of the principles and that there are some applications that may have wider applicability to the population as a whole, the reality is that it will be a slow and grinding process and that there will still a large number of individuals and the population as a whole that just won’t be engaged until they become very sick/ill.
- “Open source” is another term that gets bantered about endlessly and seems to somehow manage to stay free of any type of negative association. Markets certainly do need some level of healthy competition in them to push innovation (FTC has fallen asleep at the wheel during the last decade on this issue in just about every American industry) but there are several real drawbacks to “open source” too. There isn’t some mythical domain on the Internet where is code is magically maintained by an army of magical programming gnomes. It is alot more subtle than that and messy too. Need somebody or something to provide some rational structure and boundaries to any Open Source project.
Example of where sensitive personal information is exchanged via an open standard and there is a good bit of interoperability: Criminal Justice.
The FBI and the DOD’s Electronic Biometrics Transmission Specification along with its underlying specification from NIST are public domain and they do a great job exchanging information.
State, Federal, and international groups are readily able to exchange criminal information as a result. Under this system, vendors, government (global), and other stakeholders still participate in the development of the standard. It costs a little money to attend the meetings etc, but the result is a free download which can be accessed by anyone. It wasn’t always that way, until the FBI decided to put their foot down in the 1990s so they wouldn’t end up with a proprietary system or only one vendor. In their case, an open architecture is a matter of national security. I know the FBI and many others in the identity/biometrics space emphatically agree that their decision to open things up dramatically sped up interoperability. In this case, the general public can get the format..in XML or otherwise. You are right most people aren’t going to do anything with it, but its nice to know that it is in the public domain. Regardless of your opinion on the FBI, DOD, or the criminal justice world, they are light years ahead of the health care industry in terms of open formats and interoperability.
I still can’t see where HL7 is open or free. Look at the licensing. I do applaud your personal open source efforts. HL7 is just not open source like the spec for TCP/IP, HTTP, etc. (Those are also “messaging” formats. I think all health related formats should follow the same model.
The government needs to fundamentally reconsider the role of standards development organizations in regard to health informatics formats.
In general I think we need a BSD-style open source licensing (over GPL) because it preserves the right to package the code in commercial applications. I may not be a huge fan of Microsoft, but I sort of agree with the fact that GPL can be dangerous to businesses. BSD on the other hand is a different. A perfect example is Mac OSX, which is based on BSD Unix. Its also generally compatible with other Unix/Linux OSs.
Microsoft HealthVault is format agnostic, but has support for CCR and CCD. Google Health is CCR. These are PHR applications, not EMRs but PHRs. Personally I think Microsoft would make a pretty good EMR and for a lot less money…and this is coming from a Linux geek. These issues barriers I raise are so big they even help keep Microsoft (and small start ups) from penetrating the EMR market. Eventually though I think outside companies will chip away and help modernize the health IT industry.
Sure there are incremental success stories all over, but the overall interoperability picture is still pretty bleak. Just because a horse can carry supplies over a mountain, doesn’t mean a truck can’t do it faster and better.
Cindy: I’ll have to take a look at DDI. Looks interesting.
The weather outside is too nice on this holiday weekend. Are we w/o a life, geeks, or just passionate in our beliefs? Not sure, but its family time. I’m also hungry, but I won’t eat at McDonnalds.
I love you all…nothing personal.
-Alan
Great debate, guys, but… So much of this conversation is taking a tastes-great-less-filling sort of absurdity, though. We can all agree that the HC system is broken and that we’d all love to have a hand at fixing it. Arguing “this is the problem, — no THIS is the problem” is pointless, because everyone’s right.
The biggest issue with open source is not that it’s the solution to everything, but that it’s *a* solution — and an important one at that. In an ecosystem of EHR/PHR/interoperability technologies, open source is an important choice necessary to bring down the costs of small-scale implementations, particularly for small practices and PHR start ups. What gives me HUGE cause for concern is that it’s been completely left out of healthcare reform. If memory serves me well, there hasn’t one mention of open source by Dave Blumenthal, or in all of John Halamka or David Kibbe’s great posts on THCB. For the time being, open source has been shut out and we’re going down a road that’s more expensive, less innovative and in the end less effective than it could be.
More important than software, it’s critical that standards be free, open and easily available to developers. Every restriction on cost or access is a brick wall in the path of innovation.
Open standards and open source are also reciprocal: as you take from the community you’re expected to give back. Corporations using open source are expected to fund initiatives, and individual developers are enticed to contribute their code back to the community. In fact, any organization that takes from the OS community and does not give back is quickly seen as a pariah.
Cost of maintaining standards is an issue, there’s no argument. However, the very business model of how the standards bodies work is flawed to the core. For example, IHTSDO spends millions of euros to keep expensive full-time medical experts on its payroll. It could be allowing qualified medical experts in the field to make contributions, and then maintain a much smaller staff to review these contributions and add them to SNOMED as it sees fit. (In fact, it would probably be a badge of honor to be a SNOMED contributor, and give you more letters you could tack on after your name. ) IHTSDO’s closed-door approach is out of date, and will have to change towards a more open model. It’s inevitable.
Open source and interoperability are part of a larger, more difficult problem: healthcare has set itself apart on an island where skills, norms, and standards from other industries don’t apply. You can be a world-class developer, designer, information architect, or illustrator but your skills wouldn’t translate to healthcare, often because your expertise is viewed as worthless without a medical degree.
As a result, Healthcare can’t learn from other industries, nor can it benefit from the innovations that have been made elsewhere. EMR systems shouldn’t run on MUMPS, not because the technology’s old, but because it completely out of the mainstream everywhere else. A sensible approach would run on common standards that are ubiquitous.
Personally, I believe that EMRs and PHRs have everything in common with content management systems (CMS’s) and e-commerce solutions that have been around for years. For example, I’m convinced that an inexpensive EMR could be built on a Linux-Apache-MySQL-PHP platform (a LAMP stack) using a combination of an existing CMS with an ecomm solution like Ubercart.
To me, it’s absurd to go through the costly process of reinventing the wheel for healthcare IT, when 6th and 7th generation solutions are everywhere. It’s equally absurd to define reform in terms of pay-for-standards monopolies held by a clubby minority that’s offered very little innovation in the past.
‘No hay que pedirle peras al olmo.’
- Spanish proverb: you can’t ask for pears from an elm tree
Claudio:
¿Podrías decirnos que fruto da el olmo?
The HL7 Guy (a.k.a. El Catalán) is from Spain too!
Alan:
By the way, I have a life, but my boat had to go to the shop and of course they couldn’t get it ready for the Memorial day weekend!
So I decided to finish about 6 blogs that are partially done and you guys have distracted me from my purpose!
On another note:
I have extracted a few paragraphs from the HL7 website that explain HL7′s involvement with other open standards:
********************************************************
“XML
Health Level Seven has been actively working with XML technology since the formation of the SGML/XML Special Interest Group in September of 1996. Since that time, the SGML/XML group has evolved into two separate groups:
the XML Special Interest Group – which supports the HL7 mission through recommendations on use of Extensible Markup Language (XML) standards for all of HL7′s platform- and vendor-independent healthcare specifications.
the Structured Documents Technical Committee – which support the HL7 mission through development of structured document standards for healthcare.
Both of these groups have produced work items that have been ratified by the HL7 membership. In 1999, HL7 endorsed a recommendation for using XML as an alternative syntax for HL7 V2.3.1 messages. This recommendation was informative in nature (not required for compliance) and was thus not submitted to ANSI for approval. However, an XML encoding for Version 2.4 was balloted and submitted for approval to ANSI in early 2001.
In September 2000, the HL7 membership ratified Version 1 of the Clinical Document Architecture, which defines an XML architecture for exchange of clinical documents. The encoding is based on XML DTDs included in the specification and its semantics are defined using the HL7 RIM (Reference Information Model) and HL7 registered coded vocabularies. In January 2005, the HL7 Membership ratified Release 2 of CDA.
The initial release of Version 3 will use only XML encoding.
HL7 actively participates in and supports the W3C, the organization responsible for the development of XML.”
*******************************************************
This other one explains its usage as indicated in its ByLaws:
*******************************************************
02.03 Approach
02.03.01 Methodology
HL7 shall develop and publish protocol specifications based upon the knowledge of existing and
evolving methods and systems, but shall not establish protocol specifications that intentionally
favor the proprietary characteristics or interests of specific systems or entities.
02.03.02 Ownership
The information content of HL7 protocol specifications shall be in the public domain and be
actively promoted for use by any interested party. The protocol specification documents and
other productions thereof shall be the intellectual property of HL7. HL7 expressly reserves sole
right to publish and sell the documentation of its protocol specifications, and shall exercise all
applicable copyrights to said materials.
02.03.03 Operations
HL7 shall operate primarily with funds derived from membership dues, meeting registration fees,
and publication fees. HL7 is a not-for-profit enterprise and no part of the HL7 revenues shall
accrue to the benefit of any officer, participant, or contributor.
*******************************************************
Any questions?
The HL7 Guy (a.k.a. El Catalán)