Having a single model will certainly solve the inter-operability problem. However, countless smart people and many millions of dollars before us have not been able to achieve this solution.
Well, either it has no solution, or no single model has emerged yet that fits the medical data models in a satisfactory way.
I am most interested in your perspective!
Ok, please grab a coffee, or a nice cup of tea, because this email will be long. Perhaps it would be better to print this email, as we all know that our cognitive abilities aren't at their best while reading from a computer screen. And I assure you that some pretty strange ideas will emerge.
So here it goes...
1. About the nature of a Hospital Information System: I have a simple concept of a health information system (HIS), I see it as a system in which qualified people use a set of tools - hardware and software - to collect, change and store data to produce information about health, be it people/user's health, healthcare organization health, and even, HIS own health.
1.1. A HIS is a living organism. Some of us see it much like a symbiont, made of hardware, software and more or less knowledgeable people. Humans are this organism's most precious asset. They are the brain of this HIS organism. As such some of us do not value as much the tools that those humans must use to produce information, i.e., to me the (hard|soft)-ware part of the HIS organism is kind of secondary.
1.2. HIS do not diagnose, HIS do not treat people, HIS only produce information.
1.3. For some strange reason it seems that health organization's "decisors" are not the only ones that fail to acknowledge that in Health IT there is no such thing as "plug&play". And yet it seems pretty clear that even if you could clone the hardware and the software, it would still be impossible to duplicate all the people that made possible a Health IT system to succeed in any given healthcare environment.
1.4. Now comes the strange part, as it relates to people that, at least for the fact of being caregivers and/or main stakeholders at healthcare organizations, should know better: 1.4.1. Whenever we talk about a HIS we tend to talk about the more or less clever model (the hypothesis) that we proposed to represent it. 1.4.2. We tend to speak about the software (material & methods) that we are using to validate the model. Sometimes we even talk about the "right" hardware to run that software. 1.4.3. What we tend to forget to talk about, or fail to see as much enthusiasm about, is the fitness or well being of the living part of the HIS "animal", i.e., the healthcare institution people.
1.5 Some of us see these as fundamental reasons to the failure of more than 80% of all health IT projects. Could it be because they are not "plu&play" things? Could it be because they are living organisms that must strugle to adapt to new environments?
2. About real world limitations: 2.1. I am sure that I would not be able to invent a relativity-like theory to holisticly explain the medical knowledge model and the medical "praxis" dataflow. I am not even sure that I could make good use of information theory to correctly model my ideas in simple code and/or a few robust database principles. But, I wonder, had the first "talking" man used the same convoluted approach that we tend to use today, perhaps we were still debating about the best and absolutely most adequate way to express the whole complexity of the human ideation in the utterly perfect spoken language. And that would most certainly need to include present caveman ideation like "I-have-to-survive-another-day", to a future proof "moon-team-2-calling-earth".
2.2. But, as I do not know better, I am involved in the research and making of the software part of the HIS organism. And in my "caveman" I see the software part of a HIS as simple thing, having 3 simple components: a. Clinical, b. Administrative & Operational c. Financial
2.3. A healthcare organization's main business (even when it is non-profit) are Clinical Services. In order succeed at that people need operational support and an administrative structure to keep track of the where and whys. Finally, to do all those beautiful things we need to find ways to fund our organizations and to wisely manage those funds, and that is the financial part of the HIS organism.
2.4. Once again, I try to not forget that real people are the living force of those 3 components. The quality of their work will, in the end, condition the value of the information produced.
3. About the objectives of a developer of HIS software: 3.1. So, at this time I only aim at giving those people the most adequate and robust tools that I can find. As a caveman I am not yet worried about good information models or knowledge theory vanguardist solutions. I am just seeking simple, fast, reliable, secure and sheer user friendly tools to help people that work at the HIS organism.
3.2 As I do not know any better method to improve knowledge, I do trust the scientific method to produce these tools, i.e., the software part of this HIS organism. That's why I favor the open source approach as to me it is simply the scientific methodology applied to the making of software.
Now, for the pragmatic part of this monologue:
4.The best HIS software tool that I have meet to attain the above stated objectives is, so far, the VA Vista project. But its infancy and greater part of its adult life were not developed under the scientific method (meaning that the main body of development was made under the proprietary-like model of development, i.e. the opposite of the open source approach). And now it has grown to a mature beast, but sadly, much like people, it has an old body but likes to think that it is still "very young at mind". And, as an european, I could easily add that the Vista organism has been working with american people and idiosyncrasies for too long... but I will not say that because I really do value all my american colleagues and friends.
5. Now enter Care2x, the "web based", "standards based", all open source, "everybody-knows-how-to-contribute-to-it" approach.
5.1. Add to it a slightly philosophical shift from a traditional technological centered approach, to a far more pragmatic people/procedure centered approach. Gather people around ease-of-use and user model centered interface design, add procedure safeness, "point&click", "copy&paste" and... we end up with a monster.
5.2. At least it must be a monster, as it is not based in strict, orthodox, principles of information theory. Its database isn't even well structured. It was even developed mostly from inside the clinical community and to solve those little day-to-day problems found in ordinary healthcare organizations.
5.3. But guess what? It seems that people are loving it! Go figure! It is already present in 20 countries. It has at least 107 real persons that regularly contribute code, ideas, algorithms, suggestions, critics. Is it perfect? No, far from that. Has it a good conceptual model? Well, here we diverge. It certainly does not have the most accurate medical data representation model. But its software engineering is elsewhere as solid and as robust as currently available FLOSS permit: 5.3.1. It uses a 3-tier architecture; 5.3.2. It works over TCP/IP and HTTP locally in a LAN or trough the Internet or any other support in a WAN as large as we want; It even lets you interact with it from a plain mobile phone (via WAP); 5.3.3. At a higher level, critical data is safely carried inside S-HTTP; 5.3.4. As coding languages it uses mostly PHP and a few lines of javascript; 5.3.5. It is mostly hardware and OS independent, meaning that you may use it over Windows and i86 derivatives, or over Debian or BSD and all the panoply of HW architectures that they support.
6. And, most important... Care2x code works, or should I say... it lives.
Now, tell me, is there anything wrong to wish an even more eclectic approach? Is there anything wrong to be willing to try someone else clever ideas, which I am sure that we all agree to exist in approaches like GNUMed, OIO or Care2x? Isn�t that what we all do at our hospitals when we try some clever new aproach that we found in the newest literature?
Best regards,
J. Antas
------------------------------------------------------- SF.Net is sponsored by: Speed Start Your Linux Apps Now. Build and deploy apps & Web services for Linux with a free DVD software kit from IBM. Click Now! http://ads.osdn.com/?ad_id56&alloc_id438&op=click _______________________________________________ Care2002-developers mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/care2002-developers
