Health IT

I'm going to be controversial and provocative here, please bear with me :)



EMR's by their very nature are technology intensive. Their development has largely been driven by technologists with very little experience with the human condition as applies to the clinical setting. Those physicians who design EMRs are also "limited" by this technological approach. This, in my opinion, is the reason for the low level of acceptance of these systems. The VA system is an "enforced" system. Those in the VA do not have a choice and hence are forced to use it. It's "success" does not in any way reflect any excellence in design. And as we all know , "free" does not always translate to "best".



The conventional paper based records format, though imperfect for mass retrieval of information, are perfect for the recording of information. This format has evolved over many many decades of physician consensus based design and many millenia of human development in putting "writing implement to substrate to record events and concepts". The paper record lies at the very core of the training as a physician and hence it is second nature to doctors around the world. It is a cheap and ...

I'm going to be controversial and provocative here, please bear with me :)



EMR's by their very nature are technology intensive. Their development has largely been driven by technologists with very little experience with the human condition as applies to the clinical setting. Those physicians who design EMRs are also "limited" by this technological approach. This, in my opinion, is the reason for the low level of acceptance of these systems. The VA system is an "enforced" system. Those in the VA do not have a choice and hence are forced to use it. It's "success" does not in any way reflect any excellence in design. And as we all know , "free" does not always translate to "best".



The conventional paper based records format, though imperfect for mass retrieval of information, are perfect for the recording of information. This format has evolved over many many decades of physician consensus based design and many millenia of human development in putting "writing implement to substrate to record events and concepts". The paper record lies at the very core of the training as a physician and hence it is second nature to doctors around the world. It is a cheap and ubiquitous technology that is imbued with ease of use and with many many more levels of sensory interaction with the physician than EMRs allow (how do you mimic the sensory input from thousands of cutaneous receptors every time a pen stroke is made). Visual display technology is still evolving but has not reached the level of full sensory experience that paper has. As monitors and other visual display technology improve and incorporate tactile and haptic feedback, EMRs will become more acceptable. However, for an EMR to become second nature to doctors around the world, "something more" has to change fundamentally/philosophically. IBM's development efforts in this area are noteworthy. (How many physicians or EMR developers have heard of IBM being in the medical records business?)



But, as I said, provocatively in an earlier post, until these as yet undefined fundamental changes are defined and instituted as required standards, the EMR industry will continue to stumble along.



Jay Devasundaram

The idea of attributing EMR's slow spread to the lack of 'sensory' feedback in filling in forms is unconvincing given society's push towards technology in practically every format for information recording, storage, and transfer. Mankind's precedent for information storage has little relationship to what is our best option now - I'm sure the technological advantages of transfer, duplication, backup, sharing, and multiple real-time access would outweigh the satisfactions of filling in a paper form and adhering to 'what we've always done'. To the last point, if IBM does release the best market product for EMR, I doubt it would fail simply because it does not have a health-associated brand. Google Health is one of many successful and recent examples.

The factors inhibiting the progress of EMR systems in the US include the fragmented state of existing systems, lack of suitable hardware and process infrastructure for web-based EMR, a lack of time within the medical community for training and integration, and the lack of a clearly "Pareto" optimal EMR solution to encourage a positive network effect. I imagine many hospitals would be loathe to invest the money and time to integrate one solution when another may become ...

The idea of attributing EMR's slow spread to the lack of 'sensory' feedback in filling in forms is unconvincing given society's push towards technology in practically every format for information recording, storage, and transfer. Mankind's precedent for information storage has little relationship to what is our best option now - I'm sure the technological advantages of transfer, duplication, backup, sharing, and multiple real-time access would outweigh the satisfactions of filling in a paper form and adhering to 'what we've always done'. To the last point, if IBM does release the best market product for EMR, I doubt it would fail simply because it does not have a health-associated brand. Google Health is one of many successful and recent examples.

The factors inhibiting the progress of EMR systems in the US include the fragmented state of existing systems, lack of suitable hardware and process infrastructure for web-based EMR, a lack of time within the medical community for training and integration, and the lack of a clearly "Pareto" optimal EMR solution to encourage a positive network effect. I imagine many hospitals would be loathe to invest the money and time to integrate one solution when another may become the market leader in a few years. The cost to adopting any EMR system is massive for high-stakes high-cost environments, and I agree with Hamish that it would be beneficial for many hospitals to adopt an open source system such as VistA. I imagine that the practical hurdle to gaining traction for any of these systems might require one of three events: a centralized effort from the medical community, a government action, or a 'bundling' move by a market participant (analogous to Sony winning the Blu-Ray war against HD DVD partially by bundling with the PlayStation 3) that may push one solution forward. Any thoughts?

Jay,
Lovely points, but if we are going to have to rely upon doctor's handwriting we are all in big trouble! Haha
There are other aspects of this too. In medecine we don't have a 'scribe' constantly drawing sketches of EKG output. Low res lcd screens give us quick and easy access to representations of complex data.

Agreed that VA is forced into a system, but ultimately it was a hierarchical decision for efficiency and information sharing.

This is a very interesting time to be discussing input methods, data normalization and quality, recordkeeping tactics.

Nothing yet approximates pen(cil) on paper, granted, but I have also seen growth in the voice transcription systems since 1993 or so, e-ink is coming closer to paper resolution, voice recognition is coming along nicely too, not to mention ink pens that capture motion and digitize it.

I think personally that lack of uptake has to do with institutional memory of klunky technology, poor interface design, redundancy, and as you say, the reliability of hiring a human to retrieve the data. There is only one patient, so a single file can be effective. We also have to deal with specialization however, and this ...

Jay,
Lovely points, but if we are going to have to rely upon doctor's handwriting we are all in big trouble! Haha
There are other aspects of this too. In medecine we don't have a 'scribe' constantly drawing sketches of EKG output. Low res lcd screens give us quick and easy access to representations of complex data.

Agreed that VA is forced into a system, but ultimately it was a hierarchical decision for efficiency and information sharing.

This is a very interesting time to be discussing input methods, data normalization and quality, recordkeeping tactics.

Nothing yet approximates pen(cil) on paper, granted, but I have also seen growth in the voice transcription systems since 1993 or so, e-ink is coming closer to paper resolution, voice recognition is coming along nicely too, not to mention ink pens that capture motion and digitize it.

I think personally that lack of uptake has to do with institutional memory of klunky technology, poor interface design, redundancy, and as you say, the reliability of hiring a human to retrieve the data. There is only one patient, so a single file can be effective. We also have to deal with specialization however, and this is where the pain is really felt I think.

Getting facts from one office or department to another to create a complete picture of the patient.

This brings up another issue. The training and modern legacy of North American medecine is symptomatic and fractured, rather than holistic.

I think this comes across in the design of IT surrounding health care from an institutionalized psychology.

With a more human centric approach really useful systems can certainly be made.

Om
"Pollution is a symbol of design failure."
-William McDonough

The response provided is "technology jargon" and completely sets aside the complexity of the human sensory experience and the importance of the UI in the acceptability of EMRs. Remember the DOS days when all data entry was on a tiny, highly pixellated two tone screen with command line entries? Why do you think the current Graphical User Interface is preferable to that DOS interface? It's the human sensory experience. Apple is the leader in this GUI implementation that is closest to the human sensory experience. They do this by controlling every aspect of the design and manufacture of the hardware- the Apple mouse and key board (tactile and auditory sensory experience) and Apple monitors (visual sensory experience) are way better interfaces sensorially than any of the other generic windows based offerings out there. But this is not good enough. No amount of technological banter is going to convince folks like me to accept a tedious and technologically mechanized EMR that does not take into consideration more of the senses that I am endowed with, integrated tightly with the clinical business processes that I am comfortable with. This is the design philosphy change I am talking about. Build such a ...

The response provided is "technology jargon" and completely sets aside the complexity of the human sensory experience and the importance of the UI in the acceptability of EMRs. Remember the DOS days when all data entry was on a tiny, highly pixellated two tone screen with command line entries? Why do you think the current Graphical User Interface is preferable to that DOS interface? It's the human sensory experience. Apple is the leader in this GUI implementation that is closest to the human sensory experience. They do this by controlling every aspect of the design and manufacture of the hardware- the Apple mouse and key board (tactile and auditory sensory experience) and Apple monitors (visual sensory experience) are way better interfaces sensorially than any of the other generic windows based offerings out there. But this is not good enough. No amount of technological banter is going to convince folks like me to accept a tedious and technologically mechanized EMR that does not take into consideration more of the senses that I am endowed with, integrated tightly with the clinical business processes that I am comfortable with. This is the design philosphy change I am talking about. Build such a better mousetrap and the people will come.



Jay Devasundaram

Om,

I like your response. Perhaps we can discuss some of these philosophies off-line (since I really don't want to indulge in lengthy "back and forths") in deference to other members of this list who may not wish to be be drawn into this aspect of health.





My email address is :



Regards,



Jay Devasundaram




Checkout my photography on the web : http://community.webshots.com/user/jayanthd

Excellent.

I would also advocate creating a channel, or similar technology where it would be a group conversation in an environment where lengthy banter is encouraged.

The advantage to this that I see would be that the conversation would then be 'discoverable' not only inviting other perspectives, but hopefully, being influential on current or future designs.

Om

"Pollution is a symbol of design failure."
-William McDonough

Jay,

I liked your second point about the importance of user interface in the medical context and I regret missing that perspective in the list I put forward - Stanford has an entire branch of its computer science department dedicated to 'Human-Computer Interaction' where they study how to make computers as user-friendly as possible. The digitized pens Om mentioned could be a great way to translate your preference for paper straight into a digital format where you can send it onto another specialist, as long as the EMR system works with at least one type of digital pen!

I have approached digital systems for medicine from an economic and technological perspective, but I am very keen to understand what your perspectives are as users keen to adopt new programs but with concerns about interface. What kinds of things do you feel the need to do with a patient's medical record that Windows text-based systems might prevent you from?

Om, Jay,

These are the sorts of discussions that are perfect to hold on GHDonline,
rather than taking them offline or creating a separate channel -- we
certainly encourage lengthy banter! The Health IT community does have
mailing-list-like functionality, but also provides a web interface (
http://www.ghdonline.org/tech/) where--to borrow your words, Om--"the
conversation [is] discoverable". Others are able to not only read the
entire discussion and contribute to it via the web, but can also use the
GHDonline search interface to find this and other relevant discussions.

-Aaron

San,

Well, for one, text is "stupid" in that it doesn't know what it is. You have a patient 'file' with entries in it named june visit or some such.

Having a system breaks up each fact in a way that is reusable and only shows the relevant part to each viewer. You can also ask questions like "How many patients with hypertension also have colitis?"
and get near instant answers.

Or did I misunderstand the words 'windows' and 'text based' ?

Om

Thanks Aaron,

I especially like the email interface to your forum system. Wise choice!

Om

"Pollution is a symbol of design failure."
-William McDonough

Hi Om,

By Windows and text-based I mean medical software where you enter in patient information in a series of boxes or 'fields', such as Patient Name, Date of Birth, Date of Appointment, Doctor's Comments, and so on. You would like to be able to search through these notes within a patient's medical records, which should be possible I'd imagine. With regard to statistical analysis, doctors write their personal notes in different formats so I wonder if a search function could be designed to find the right numbers to look for in a mass of notes.

Sanjee

Hi All,

The Meaningful use workgroup of the Health IT Policy Committee (in charge of making recommendations to the National Coordinator for Health Information Technology on a policy framework for the development and adoption of a nationwide health information infrastructure, including standards for the exchange of patient medical information) yesterday released initial recommendations for defining "meaningful use" of electronic health records: http://healthit.hhs.gov/portal/server.pt/gateway/PTARGS_0_11113_872720_0_0_18..., and also a “meaningful use” matrix: http://healthit.hhs.gov/portal/server.pt/gateway/PTARGS_0_11113_872719_0_0_18...

What do you think of these recommendations?