The high implementation failure rate of EMRs suggests that there is something fundamentally wrong with the user interfaces ...
What is it that you don't like about your EMR? Too many tabs? Too many popup windows? Too cluttered?
Graham http://www.synapsedirect.com/ Synapse - the EMR for smart users
I don't think many physicians could say. A bad user interface is a UI that does not provide a fulfilling human-computer interaction. What makes a fullfilling computer interaction is a complex science that only a few development teams manage to nail. You know it when you have it, but how to get there...
My programming tools (IDE) is to me what an EMR is to a doc in so much that it is a tool to complete work for the day.
Having programmed use a number of IDE's and a number of languages over the years they have of course evolved. For Java I started with a regular notepad or vi (unix notepad) - the equivalent of paper for an EMR.
Then came JBuilder which was frustratingly slow and unsatisfying - the equivalent of that first EMR the solo doc mistakenly implemented because it was local to them.
Visual Age from IBM was next and what a marvelous tool - drag and drop UI building, API searching and a host of other things. Increased productivity (so it seemed). It was not cheap though. The equivalent of todays EMRs - eCW, Cerner etc.
But now I have Eclipse.
It is a dream tool. It is not bloated, rather the UI is simple with subtle icons and shortcuts that have a host of features for the advanced user to explore. My jaw dropped so many times as I hit keystroke combinations and the system guessed (correctly) what I wanted, auto filling this, auto completing that, hell even refactoring my code to make it right !!! And fast, so fast.
Eclipse is the next generation of IDEs that puts the old commercial products to shame.
And it is open source aka Free. And developers like free, we care not to brag about how much money we spent on our IDE. Rather we prefer to show we are smart and can pick tools that increase productivity, integrate with the other open source tools and is *fun* to use. Today there is no EMR that delivers the Eclipse experience but the projects are in place, be it PatientOS, ClearHealth, OpenMRS, GNUMed or new ones to come.
It is only a matter of time before one will ecilpse the commercial EMRs and the next generation of docs will shake their heads at the thought of spending thousands on an EMR. One will eclipse the commercial Healthcare Information Systems and the next generation of IT Directors will shake their heads at the thought of spending 10's of millions on an HIS that does not even play well with the legacy applications.
Greg
http://www.patientos.org
I think many vendors are clueless. I ran into a GE exec at an EMR awards dinner hosted by Frost and Sullivan and attempted to politely ask her if GE had any plans to simplify the Centricity UI. She was insulted at the suggestion and defiantly crossed her arms and said Centricity was an outstanding product and why would they consider changing a perfectly functional, well selling product.
I thought to myself, yeah, right. She's obviously clueless. If they fixed their UI they could sell twice as much. I probably would not do well in the large corporate setting. I can't stand mediocrity or stupidity.
Disclaimer: I am the founder of e-MDs. Highest rated by doctors. All posts are opinion only
gchiu: The high implementation failure rate of EMRs suggests that there is something fundamentally wrong with the user interfaces ...
While I do agree that most EMR interfaces are terrible and that most EMRs are terrible, I think that the problem may be more challenging than just some tinkering here and there to make EMRs work in the Medical Profession.
I think it's silly when someone says, "all industries have become electronic but medicine was left behind".
The point I want to make is this: Electronic Documentation/Management in Medicine is far more challenging than other industries.
Three major problems are: (1) the complexity of the documentation (2) the time constraints, (3) the challenge of having to interact with the patient during the visit.
Considering these constraints which likely have never been accomplished in any industry before, it's possible that an EMR that "works" for most offices is more wishful thinking than reality.
(of course, if you could see 1/2 the number of patients per day, most EMRs would work).
Electronic Medical Records is some of the most complicated software to create, not from the fact that we know what needs to be there, but the user interface is everything. If it doesn't get used, by the end user, then it's no good for anyone. I remember a long time back a thread here on what was called "click wars", keeping the number of screens and clicks down to a minimum and that still holds true today.
From the programming side you need to get as much information on a screen as possible, but yet have it clear enough to where it can be seen and used. Back when I was writing and learning how to code I remember coming up with something I thought was the cat's meow, but yet when going back to my MD partner, it didn't work, so back to the drawing board with a new plan, but all was not totally lost as it got him thinking as well and together we would come up with what worked for him, as perhaps my idea was not totally a bust, but it was not written to where it was totally functional in the way he needed it to work.
On the other side of the coin, sometimes I could not return the exact method he was looking for as well, but again 2 heads together and working back and forth would bring something good together. Programmers and end user thoughts are not always equal so both need to work close together for the winning solution.
In time, I am thinking that somewhat of a common user interface will evolve with all software companies, maybe not 100% the same, but similar, will evolve so it won't be like having to learn numerous systems, kind of like driving different cars, they all do the same thing and get you from point A to B, but all have similarities and minor difference so we can adjust. The controls for the windshield wipers might be in a different place, but we all know they are there and find the controls quickly and move on.
Barbara Duck Ducknet Services TabletKiosk Sales Information
Blog: http://ducknetweb.blogspot.com/
The Widget Blog Site: http://ducknet.net/default.aspx
Skype Address: Ducknetservices Phone: 714-898-1917 Email: barbaraduck@ducknet.net
The palest ink is better than the best memory. - Chinese Proverb
I'm sure we will end up with the perfect UI eventually but it's a question now of determining what works and what doesn't.
Fortunately there are no companies dominating this market so this allows innovation from startups.
Considering there are millions of different scenarios, you'd need millions of perfect UIs.
One thing that is annoying in our system is the inability to multitask. You can do one thing on one patient. If interrupted for another patient's needs, you have to exit the chart (the paper equivalent to closing the chart) so that even relatively simple questions may take extra time.
The system was built on the concept of entry in a database for the purpose of coding. It is painfully slow.
The system was not designed to carry all data forward. For example, our Health maintenance module is suppose to have the dates various tests, procedures, immunizations were done, when they are entered elswhere in the system. However, some of these have to be manually populated.
Our EMR cannot be modified "on the fly.
Scanned documents are TIFF files that are very small, and one cannot see the entire document making them difficult to read.
Our prescription module pops up inaccurate alllergy alerts.
Our system cannot manage multiple visits on the same day (ie providers from 2 specialties)
imemod:One thing that is annoying in our system is the inability to multitask. You can do one thing on one patient. If interrupted for another patient's needs, you have to exit the chart (the paper equivalent to closing the chart) so that even relatively simple questions may take extra time.
FWIW, imemod uses Medinformatix (If I remember correctly).
Surpisingly most EMRs have trouble with that one.
Patients constantly asked me, how was my husband's shoulder Ultrasound ?
etc. I try to answer these questions "on the fly". Synapse does well for this but I can see how most EMRs would be cumbersome in that regard.
My EMR is not paper. There is nothing like paper; i took good notes and kept a good med list and problem list. I could visually scan a page and see everything in a glance especially why there were back to see me. I could flip to the consult section and quickly see letters from docs. I could flip to labs and in a glance see abnormals. I miss paper [:'(]
hirschr: There is nothing like paper; i took good notes and kept a good med list and problem list. I could visually scan a page and see everything in a glance especially why there were back to see me. I could flip to the consult section and quickly see letters from docs. I could flip to labs and in a glance see abnormals.
There is nothing like paper; i took good notes and kept a good med list and problem list. I could visually scan a page and see everything in a glance especially why there were back to see me. I could flip to the consult section and quickly see letters from docs. I could flip to labs and in a glance see abnormals.
Same here. I have a difficult time explaining the convenience of visual scanning to those who do not use an EMR. We take it for granted. I ask them to imagine looking at a record through the end of a rolled up piece of paper. It is not identical, but you do get the feel for limited information presented and the need to intreract with the "system" in order to find what you need/want.
I’ll admit that it might take a little longer to gather all of the information on a patient in an EMR. But with time, I have learned how to navigate the screens in such a fashion that it is quicker than paper. The real difference is that EMR’s must necessarily gather all the information to properly code. Paper does not have to do this. Using a mouse is a very slow way to navigate and this was quite apparent to me many years ago when I selected an EMR. It is very unlikely that we will every see a universal interface. EMR’s will have to be designed around touch screens to realize their ultimate potential. Most users have a multitasking operating system and it would seem to me that unless your program is overly cumbersome in the log in process, you could open another instance of the EMR if you want to quickly look at another chart. Again, with the design of my touch screen EMR, you can exit a chart and it will remain available to use again and return to the same spot until you close out the chart completely. This is the same as paper in that the chart remains on the desk until you close it and file it away.
Most EMR’s missed the boat in initial design to give physicians what they need and most physicians have no idea what the boat should look like when shopping.
There’s very little that I don’t like about my EMR because it was designed on the correct platform from the beginning. The only issue that I have is the ability to generate good looking notes without editing. This is possible, but I have no desire to learn HTML programming. I have yet to see a company willing to sit down with the end user and provide the time necessary to fix the final 20% of the Pareto principle.
imemod: Same here. I have a difficult time explaining the convenience of visual scanning to those who do not use an EMR. We take it for granted. I ask them to imagine looking at a record through the end of a rolled up piece of paper. It is not identical, but you do get the feel for limited information presented and the need to intreract with the "system" in order to find what you need/want.
I think we need to understand what it is that you are actually doing when scanning your paper notes.
I do remember when I had to see new patients with thick hospital notes that it was impossible to extract the information I needed in the time allocated.
gchiu: imemod: Same here. I have a difficult time explaining the convenience of visual scanning to those who do not use an EMR. We take it for granted. I ask them to imagine looking at a record through the end of a rolled up piece of paper. It is not identical, but you do get the feel for limited information presented and the need to intreract with the "system" in order to find what you need/want. I think we need to understand what it is that you are actually doing when scanning your paper notes. I do remember when I had to see new patients with thick hospital notes that it was impossible to extract the information I needed in the time allocated.
I agree. Hospital charts and outpatient records are different beasts. I interpreted the issue relative to the question of an outpatient EMR. I notice you referred to hospital notes in the past tense. I assume that you now see patients with an easily searchable electronic record. If that is true, what are you using?
imemod: I agree. Hospital charts and outpatient records are different beasts. I interpreted the issue relative to the question of an outpatient EMR. I notice you referred to hospital notes in the past tense. I assume that you now see patients with an easily searchable electronic record. If that is true, what are you using?
I no longer see patients with hospital notes, but even then after over a decade, my electronic notes tend to become quite large. And so I needed a quick way of searching thru a patient's electronic chart. I do this by tagging each encounter, and the system can quickly rebuild a patient's tags so that I can filter down on to the relevant encounters.
I use Synapse EMR