This is a health card concept I came across today http://www.flickr.com/photos/juhansonin/393271975/sizes/o/in/set-72057594067707294/ Note that it uses sparklines ... quite tidy, but wonder who would actually carry such a thing with them!
All this on a Health Card ? Will never fly. A woman with her weight on a laminated card ? That's crazy talk. All this information on portal, on a page that looks like this ? That'd be awesome.
The creator refers to it as half-baked, and it is just that. He mentions it in this critical post of MS's CUI lamenting the lack of design excellence.
You can of course comment on his health card too at http://www.flickr.com/photos/juhansonin/393271975
There is nothing special at all about MS CUI. There would be possible potential benefit to Synapse if Synapse followed it and MS CUI became popular. Of which I doubt. We've seen some fairly large negatives of following the MS CUI - it slows the UI down considerably. If we "MS CUI"-ed all Synapse windows / Tabs, it might just overload the GUI entirely. Certainly a defining element of Synapse is that almost anything can be done instantly (minimal server requests). Throwing away the instantaneous feel to Synapse to bring in a GUI that isn't all that special or common seems premature. As I've talked about at length, a weakness of many "Tabs" etc, is that the buttons that show are not always available and do not apply to what the User can do next. Unlike almost every other application I use, Synapse really doesn't have any "Hidden Buttons" (ie. Drop Down menus). The benefit of that is all options are visible, the drawback is that you have to limit the number of options and that options that are not available in the current context are visible. A simple UI with "Just in Time" Buttons is something that appeals to my eyes, as well as some dropdowns here and there. Good GUI development isn't copying MS CUI it's deciding what is needed for your app and what your IDE has to offer.
This is a great example of a patient directed PHR type entity - Essentially irrelevant to the doctor. Only useful stuff on there is:weight, lipids, vaccinations, maybe Hgb. The medications and medical history and allergies (the real important stuff) are unrealistically small. I'm not even going to ask why he is tracking his CD4 count. This stuff is fine for him and his health, but if he is trying to pioneer something that will work for a wider audience, he isn't heading in the right direction. Is this supposed to be a laminated CCR ?
[There are always tradeoffs .. in this case, the decision was made to try and increase usablity at the expense of speed. I am happy with the outcome. The MS CUI's main problem is that it is being rendered by Silverlight, and this combination needs a minimum core2duo processor with 3Gbs of ram. I guess maybe one day Nvidia etc will produce add on processors specifically for rendering Silverlight like they have for H.264 hi definition video [] And of course Rebol3 is in alpha now, and is significantly faster than the current version of Rebol including in the graphics arena. As for buttons showing on tabs that aren't relevant, that's a premature optimization on my part using the same buttons for each tab, rather than coding buttons for each tab. It does lead to less memory usage, but at the expense of being visible when they need not be. Dropdown menus - don't like them. They make using a tablet so much harder.
The HealthCard referred to above (http://www.flickr.com/photos/juhansonin/393271975/) was an early prototype for a digital and physical service. It has a ton of problems... but has several tasty ideas. Different representations are needed for different patients... from tiny tots to kids to young adults to mature adults to geriatrics (and many shades in between). My history and allergies are going to be small versus someone twice my age who suffers from dementia. A driver license-sized card is one output; only critical data would be printed... with a possible add-on, print-at-home sticker for yearly changes. An embedded chip could house the data. Citizens should have a certain slice of their health data PUBLIC. Privacy is a ruse. Take a peek at PatientsLikeMe as an example. The digital service = where the fun starts. We're getting more health data at home... and as the cost of devices nears zero, the cost of collecting data nears zero. We're lacking an open source health record service: ... that a patient (the atomic unit of Health) can push data into, see trends, get help from wiki-Docs, and ultimately change their behavior for good ... that a clinic could setup in 1 day The US [insert country of choice] needs to mandate a yearly health form for every United Statesian... just like taxes. Now we need a simple health data standard (CCD = ugly, CCR = not horrid but needs help). -Juhan, juhan@mit.edu
Welcome Juhan Not sure how you found us here [] Not quite. Representations will depend on how ill the patient is .. not necessarily their age. Disease is no respector of age. There is a significant barrier here on maintaining such cards. Who's going to program the chip? The family practitioner is the one holding the data ... one more unfunded task for her/him? Seems to be a number of FOSS PHRs out there ...or in development. But my experience with providing a place for my patients to store their data has been disappointing. Hang on . there is very little evidence to support yearly physicals ... what evidence is there that a yearly health form will benefit?
Agreed. Age is one of many ways to slice the data. Disease, conditions, geo, and n other topics can baseline/filter the representation. This is a DIGITAL service; slicing and dicing the data and representation is key to getting the right view for you. I dig the family practitioner updating the data, but ultimately it's about the patient and collecting their own data... and adding huge value to the hospital and doc-collected data. Docs + hospitals will not be the only hub for the information. Doctor Mom (often the head of health of a household) is a data gate. I'm a data gate (for taking my BP, weighing, activity levels, etc). One minute clinics are another. The high ceremony engagements with doctor gods is changing and patient-collected data will be highly valued. You're spot on - an OS health services are game changers. I haven't seen 1 amazing FOSS PHR... YET. Several are available, but none rock my world from the data standard to display of the data to usability to installation. A few have fantastic ideas like Tolven with it's wiki-esque infrastructure. The OS community will get there. Where's the itunes for PHR? It will be the iPhone (and when the Gphone gets its act together).... Graham, shoot some examples out of good systems; I would love to see them in action. People need incentive... especially around health (most of us are in denial). One way - not THE way - is to offer a tax break if they submit a record just like we do with taxes once per year. There are a bunch of other mechanisms that will kick this idea to the curb. But we need a standard way to SEE (and GET) the data. Yearly physicals == monthly, weekly, and daily micro updates to our health records. .... data from our Toto toilets, GoWear fits, Walmart 1-minute clinics, sensors in our bathrooms+kitchens, mail-in and instant result blood tests, etc. This is INSTRUMENTATION. A good US-centric overview of the PHR scene: http://www.slideshare.net/John.Moore/phrs-platforms-consumer-trends-presentation -Juhan, juhan@mit.edu ps: The value of good instrumentation == Flickr supplies all the traffic surrounding images. Voila, up popped synapsedirect.
I'm unclear as to who the intended audience for a health card really is. Once that is determined, then we can decide on what data should be collected. As for patients collecting data, our concern is that they are not discriminating in what they collect ... are the docs really interested ?? As for FOSS PHRs, I was looking at using http://www.wik.is ... as the underlying infrastructure ( free wiki ), and creating standardized plugins to display the data. Eg. we already using sparklines for tracking serial data in Synapse, and I recently added Simile's timeline ( from MIT's interoperability project ). Mindtouch is FOSS, and basically has the underlying architecture to allow us to add any web service we like. So, perhaps my ideal health card would be some very basic data, and a URL to wik.is encoded in a bar code ... I just scan it with a bar code scanner and my browser/emr displays/grabs the data ... no need for embedded chips that need re-programming. Re-use existing tech.
I noticed you mentioned Tolven which has a wiki architecture .. I guess I should check it out again, it's been a long time since I visited their website. We did discuss PHRs briefly a while back, and I looked at that slide presentation which is interesting but with some rather complicated slides! I think expanding on the above, my starting point would be a wiki too, and I've already mentioned Mindtouch's Deki as the underlying architecture. It has a rich gui which covers a REST based server ( the php wiki does all its work via REST calls ). The user is able to upload heterogenous files and can define permissions for other users to view etc. Not https though, have to pay for that. So, the user has a place to collect all their health data. All we need is to add extensions to allow them to view the data in ways such as your health card which I suspect is more suited to this purpose, than to carry around. I've created some dummy timelines to illustrate the type of displays I would like to see eg.http://compkarori.com/timeline/timeline2.html The important step is to settle on a standardized API for EMRs to upload and retrieve data from such collections so that the PCP can painlessly transfer data from their EMRs to the patient's PHR. I am not convinced that patients can collect significant data themselves. I know PHRs are supposed to be patient driven, but my experience is that few are driven enough to do the work.
This version looks a lot more usable but like the others, the sparklines need a start date indicated somewhere. Personally I would ditch the immunizations and just do medications and medical conditions, and allergies. In an emergency, that's way more important to know than whether you were vaccinated against poliomyelitis.
