louisrosenfeld.com logotype

Home > Bloug Archive

Feb 08, 2004: Hillary Clinton, Information Architect?

Speaking recently at the World Health Care Congress, the Dems' presumptive 2008 presidential nominee offered her vision for improved health care. Senator Clinton wants to "build an information technology infrastructure that puts the right information in the hands of the right people at the right time." Mmmm, good, sounds very IA-ish.

Even better: she said that national government leadership should "ensure interoperability and the use of common standards and terminology." Like XML schema and MeSH headings? Harry and Louise commercials aside, was this what was missing back in 1993?

Cool; maybe Hillary or someone else will get it right next time by hiring scads of information architects. Anyway, many thanks to Socialtext's Ed Vielmetti for the tip.

email this entry

Comment: Dean Karavite (Feb 10, 2004)

Hi Lou,

As you know, I used to work with medical information and physicians as users (try it some time!). While people go on and on about the wonders of the electronic chart, some of the most basic issues with medical data remain unresolved. For example, standards in data formats are a distant pipe dream yet essential. I mean, how hard is to decide and then follow a standard for gender? Does Female = 1, 0, F or 2??? How about a code for "did the patient die during their hospital visit?" Seems a simple binary data field (1 =Yes, 0 = No), yet many systems use complex mortality fields that include the cause of death (often impossible to really know) along with the simple issue of whether the patient died or not. The result of this type of thing was seen in a hospital in Grand Rapids MI. They migrated to a new system and had to translate all the fields and codes from the old system to the new system. The result? 2000+ patients were suddenly indicated as dead and that same information was reported to Medicare where their benefits were cancelled. Problem was, they weren't dead! Oops! :-)

While some people like Hillary argue for standards, others think it will simply never ever happen. There are companies that create amazingly sophisticated software to merge incompatible healthcare databases in order to derive regional or national statistics. http://www.hmsonline.com/ is one (forgive the lousy site and note they are hiring a usability person!). This same type of software has the potential to work with all kinds of data - not just medical data.

I could talk about this subject for weeks on end, but when approaching physicians or healthcare professionals as users, always keep in mind that they tend (like all users) to tell you how they think things should be done vs. what it is they really need from their data and information. This is further complicated from medicine being such a complicated subject! In addition, physicians are notorious for wanting absolutely everything whether they really need it or not. I cannot tell you how many times a physician gave me his/her list of "400 essential variables" for a study of perhaps 120 patients. Also, many physicians do not understand how data can be combined or parsed to answer specific questions. A few, rather than admitting they are not real research scientists, ask for any and all data hoping to simply find a few good p values < .05 in SPSS. Simpler data fields that can be compared/crossed... are the key, yet ICD billing codes are a prime example of consolidating a ton of information into a format where the details that went in simply cannot be extracted out. Here is an interesting book with a section on many issues with classifying medical data: http://www.amazon.com/exec/obidos/tg/detail/-/0262522950/102-6136277-4755332?v=glance

All of this, if it is not understood, often drives unmanageable lists of requirements in any medical information system development. It takes guts (let alone supprting data) to stand up to a doctor and tell them they are mistaken or should consider changing anything about how they do anything and they certainly aren't used to getting this kind of feedback in the healthcare culture, but it is the first and most important step to resolving the gigantic mess that is medical information systems.

Please take my ribbing of physicians in good humor. The docs I worked with at Umich were outstanding people and some of the smartest and most dedicated people I will ever met. Still, they are a little too busy figuring out the complexities of the human body to spend much time with lousy technology and information systems - they need us almost as much as we need them!

P.S. I missed seeing Hillary Clinton up at Marist College in Poughkeepsie just yesterday! I was going to go there for some project work, but decided to call in to the meeting instead!

Comment: Dean Karavite (Feb 10, 2004)

P.S. Don't shoot me for being a largely IA ignorant usabilty person, but while there is certainly a place for IA in medical information, a deep understanding of data, databases and stats is essential. Can you explain to me how IA's approach *data* vs. "information" - I'm sure many similar techniques would apply, but it seems there would be limits.

Comment: Lou (Feb 10, 2004)

Hi Dean, I feel your pain: I've worked with physicians on information systems design too. Nothing like trying to tell God that he's misinformed...

Regarding data and information, you might find this article useful: http://www.ddj.com/documents/s=3340/nam1012432676/

Comment: Dean Karavite (Feb 10, 2004)

I swear I'm not a data bigot!!! Still, I'm not 100% sure - the link "13 Reasons Why Data and Document Retrieval are not The Same" is broken. When and if you have time, I'd like to see this!

Comment: Lou (Feb 10, 2004)

Sorry Dean, I should have realized that link checking was not Doctor Dobbs' strong point. (Boy, all this griping about doctors today!)

Try this one instead: http://www.catdancers.com/webmags/webrevu/1999/04_02/developers/04_02_99_4.html

Comment: Lou (Feb 10, 2004)

Sorry Dean, I should have realized that link checking was not Doctor Dobbs' strong point. (Boy, all this griping about doctors today!)

Try this one instead: http://www.catdancers.com/webmags/webrevu/1999/04_02/developers/04_02_99_4.html

Comment: Matthew Oliphant (Feb 11, 2004)

Another take on the "right information in the hands of the right people at the right time" thought is that this is the same purpose underlying Information Risk Management and Privacy issues.

Not that IA doesn't have a big hand in these issues, which it does of course.

Meh. Random thought.

Comment: Joe (Feb 18, 2004)

Ever since the 2nd IA summit, I've tried (and failed) to get a presentation about a failed project in the summit. The project was one I did for Siemens Medical--it was an IA and usability project. I wanted to present it as a case study of what went wrong, when the IA work was good and the usabililty work was good...and yet the customer rejected it. Similar to Dean's comments about physicians, we were dealing with radiologists, oncologists, ob/gyn, and their support staff.

Maybe I'll write up the case study & post it on my site...


Add a Comment:



URL (optional, but must include http://)

Required: Name, email, and comment.
Want to mention a linked URL? Include http:// before the address.
Want to include bold or italics? Sorry; just use *asterisks* instead.

DAYENU ); } else { // so comments are closed on this entry... print(<<< I_SAID_DAYENU
Comments are now closed for this entry.

Comment spam has forced me to close comment functionality for older entries. However, if you have something vital to add concerning this entry (or its associated comments), please email your sage insights to me (lou [at] louisrosenfeld dot com). I'll make sure your comments are added to the conversation. Sorry for the inconvenience.