Slow versus Fast Knowledge

By Abraham Verghese

medical charts.JPGThe time of year has come when we interview final year medical students from across the country applying for internships. I experience deja vu when I see the candidates appear in their suits, because we have been doing these interviews twice a week for some weeks. And also because I feel kinship with them, as if I have just been in their shoes, although (I am shocked to realize) it has actually been 25 years.

I try to imagine what these interns-to-be will experience when they start in the wards in six months. I suspect it will be much like what I felt: exciting, overwhelming, scary. What hasn't changed that much is the kinds of problems their patients will present with--the common conditions that manifest with regularity on an internal medicine service: pneumonia, heart failure, cirrhosis and stroke to name a few.   

What has changed dramatically is the fact that the charts (now largely electronic) have tons of information in them, and they mushroom every hour that a patient is in the hospital. The electronic medical record (EMR) contains blood tests of every kind, problem-lists that scroll off the page, medication lists that fill several screens, and notes from an army of people. The great challenge is figuring out what is critical, what is of current importance, and what isn't. (It's very much like the cancer screening problem I wrote about in my last post: so many tests coming back with information that we don't always know what to do with.)  

I finished a stint on the wards two weeks ago, and I start again in a day. I still find the best way to understand a hospitalized patient whose care I am taking over is not by staring at the computer screen (or not by that alone) but by going to see the patient; it's only at the bedside that I can figure out what is important. When I know the patient well, I become aware of how much detritus has accumulated in the chart obscuring the data that does need our attention. (In the health care reform debate, proponents of the EMR don't talk too much about this, implying instead that EMR is a panacea for what ails us.)

David Orr makes a wonderful distinction between "slow" knowledge and "fast" knowledge, which he explains in his book, The Nature of Design.

In brief, fast knowledge (and I am interpreting Orr's work to the medical setting):

  • Celebrates lab tests, imaging, consultations and the more the merrier--you can never have too many tests or images.
  • It suggests that what counts are only the things one can measure (and counts more than the patient's or the family's subjective observations and their verbal reports).
  • It presumes that an error made from misinterpreting the existing data can be overcome with even more data. (More tests can help you claw your way out of a clinical impasse, in other words.)

Slow knowledge by contrast has a different purpose:

  • It celebrates wisdom more than cleverness, a sense of the individuality of the patient and the need for a tailored treatment, rather than one-size-fits-all algorithms.
  • It recognizes that the volume of tests ordered and the results that come back can compound mistakes.
  • It suggests that mistakes are often generated in part by the fact that there is no filtering function to the data.

Clinical medicine is of course art and science; quantitative data and the subjective assessment of how the patient is doing; listening to consultants and listening carefully to the patient and family. It is that balancing act, that human act, that makes medicine so challenging, humbling and rewarding. It is the experience I wish for our future interns.

Photo Credit: Flickr User The National Guard
 

This article available online at:

http://www.theatlantic.com/technology/archive/2009/12/slow-versus-fast-knowledge/31808/