Physicians Resist CDSSs

I was doing some research for a presentation titled “Overcoming Resistance to Incorporating Technology Into the Hospital Environment” that I am giving shortly (today) and uncovered some really interesting information.

First, getting Doctors to use new technology is a big deal and hard to do. Thus a large number of articles in medical journals are dedicated to the subject. If you don’t believe me about the resistance, check out Wendy Wickham’s post on her blog leading to many stories along this topic. In the Middle of the Curve: Archive of EHR / EMR Posts

Second, one of the newest medical innovations out there are CDSS (Clinical Decision Support Systems) in training we’d call them Electronic Performance Support Systems (EPSS). Same concept, the system provides the doctor with information he or she needs to make decisions on how to treat a specific patient. But here is a really interesting finding:

An empirical study was conducted where physicians were required to use a computer system to solve a number of clinical patient problems with the help of a diagnostic decision support system [CDSS].

Although 75% of the participants believed that computers were useful in daily clinical work, only two-thirds of the group could imagine a theoretical possibility of computers supporting physicians in their diagnosis.

Of greater interest was that, in those cases, where the physicians’ diagnosis was incorrect and the system’s differential diagnosis was correct, the physician did not reconsider his/her own diagnostic opinion. When asked for their opinion about the system’s differential diagnosis, the participants indicted that they found it very useful when the system confirmed their diagnosis but marginally relevant to their judgment when the system refuted their diagnosis

–Ridderikhoff J. and Ven Herk, B. 2000 in article titled “Who is Afraid of the System? Doctors’ Attitude Towards Diagnostic Systems”. International Journal of Medial Informatics as found in E-health Systems Diffusion And Use.

Fascinating, when the system agreed with their diagnosis, it was helpful and correct. When the system disagreed the system was “marginally relevant.”

As we rush to create electronic performance support systems to make learning “part of the workflow,” we may want to consider the human element and think of how we can help our learners to become more comfortable with automated decision support and not just assume that they would “love” a decision support system…maybe they won’t…maybe they will resist the “support” and still perform as they have always performed. As designers of these systems, we need to think of ways to encourage acceptance and use.

Any ideas?
__

Catalog of Recommended Books, Games and Gadgets
Recommended Games and Gadgets
Recommended Books
Content Guide