The Healthcare IT Guy has just posted a note based on a recommendation from a reader about a presentation on the web by Zen Chu, a venture capitalist and medical device entrepreneur. It's entitled High-Impact and High-Value Medical Innovation. He makes the point in slide #5 of the set that innovations by physicians are critical for healthcare with "99/100 of the top Medtronic products" invented by them. I found the entire presentation to be original and was particularly impressed by slide #11 that is captioned Cultural Barriers to Medical Innovation. I copy the content of this slide below that lists these barriers:
- Interdisciplinary collaboration takes effort
- Culture of academic and clinical research
- Grants and publishing mentality
- Leads to secrecy and slow progress
- Defining conflicts of interest too broadly
- Fear of failure
- Better to fail quickly and learn faster
- Experimental model bias
- Human data of efficacy paramount
- Only use predictive models that FDA accepts
- Physician's acceptance of current standards of care
- Overvaluing initial intellectual property
Each of these barriers to innovation in medicine is worthy of greater discussion and thought, particularly in relationship to healthcare IT. For various reasons, products in this area seems to suffer from a lack of innovative spirit and efficacy, particularly electronic medical records (EMRs). Office EMRs are particularly bad as measured by adoption rates by physicians. This void has always puzzled me because innovation infuses most other aspects of healthcare and also because most LISs, in contrast, are generally cost-effective and useful products.
I was very interested in Zen Chu's first requirement for medical innovation, the need for interdisciplinary collaboration. Over the years, I have been struck by the fact that EMR deployments seem to evoke a competition among various hospital professional groups including physicians, nurses, and central IT personnel about who will be best served by the system and who will exercise the most control over it. This often results in an over-engineered and customized EMR that proves to be too complicated to manage and serves none of the individual groups adequately. Perhaps because of greater professional cohesion in the clinical labs or because LISs are less complex, such tensions tend to be less apparent in LIS deployments. As a result, failed deployments are extremely rare these days.
Commenting on the success of the LIS vs the EMR:
The common objective of the laboratory is to provide accurate results in a timely and cost effective way. Everyone working in the lab can agree upon the goals, and behave as a part of the overall solution.
The EMR’s goal is not that well defined. Although it is often stated to be improving healthcare for the patient, I believe the working goal is to make each healthcare professional using the system more efficient at doing their job. Naturally, if the goal is to make you personally more efficient, there is a different goal for each person using the EMR.
The mentality of the people involved is very different. In the lab the people ask “What can I do to help with the process?” With the EMR, people ask “What can the software do to make me more efficient?”
Posted by: Rob Bush | June 04, 2008 at 03:18 PM
I wonder if this is a good argument for best-of-breed?
Although that approach without interoperability would lead to less collaboration.
I have wondered if simple collaboration software concerning a patient is the way to go. What if each patient had a wiki concerning them? Too simple perhaps.
Posted by: John Norris | June 04, 2008 at 12:13 PM