Now Reading: Challenge Paper – Failure to Provide Clinicians Useful IT Systems: Opportunities to Leapfrog Current Technologies

While at Johns Hopkins, I spoke with researcher Marion Ball, Ed.D. who asked me to read the challenge paper she authored regarding the failure of current clinical information systems to support health care providers, which I was happy to do.

Dr. Ball and I began our conversation about the fact that HIT adoption among physicians in the United States is at a level that most leaders consider unacceptable. The paper talks about the importance of human factors research in creating usable Health IT systems. The premise (or the challenge) is that this is the principal reason why adoption hasn’t taken off.

The paper cites two corollary articles that I also read as part of this review that touch on an emerging issue of patient safety being impacted negatively by HIT systems. See:

  1. Michael I. Harrison, Ross Koppel, and Shirly Bar-Lev, “Unintended Consequences of Information Technologies in Health Care An Interactive Sociotechnical Analysis,” J Am Med Inform Assoc 14, no. 5 (September 1, 2007): 542-549,
  2. Jonathan P. Weiner et al., “”e-Iatrogenesis”: The Most Critical Unintended Consequence of CPOE and other HIT,” J Am Med Inform Assoc 14, no. 3 (May 1, 2007): 387-388,

The latter paper cited here brings in its own challenge, in attempting to coin the term e-iatrogenesis. On that point, I thought about my own experience working with patient safety issues in HIT, and I am not sure that this is a good term for this issue. Some issues are physician dependent, some are IT system dependent, and some are management system dependent. I think HIT Patient Safety more accurate. However, the issue is a serious one, and I am both sorry I missed these papers in print, and happy that it is being described now. We should not assume that HIT by definition always enhances patient safety. Any system can be challenging to a patient’s well being if not implemented and monitored closely.

In the background of all of this, the question is a deeper one, then: “What is responsible for the low adoption of HIT in physician practices?”

I believe Dr. Ball’s points are well articulated and do reflect somewhat of a reality in the health information technology sector, that human factors are not as well studied and implemented in this industry. Something that I think adds to this challenge is that some vendors work to market and differentiate their products based on the interface, which might lend to reduced standardization across medicine. Imagine that a community physician practices in an ambulatory medical center with one EHR, a hospital with another EHR, another hospital with another EHR, all in the same day. It happens.

What I am not sure of is whether IT system design is the principal reason, especially in a complex adaptive system like health care. In the system I have worked in, we have tackled issues of bringing clinicians into the design process, and even a leadership approach that involves their input and experience across continuous improvement in general (see this post from my DailyKaizen blog for an example). However, to support the paper’s assertions, even with a management system like LEAN, there are aspects of the human factors environment that we cannot control with a vendor purchased system. Also, we should recognize that many of the personal health record systems that communicate with EHRs have had extensive human factors research behind them. As I found out when I talked with Northern California HIMSS, there is a lot of experience outside of health care around serving customers using IT.

I would probably create a fishbone diagram that shows the contributions of human factors in systems themselves, the leadership and management approach, and the external environment as contributions to the root cause of “low adoption.” I think the paper does an excellent job describing the human factors problem, and I recommend it as a read for those who want to understand it better. At the same time, knowing what I do about some of the IT failures cited in the paper, there was more to what happened than poor system design. Even the best designed systems can fail as a result of the management system and continuous improvement methodology (or lack thereof). HIT, after all, is just an enabler of a great health care system.

There should be a robust partnership between human factors experts, clinicians, and business experts to make this successful (I’m envisioning a big A3 document here).

Thanks to Dr. Ball and her colleagues for writing about how we can do better in the provision of usable systems for our patients and our providers.

Ted Eytan, MD