Physician Satisfaction with Order Entry Systems

To the Editor:—I read with interest the recent article by Murff and Kannry on physician satisfaction with computerized order entry systems.¹ In brief, house staff were fairly satisfied with one system, and displeased with another. The authors conclude that user satisfaction is an important marker for the usability of any system, and that not all order entry systems are created equal. They further stress that only well-received, high-quality order entry systems should be implemented.

When I was a resident in 1990, our tertiary-care teaching hospital in Calgary, Canada, installed a commercially available patient information system, complete with physician order entry. The institution's official name for the system was OSCAR (Online System for Communication and Records), but it soon became known as FRED (Friggin' Ridiculous Electronic Device). Our travails with FRED were reported in the medical literature²; the University of Virginia had a similar experience with the same system around the same time.³

Similar to work reported by Murff and Kannry, a survey of residents at the hospital was performed after FRED had been in place for a year. About 120 residents were surveyed; more than 90 questionnaires were returned, for a response rate of 75 percent. Of those who completed the survey, 90 percent reported that working with FRED took more time than working with paper charts, 95 percent thought the system was difficult to use, and 63 percent thought the quality of patient care had declined since its implementation. Rating overall satisfaction with FRED on a scale of 1 to 10, residents gave it a mean score of 2.4 (95% CI, 1.4–3.4).

In written comments returned with the survey, residents frequently remarked that FRED appeared to be an antiquated system. At the time, I had just purchased a state-of-the-art 386 personal computer, with a sizzling 16 MHz chip and a whopping 2 MB of RAM. The awesome power and speed of my desktop machine put ol' mainframe FRED to shame. I predicted confidently that FRED, like the Edsel and disco boots, would soon be consigned to the dustbin of history. A few years later, as I completed a master's degree in medical informatics at the University of Utah, I discovered that many of the faculty were familiar with FRED and largely shared my impressions.

Imagine, therefore, my surprise when I read in the paper by Murff and Kannry that FRED is alive and well and living in New York. Although not named specifically, FRED is instantly recognizable by its pathognomonic clunky user screens, shown in Figure 1 of their paper. According to the article, this antediluvian system was implemented and its use made mandatory in the Mount Sinai Hospital in 1999, almost a decade after the fiasco in our hospital. Predictably, the results of the survey among Mount Sinai house staff closely parallel those of our assessment 10 years earlier.

Surely the choice of FRED so recently was a triumph of marketing over prudence, since far superior systems now exist. I wonder if the decision makers at Mount Sinai were aware of the Calgary and Virginia experiences with FRED, both published in medical journals. If not, I doubt that the vendor took the time or effort to inform them.

I believe strongly in the concept of computerized physician order entry, but think that the physician community will make the migration from paper-based records only if they are enticed by carrots, not beaten by sticks. As noted by Murff and Kannry, user satisfaction is crucial to acceptance of hospital information systems, and it, in turn, is dependent on intuitive and responsive software. Unfortunately, inefficient proprietary systems such as FRED do little to impress doctors and serve only to give the specialty of medical informatics a bad name.—robert patterson, md, msc, frcsc