Lack of Good Paradigms and Experts Combine to Slow EHR Evolution

April 22, 2013

Einstein was right when it came to EHRs: Insanity is doing the same thing over and over again and expecting different results.

With EHR as with anything else, there are experts and there are experts. Some become experts by spending hours dealing with a subject on the front-lines, in the trenches. They understand the delayed consequences of early decisions and actions. True experts have knowledge born of both learning and experience.

Others acquire the status of expert by making a career of attending conferences and trade shows, participating in committees and work groups that develop standards and guidelines, and attempt to influence government action, in short - being politically active. Advancement and reputation are often determined more by whom you know than by what you know or the extent of your experience. Those practicing meeting attendance instead of being hard at work on the innards of EHR will become expert meeting-goers but are not likely to become truly expert at EHR.

It takes about 10,000 hours of practice to become an expert. It's no surprise that the number of true EHR experts is relatively small. Complicating matters, several distinct areas of expertise are necessary to move forward with EHR:

1. Conceptual expertise - What should an EHR do and what internal structure and design are necessary?

2. Development expertise - Choosing and using tools, translating concepts into functioning systems

3. Political expertise - Influencing standards and policies, and working to get them adopted.

No single person is likely to be expert in all three areas but it seems logical to me that solid concepts are the prerequisite to successful development; but this is not the way EHR has evolved.

In the 1960s, there was great eagerness to prove that computers could be applied to healthcare. Groups at several institutions each chose a target. For some it was decision support, for others it was interoperability and for many, such as the group at Massachusetts General Hospital (MGH), it was hospital admissions, order entry, and result reporting.

In my opinion, the work at MGH is largely responsible for the paradigm that guides most EHR developers today. I'm sure that they were not thinking abstractly about paradigms when they appropriated the one then in use by their contractor, Bolt, Beranek & Newman, Inc. (BBN). The MGH group was conditioned both by their mentors from BBN and by the very nature of available software models to conceive of their problem in terms that were meaningful in the context of the computing paradigm that prevailed at the time, as they came to understand it.

MGH had early success. Seeing this, many of the emerging vendors adopted both the MGH paradigm and techniques. Gradually, the entire industry became fully committed - effectively locked-in. There is no visible evidence that any significant EHR vendor has considered, or is considering, an alternate paradigm.

There is just one problem. The prevailing computing paradigm is inappropriate for medicine. It's useful for certain highly focused transaction-oriented applications but wrong as the fundamental organizing principle for medical systems. Despite the goal of improving healthcare, today's EHRs are weakest in those areas that are most uniquely medical because the paradigm does not account for them. This is the root cause of the dissatisfaction with today's EHRs and, until a new paradigm emerges, EHRs will continue to disappoint.

Einstein is reputed to have said: "Insanity is doing the same thing over and over again and expecting different results." The past 30 years of following the same EHR paradigm over and over has left us with increased costs, little evidence of benefit, continuing dissatisfaction, and has totally ignored important requirements. Novices in a hurry have gotten us into this mess, it's going to take some experts to get us out, and they are going to need time to do it.