TDWI Upside - Where Data Means Business

How Narrative Text Limits Electronic Medical Records Analytics

Automated medical records left healthcare with a glass-half-full conundrum. Why is the world of medicine so slow to mature in terms of technology?

Why has the world of medicine been so technologically immature, and how do electronic medical records hamper analytics?

It wasn’t so long ago that the medical community was focused on building the EMR (electronic medical record). Considerable resources and attention were placed on taking what once were manually created and maintained medical records and automating them so that they could be stored and managed by a computer. In most places -- doctors’ offices, hospitals, insurance companies -- medical records are (or are being) automated today. It is only now that people are starting to realize that many of the promises and expectations of automating medical records have not and are not being met.

People thought they were automating medical records for one purpose. However, now that many medical records have been automated the reality is becoming apparent that a whole different scenario than what was promised is playing out.

For sure, there have been some advantages to the automation of medical records. For example, medical records can be stored inexpensively and accurately for a long time. Another advantage is that records can be easily exchanged among healthcare providers. Furthermore, medical records can be automatically (or at least semiautomatically) captured. These advantages are real and certain with today’s EMR.

In many ways, the EMR of today is a glass that is half full. There is still an empty half. What is missing from today’s EMR records? EMR records don’t really facilitate data analysis. Although EMR records are on a computer, much of the data found in the EMR record is in the form of a narrative. The doctor describes a procedure or a line of treatment, describes a diagnosis, specifies a medication plan, and so forth.

Not all of the medical record is narrative, but major portions are. Although narrative can be captured electronically, text in that form cannot be easily used for analysis. Stated differently, as advanced as the computer is, the computer still chokes on data when working with narrative data.

There are many reasons why the computer stumbles on narration. There is sentence structure, for example, and terminology. There is the barrier of different languages. Most of all, there is context.

When we speak, we take context for granted. All of our lives we have learned to interpret speech, and our brains automatically contextualize what we hear. We don’t consciously think about contextualization. Our brains just do it automatically.

When we commit narrative to paper or a computer, there is no such automatic contextualization process, so the written word does not have the advantage of the human brain’s processing.

The result is that any form of narrative suffers from the lack of contextualization and much of the EMR is narrative. Is it any surprise, then, that merely automating the capture of medical notes -- from the doctor, nurse, or administrator -- leaves a lot to be desired?

It is predictable that the EMR initiative will undergo a maturation process, from automating the capture of data and exchanging data to analyzing it.

It is noteworthy that other organizations have experienced a similar maturation process. The world of medicine is just another discipline when it comes to processing data, but the world of medicine has lagged markedly from other disciplines when it comes to maturation. Stated politely, the world of medicine has matured far more slowly than other worlds, such as finance and banking, insurance processing, and manufacturing.

What is it about the world of medicine that has hindered its progress? There are several factors that have to be taken into account.

Profit motive: Other institutions have a strong motivation for profit. Medicine has always had the care and recovery of the patient as its primary objective. Although patient care as a priority is commendable, in many ways that priority has retarded the forces that lead to technological maturation

No General Motors: The organization of medicine is very different from that of most businesses. Much of medicine is organized around hospitals and doctors’ offices. Although there are a few large medical institutions, there is nothing like the General Motors or the Standard Oils of industry. In a General Motors or in a Standard Oil, a manager can make a decision and the whole company follows in turn. There are no central authoritative bodies in medicine as there are in other industries.

Management by committee: In medicine, most decisions are made by committee (such as a board of directors). Unfortunately, in my experience, committees seldom have vision. To make important decisions, there must be a vision of management.

These are some of the reasons why the world of medicine has lagged behind other communities in terms of maturity. There is, however, one other important factor. The decision making in the world of medicine is almost always left to a doctor. Certainly doctors are the ones that ought to be making decisions when it comes to healthcare. It is truly the rare doctor who is qualified to make visionary technical decisions. Doctors spend their professional lives working with and understanding the human body. Most doctors just don’t have the time or inclination to understand the nuances and advances that are being made in technology. Yet the final technological decisions are being made by doctors when it comes to the profession of medicine.

This then is yet another reason why the world of medicine has been slow to mature in terms of technology.

TDWI Membership

Accelerate Your Projects,
and Your Career

TDWI Members have access to exclusive research reports, publications, communities and training.

Individual, Student, & Team memberships available.