The technology relating to health records is moving along rapidly having been mandated by HCFA to get the ball rolling.  Millions of dollars have been allocated and spent to create useful electronic medical records and efforts are continuing to determine what and how information will be retained and accessible.  The government is deeply involved in promulgating and monitoring the progress.  Government agencies are interested in information but concerns about how records will be protected remain a primary issue.

Here at www.med-certification.com, we stay on the cutting edge of training and certification to make sure curricula and testing address the innovative happenings in health care.  Here’s the latest:

Electronic Health Records Data Mining

You hear much about the whole migration of medical records to Electronic Health Records, but so far, the information is largely unstructured.  It is often referred to as simply “text blob.”  Within all that information is extraordinarily valuable data, clinical, statistical, predictive (outcomes) and even the expanded coding and billing information.  Technical people and of course providers are intrigued by the challenge of unlocking the salient data from those records.

Natural Language Processing (NLP)

Natural Language Processing (NLP) is a technology powerful enough to unlock specific data from the EHRs.  It serves nicely as a query tool, for example, if a patient is taking a specific medication, it would be possible for the user to find the prescription in question by using a pull down menu and selecting the drug, dosage, route of administration, and any potential drug interactions.

Keep in mind that the provider concerns about EHRs relate to losing the historic unique patient histories because of the structured template requirements established at the current time (as a bare minimum).  The physicians state that using the provider narrative (interpretative and informative) is critical to the elements of patient care.  They rate that need on a scale of 100 at 94% importance.  Their need to be able to access the information on the EHR side is critical to patient care.  The structured elements of the record are pretty simple to analyze and derive information from but less than adequate for the provider treating the patient since the unique patient data is not available under the current template.  Natural speech documentation, combined with the NLP is an alternative that delivers the ability for the provider to tell the complete story and then the facts are all deliverable in the best possible way.

Clinical Language Understanding (CLU)

In the medical field, Clinical Language Understanding (CLU) is a methodology tuned to identify various medical facts so the NLP process can better understand what the provider said.  It has properties like a relational database with intelligence enough to determine for example that “diabetes” is a disease and would then be able to populate that fact into the data stream.  CLU would allow the provider to better efficiency with documentation on input and retrieval.  Hopefully, the purely structured EHR (with its relative uselessness for standards of good patient care) would benefit tremendously from the CLU process allowing doctors to provide and review comprehensive medical records.

Impact on Medical Coding and Billing:

The NLP technology allows for immediate and retrospective analysis of the patient encounter by querying data used in the record.  Typically, when a provider uses vague information to document services, the coder is stuck trying to make sure the service is fully documented to support the codes selected to represent the service provided.  With doctors’  busy schedules, some time may have lapsed since the patient was actually seen,  so if the coder asks for more detailed information, the provider may simply not remember all of it.  However, if s/he were working with the NLP documentation process and the CLU interpretative technology, more specificity might have been requested and incorporated into the record at the time it was being generated.  For instance, if an orthopedist noted a fracture of the leg, the queries could have included which bones and which leg.  By prompting for more information while actually working on the record, the results are far more reliable, eliminating the need for retrospective attempts to pin down with more precision what the encounter actually entailed.

Predictive Care (and outcome analysis)

The benefit for predictive care is enormous.  Using technology, the provider might be prompted to analyze a prescription just written if the software reminded him/her of some potential drug interactive process.  Using the potential of immediate feedback would be a powerful tool for the provider and would certainly enhance the quality of care.

Summary

All of the potentials are clearly on the back burner and some elements are even available now, but the success of NLP in both documentation and improved care will hinge on high quality information input.  To date, the attempts have primarily been based on structured data based on templates.  The next step is to make sure the digital data provided is used to its fully maximum potential.  Any information is valuable for sure but does not tell the whole story.  Since natural speech is now so much a part of the healthcare provision process, the tools to make the provider information must be implemented.  When the process is workable, the benefits will be huge, not only for providers and payers, but for the patient.

It’s an exciting new world for healthcare career workers and those thinking about entering the field.  If you want to learn about transcription, medical coding, billing, office assistant/specialist, and/or medical office administration, we have what you need to succeed.