Data Integration: Consistency Concerns & Archival Policies

December 7th, 2012 / By Jill Devrick

For my final entry in this blog series, let’s take a look at some of the issues surrounding consistency and archival policies that come up during the data integration process.

Investigating Consistency

Manually making the data match in integrated systems is an issue of accuracy, as well as consistency.  For example, if the dictation system says the document is an H&P, and the transcription system says it is a consultation, is that a problem?  It might not be major, but it is confusing and can result in wasted time if not addressed while the document is in process.  I have found that consistency becomes an issue more often after the fact than during the document creation process.  It’s when you have to backtrack to find something or do reporting that you find inconsistencies that don’t make a lot of sense to those who were not involved with the document initially.  Be sure to think through those situations when they occur and put a plan in place so that your systems are as clean and uniform as possible.

In addition to having a plan for dealing with inconsistency, you can promote consistency by making data entry and on-the-fly corrections as easy as possible.  For example, evaluate the prompts that physicians must answer when creating a dictation to make sure they are clear, concise, and easy to navigate.  Try to use the same codes, descriptions, and terminology across systems so that making a selection or formatting a response is as straightforward as possible.  Rule of thumb is that the technology that you use to do your job should be intuitive and not add additional stress, so if you or those you observe are frustrated by certain steps in the workflow or with chronic inconsistencies, it would be time well spent to address these issues.

Determining Archival Policies

When an organization uses multiple systems to accomplish all of their workflow needs, it is important to establish the ultimate source of historical information and storage.  Patient data and completed documents need to be retained, but for how long, and where?  For example, if the documents move from the transcription system to the EHR for permanent storage, how long is it necessary for that document to be kept by the transcription system or MTSO?  You definitely need to keep the documents long enough to ensure that they have been finalized and distributed to the appropriate systems and recipients.  You may also want to keep data and documents for a certain amount of time so that you can create statistical reports and track productivity trends over time.  But it is important to do an inventory of all the systems in which a piece of data or a document may reside, and determine the rules for retaining and interacting with that information so that everyone understands and follows the established record retention policies.

As time marches on, the integration between systems will continue to grow more sophisticated.  For example, the technology of natural language processing is making it possible to mine the content of documents and create discrete data fields from what was initially just a blob of text.  I expect that we will see more and more data mining and sharing as documentation requirements become more detailed with ICD-10 and other new standards driving us into the future.  The more you can do now to start thinking about how your current systems interact with others, the better you will be positioned to move forward with health information technology advancements in the future.

Jill Devrick is a Product Solutions Advisor with 3M Health Information Systems.