JavaScript DHTML Drop Down Menu By Milonic

Welcome to the

Media Press Room

  • Normal Size Larger Size Largest SizeText Size
  • Print this Page
  • Email this Page
  • Bookmark this Page
Press Media Alerts

If you're a credentialed journalist for a media outlet, you can receive the latest issues and topics in food and nutrition delivered direct to your inbox.



Nutrition Informatics Blog



Meaningful Use: Data Integrity vs Data Specificity

Submitted by John W. Snyder, DTR, RD

While working to implement the Health Level 7 (HL7) Role Code terminology set used to define the relationship to patient for family medical history, a question came up regarding how to map the end-user “pick-list” selections to HL7 terms. The vendor had not provided a full copy of the terminology set based on terms contained in OID 2.16.840.1.113883.5.111 and asked the following question:

What is the difference between “Son/Daughter” and “Natural Son/Daughter”?

The HL7 Role Code terminology set works as follows:

·   Child (Level 1)

  Son (Level 2)

oNatural Son (Level 3)

oAdopted Son (Level 3)

oFoster Son (Level 3)

oStep Son (level 3)

  Daughter (Level 2)

oNatural Daughter (Level 3)

oAdopted Daughter (Level 3)

oFoster Daughter (Level 3)

oStep Daughter (Level 3)


In an ideal world, all electronic health record systems (EHR-S) would provide users every possible selection, and users would select the correct value. The reality is that providing users with 80+ different types of relationships to select from when entering family history information is perhaps not realistic, and the Centers for Medicare & Medicaid Services has not mandated a level of specificity that all EHR-S must achieve for certification. This means we must map data selected by an end-user to a value in a terminology set.

In the above example, mapping a generic to a more specific (eg, Son/Daughter→Natural Son/Daughter) is not considered a good practice, because it creates invalid data. In this particular case, mapping Son/Daughter to Natural Son/Daughter has only about a 25% chance of accuracy. Mapping from the more specific to a generic (eg, Natural Son/Daughter→Son/Daughter) is an acceptable practice, because downstream it does not invalidate data even though it removes specificity.

The goal of gathering this information is to populate Summary of Care/Continuity of Care documents for interoperability, provide meaningful use data to the government via the Quality Reporting Document Architecture standard, research, and likely some other uses I have not thought to include. In all of these uses, it is necessary to consider sacrificing data specificity as an industry best practice in order to maintain data integrity/validity needs. 


Nutrition Informatics Blog
Blog Home
Subscribe to blog
Get the Blog RSS Feed

Search Blog
Most Recent Posts
Recent Comments

Member Comment Requested: EHR/PHR Nutrition Best Practices Implementation Guide

11/2/2013 9:54:34 AM
Lindsey, I am interested in reviewing this document but am unable to access it even when logged ... more »

Member Comment Requested: EHR/PHR Nutrition Best Practices Implementation Guide

11/2/2013 8:29:21 AM
the webauthor link isn't working

An Informatics Solution to Productivity Tracking

10/31/2013 7:16:49 AM
Is it possible to share this tool with us?

An Informatics Solution to Productivity Tracking

10/16/2013 8:49:19 AM
Can you share? I would like to be able to share with intern how sites are tracking.

Who needs standards? In Health Care?

7/9/2013 9:13:00 AM
I couldn't agree more. Margaret and Elaine, along with other subject matter experts, are doing a fan... more »

Useful Links