View source for Improving the usefulness of information in electronic health records; techniques used to capture and structure narrative data.

'''Improving the usefulness of information in electronic health records; techniques used to capture and structure narrative data.'''

In the evolutionary developmental process of electronic health records and clinical information systems, methods of capturing data and entering data from narrative text have been under development. [1,2,3,4] These methods were developed since narrative text in the form of progress notes or admission notes, for example, contains useful information that is often not captured. Historically, physicians enter narratives as free text, which allows for flexibility and expression, however, entering them in a structured form can be a limitation for the provider. The ability to capture data from narrative text or entering data in a structured form allows for its use in both patient care processes, such as in clinical reminders and/or decision support, and research.

An application, called OpenSDE (structured data entry), was developed by medical informatics researchers at the Erasmus University Medical Center in Rotterdam. This program is one method that allows structured entry of data. OpenSDE is an open source application that allows data entry with the use of forms generated using trees of medical concepts.[1,5,6] Users can customize these forms depending on the medical concept. For instance, there is a form for history of present illness. The application can be tailored for use by different medical specialties, i.e. pediatrics or cardiology.  

More recently, another method called structured narrative was developed at Columbia University Medical Center.[4] This method combines unstructured text and coded data using eXtensible Markup Language (XML), the HL7 Clinical Document Architecture (CDA) standard and Natural Language Processing (NLP). This method perhaps has advantages over the OpenSDE method since clinicians are allowed to use free text, which is then marked up, and data is identified by NLP. The particular NLP used by these researchers is called Medical Language Extraction and Encoding System (MedLEE), which analyses medical text and generates structured information. [4,7]

These methods, although relatively new, demonstrate how even more information can be extracted from electronic health records and I expect these applications to become more widespread in the future of clinical information systems.

 Jim Gemelas

References

1 Los RK, van Ginneken AM, de Wilde M, van der Lei J. OpenSDE: Row Modeling Applied to Generic Structured Data Entry. J Am Med Inform Assoc. 2004;11:162–165. doi: 10.1197/jamia.M1375.

2 Johnson SB, Bakken S, Dine D, Hyun S, Mendonça E, Morrison F, Bright T, Van Vleck T, Wrenn J, Stetson P. An electronic health record based on structured narrative. J Am Med Inform Assoc. 2008 Jan-Feb;15(1):54-64. 

3 Los RK, van Ginneken AM, van der Lei J. OpenSDE: a strategy for expressive and flexible structured data entry. Int J Med Inform. 2005;74:481–490. doi: 10.1016/j.ijmedinf.2005.04.005.

4 Bleeker SE, Derksen-Lubsen G, van Ginneken AM, van der Lei J, Moll HA. Structured data entry for narrative data in a broad specialty: patient history and physical examination in pediatrics. BMC Med Inform Decis Mak. 2006;6:29. doi: 10.1186/1472-6947-6-29.

5 OpenSDE. http://Sourceforge.net/projects/opensde

6 Erasmus MC. http://www2.eur.nl/fgg/mi/OpenSDE

7 MedLEE  http://lucid.cpmc.columbia.edu/medlee/
[[category:BMI-512-W-08]]