Structured data entry

From Clinfowiki
Jump to: navigation, search

Introduction

Structured data entry (SDE) is a data entry method by interacting with pre-defined forms. Compared with free text data entry, SDE can constrain clinical data entry behavior, improve data quality, readability etc.

SDE can be divided into two parts:

  • Form designer: in form designer, domain experts work on the new forms by organizing radio buttons, check boxes, and text boxes to present questions under their domain knowledge.
  • Data entry interface: end users will interact with the data entry interface to enter data. The data entry interface is predefined in the form designer.

SDE is preferred because it supports electronic storage and exchange of data that reduces variations in the data and the level of detail. It encourages clinicians to provide standard text without the use of coding.

Reference

  1. Van Bemmel. Musen Mark. Handbook of Medical Informatics. 1997
  2. S. Yamazaki1, Y. Satomura2. Standard Method for Describing an Electronic Patient Record Template: Application of XML to Share Domain Knowledge. Method Inform Med. 2000, 39: 50–5

External Link

OpenSDE: Row Modeling Applied to Generic Structured Data Entry [1]

Structured data entry for narrative in a broad specialty: patient history and physical examinations in pediatrics

Bleeker SE, Derksen-Lubsen G, van Ginneken AM, van der Lei J, Moll HA. Structured data entry for narrative in a broad specialty: patient history and physical examinations in pediatrics. BMC Medical Informatics and Decision Making. 2006, 6:29.

OpenSDE is a structured data entry application created by the Department of Medical Informatics at the Erasmus Medical Center University for the collection of narrative patient data in an electronic medical record. The use of structured data can add to all the known potential benefits of an EMR like legibility, multiple access form various locations; the possibility of using this type of data for clinical decision support systems.

Since medical narrative is diverse and can vary along a patient history or through different specialties, it is a hard task to transform medical narrative to a structured form. Previous works have shown it works in concise specialties like radiology or endoscopy, but broader specialties like internal medicine and pediatrics can be a challenge.

An application of OpenSDE was developed for general pediatrics through the customization of OpenSDE for the data obtained from pediatric history taking and physical examination. Open SDE is organized through hierarchies, the medical terms are nodes in a tree structure, the course from root to node is a medical concept in context. The branches for each node represent its descriptors. Examples of node are body height, vomiting or abdominal pain. Two of the authors developed the concepts and descriptors for pediatric history and physical examination using national standards and pediatric textbook. This was validated by five pediatricians using dummy patients.

Results: The pediatric patient history has 20 medical concepts that include the branches related to past medical history, allergies and current chief complaint among other. This concepts split in 5 to 25 sub-branches and these are described by 4 to 15 attributes. This adds up to 6312 nodes. The physical examination is organized in 11 branches with a total of 2336 nodes.

The full thesaurus has 1800 items that were used in 8648 nodes with a maximum depth of 9 levels. Data entry is usually made by selecting a concept and navigating through the branches, and there is also the possibility of adding free text to a description. The clinician is able to choose up to what degree of detail he/she wishes to input. The final output can be exported to MSWord and used as a letter or summary. In other studies the authors have described that the pediatric OpenSDE is accepted by pediatricians and has completeness and uniformity of data. Comments: Most data obtained from medical history and physical examination is usually obtained as narrative text, this type of data compromises the secondary use of it, because its difficulty for its analysis and codification. The possibility of structuring this information can be useful for more advanced electronic health records that can use the potential benefits of CPOE and CDSS.

Care delivery organizations aspire to increase the content of coded data in their Clinical Information Systems to enhance processes such as: coding for billing purposes; abstracting clinical data for Performance Improvement Efforts; and use in Clinical Decision Support. Unfortunately, care providers find coded data entry cumbersome because it interferes with individualized patient care and workflow. In addition, when compared to traditional natural language narrative, coded entry captures only a fraction of the information produced in the clinical encounter and hence there is a trade-off between sensitivity and specificity in physician coding.

Introduction

There now exist systems which are capable of coding clinical notes. For example, MediClass (a "medical classifier") is a knowledge-based system that automatically classifies the content of clinical notes within an EHR. MediClass is able to code notes by applying a set of application-specific logical rules to the medical concepts that are automatically identified in both the free-text notes and precoded data elements such as medication orders. This system can process data from any EMR system as long as data can be expressed in the Clinical Document Architecture (CDA) data standard that is maintained by Health Level Seven.

The MediClass system was built from open source components and utilizes 3 technologies:

  1. Hl7's CDA for representing the clinical encounter including both structured and unstructured data elements.
  2. Natural language processing techniques for parsing and assigning structured semantic representations to text segments within the CDA.
  3. Knowledge-based systems for processing semantic representations addressing specific subdomains of medicine and clinical care and for defining logical classifications over the semantic contents of a clinical note.

Studies have shown the application to have be similar in accuracy to trained human medical record abstractors; using the trained abstractors as gold standard, the system performed with an average sensitivity of 82% and an average specificity of 93%.

References

  1. Dolin RH, Aschuler L, Beebe C, Biron PV, Boyer SL, Essin D, et al. The HL7 Clinical Document Architecture. J Am Med Inform Assoc. 2001;8:552–69
  2. Gilbert J. Physician data entry: providing options is essential. Health Data. 1998;6:84–92.
  3. Hazlehurst B, Frost HR, Sittig DF, Stevens VJ. MediClass: A system for detecting and classifying encounter-based clinical events in any electronic medical record. J Am Med Inform Assoc. 2005 Sep-Oct;12(5):517-29.
  4. Kaplan B. Reducing barriers to physician data entry for computer-based patient records. Top Heal Inf Manag. 1994;15:24–34.
  5. McDonald C. The barriers to electronic medical record systems and how to overcome them. J Am Med Inform Assoc. 1997;4:213–21.
  6. Schneider EC, Riehl V, Courte-Wienecke S, Eddy DM, Sennett C. Enhancing performance measurement: NCQA's road map for a health information framework. JAMA. 1999;282:1184–90.


Submitted by Paula Otero