Talk to the Veterans Crisis Line now
U.S. flag
An official website of the United States government

Health Services Research & Development

Veterans Crisis Line Badge
Go to the ORD website
Go to the QUERI website

IIR 05-019 – HSR&D Study

New | Current | Completed | DRA | DRE | Portfolios/Projects | Centers | Career Development Projects

IIR 05-019
Assessing Information Value in Computerized Patient Care Documentation System
Kenric W. Hammond MD
VA Puget Sound Health Care System American Lake Division, Tacoma, WA
Tacoma, WA
Funding Period: July 2007 - September 2010

BACKGROUND/RATIONALE:
Computerized Patient Care Documentation (CPD) improves on paper, but brings challenges. CPD encompasses authenticated electronic text: notes, summaries, reports, orders and problem lists. We focused on the most prevalent form: documents managed by the CPRS Text Integration Utility (TIU). CPD is central to VA business:. Workload and performance must be documented to be credited. CPRS users spend most of their effort working with CPD. CPD is a "shared information space" dominated by "first draft" documents, some of which have mis takes, inconsistencies and inaccuracies. Template, data insertion, boilerplate, and "copy/paste" functionalities faci litate authoring but may compromise readability and document quality. There were 700 million TIU docu ments in 2003 and nearly 2 billion today. CPD is both crucial and imperfect. Qualitative and quantitative approaches were used to define, measure and validate concepts of information value in TIU documents.

OBJECTIVE(S):
(1.) Identify user information needs, value concepts, barriers, and work strategies. (2.) Assess and describe the distribution of information value in a large document corpus. (3.) Correlate perceived document quality and computed Information Value. (4.) Evaluate text mining to identify domestic violence (DV) perpetration.

METHODS:
(1.) Analyze focus interviews at 4 VA sites, extracting themes relating to information needs, processes and value. (129 subjects). (2.) Analyze 12 million TIU documents produced in 3 years at 3 sites and compute relevance metrics. (3.) Administer an instrument to 109 users at 2 sites to rate document quality and test the relationship between perceived and computed information value. (4.) Train a classifier to identify domestic violence (DV) perpetration in a gold standard set of 16,218 TIU documents.

FINDINGS/RESULTS:
(1.) Subjects reported that readability, trust, consistency, relevance and clarity were key dimensions of document value. They rely on CPD to communicate and coordinate work. (2.) The tf.idf statistic corresponds to document importance. Discharge summaries and operative reports scored higher than clinic notes. Physician notes scored higher than nursing notes. (3.) Tf.idf and quality perception correlated significantly. (4.) Compared to a human, a classifier identified DV in documents with lower precision and recall. Only 22% of known DV perpetrators had a chart entry documenting DV perpetration.

IMPACT:
CPD is central to VA healthcare, but user requirements differ and compete. All users want rele-vant, clear docu mentation and read high volumes of text. CPD enables communication and work coordination. Tf.idf correlates with perceived document importance and quality. Text mining is limited by the content of source records. Tf.idf can add value to CPRS by identifying relevant documents. Remote testing methods will aid future research.

PUBLICATIONS:

Journal Articles

  1. Weir CR, Hammond KW, Embi PJ, Efthimiadis EN, Thielke SM, Hedeen AN. An exploration of the impact of computerized patient documentation on clinical collaboration. International journal of medical informatics. 2011 Aug 1; 80(8):e62-71.
  2. Hammond KW, Efthimiadis EN, Weir CR, Embi PJ, Thielke SM, Laundry RM, Hedeen A. Initial Steps toward Validating and Measuring the Quality of Computerized Provider Documentation. AMIA ... Annual Symposium proceedings / AMIA Symposium. AMIA Symposium. 2010 Nov 13; 2010:271-5.
Journal Other

  1. Efthimiadis E, Hammond KW, Laundry R, Thielke S. Developing an EMR Simulator to Assess Users' Perception of Document Quality. [Abstract]. Proceedings of the 43rd Hawaii International Conference on System Sciences. 2010 Jan 5; 1(1):1-9.
  2. Robertson C, Efthimiadis EN, Hammond KW. Nursing documentation usage analysis. AMIA ... Annual Symposium proceedings / AMIA Symposium. AMIA Symposium. 2008 Nov 6; 1112.
Book Chapters

  1. Hammond KW, Weir CR, Efthimiadis EN. Nursing Documentation in a Mature EHR System. In: Nursing and Clinical Informatics: Socio-Technical Approaches. B. Staudinger (Ed). IGI Global; 2008. 73-93 p.
Conference Presentations

  1. Hammond KW, Ethimiadis E, Laundry R, Thielke S, Weir C, Embi P, Hedeen AN. Preliminary Validation of a Method to Measure Information Value in Clinical Documentation. Presented at: Hawaii Annual International Conference on System Sciences; 2011 Jan 1; Poipu, Kauai, Hawaii.
  2. Efthimiadis E, Hammond KW, Laundry R, Thielke S. Developing an EMR Simulator to Assess Users' Perception of Document Quality. Presented at: Hawaii Annual International Conference on System Sciences; 2010 Jan 5; Kauai, HI.
  3. Thielke SM, Hammond KW, Weir CR, Embi PJ, Hedeen AN, Efthimiadis EN. CPRS User Perspectives on Copying and Pasting the Text within the EMR. Poster session presented at: VA HSR&D National Meeting; 2009 Feb 13; Washington, DC.
  4. Cam K, Efthimiadis E, Hammond K. An Investigation on the Use of Computerized Patient Care Documentation: Preliminary Results (CD-ROM) - Proceedings. Paper presented at: Hawaii Annual International Conference on System Sciences; 2008 Jan 7; Waikoloa, HI.


DRA: Health Systems
DRE: Treatment - Observational
Keywords: Informatics, Information Management, Quality assessment, Safety
MeSH Terms: none

Questions about the HSR&D website? Email the Web Team.

Any health information on this website is strictly for informational purposes and is not intended as medical advice. It should not be used to diagnose or treat any condition.