These pages use javascript to create fly outs and drop down navigation elements.

HSR&D Study


Sort by:   Current | Completed | DRA | DRE | Keywords | Portfolios/Projects | Centers | QUERI

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

BACKGROUND/RATIONALE:
Project Background:
Computerized Patient Care Documentation (CPD) improves on paper systems, but introduces errors which affect clarity. Error sources include inserted data, difficult formatting and misleading copied text. Error propagation, data overload, and poor readability may result. The VA currently stores over 700 million patient documents. Shortcuts that aid writing can impede readability and lessen the value of CPD. Improving CPD performance while preserving its integrity is desirable. Using information theory, textual Information Content and Informativeness can be calculated. These measurements of "Information Value" may aid in distinguishing information-rich from information-poor text. We will develop these metrics using a large text collection and test their validity with CPD users

OBJECTIVE(S):
The research goal is to identify CPD users' information needs and apply information science to develop valid measures of Information Value in CPD, ultimately promoting safer, more usable patient records. Four aims are proposed:

1: Identify users' concepts of information value, their information needs, barriers to meeting needs and coping strategies.

2. Assess Information Value in a large CPD collection and describe how it is distributed.

3: Assess the association between user perceptions of document quality and measurements of Information Value.

4: Assess the association between Information Value and decision making efficiency in a document review task.

METHODS:
1: Conduct focus groups at five sites with practitioners, nurses, and administrative staff who use CPD daily. Analyze themes to identify information needs, concepts of information value, barriers to meeting information needs and strategies employed to overcome barriers.

2: Index a large sample of CPD produced in VISN 20. Calculate term frequencies, Information Content, and Informativeness. Measure human copying rates. Describe the distribution of these statistics across the VISN, by site,
discipline and department.

3: 180 users at five sites will rate information quality (readability, believability and utility) in test documents constructed to possess varying degrees of Information Value as assessed above, with and without introduced errors, in an unbalanced block design. Test documents will be presented on paper. Correlations between user ratings, document metrics and presence or absence of introduced errors will be measured.

4. In a similar design, 60 practitioners at one site will be given sets of varying documents to review via computer and asked to make care decisions. Task completion time will be compared against measures of document Information Value.

FINDINGS/RESULTS:
The Component 1 focus group interviews provided a rich set of impressions about the use of computerized documents. These interviews provide interlocking and complementary perspectives of administrative, practitioner and nursing users. These impressions influence us to adopt a systems-oriented holistic view of the VA CPD system, and the systems-oriented approach will inform our recommendations for design and policy change.

Preliminary analysis of of the interviews, which is ongoing, validates the impression that issues relating to copy and paste, readability, trustworthiness and consistency in computerized documents are important dimensions of document value held by all user groups. These dimensions, combined with quality dimensions found in the literature, will be used to construct the document quality questionnaire of Component 3.

In addition, we have identified that the document system is extensively used for real time and asynchronous communication between and among practitioners, nurses and administrative staff. This finding will influence our recommendations for system improvement but will not affect the planned questionnaire study.

IMPACT:
Anticipated Impacts: This project establishes a basis for more readable, usable and safer computerized patient documentation (CPD) in the VA. Improved patient record quality will benefit patients, practitioners, and others who review medical records. Demonstration that application of information theory can improve information performance of CPD will guide document design, reduce variability in record quality and facilitate efficient information retrieval. The findings will influence documentation practice, design, and policy.

The finding that communication is a vital part of the CPD process will be added to the recommendations made for system improvement. We will be recommending an explicit approach to supporting the types of clinical and administrative communication that VA staff have report adopting when using the VA patient care documentation system.

PUBLICATIONS:
None at this time.


DRA: Health Services and Systems
DRE: Technology Development and Assessment, Quality of Care, Treatment
Keywords: Quality assessment, Safety, Informatics
MeSH Terms: none