Format

Send to

Choose Destination
Int J Med Inform. 2010 Dec;79(12):840-8. doi: 10.1016/j.ijmedinf.2010.08.005.

Automating pressure ulcer risk assessment using documented patient data.

Author information

1
Division of Biomedical Informatics, Department of Medicine, University of California, San Diego, La Jolla, CA 92093-0671, USA. hyeoneui@ucsd.edu

Abstract

OBJECTIVE:

A rule-based prototype decision support tool; Braden-scale based Automated Risk-assessment Tool (BART) was developed to test whether pressure ulcer risk scores can be determined automatically based on the documented patient data.

METHODS:

The data items required for assessing pressure ulcer risk were identified by analyzing the parameter definitions of the Braden scale and by consulting the nurses specialized in pressure ulcer prevention and care. Documentation coverage and formats of the required data was evaluated. The decision rules were developed based on the inputs from the expert nurses, and were implemented as a web-based prototype tool, BART. The agreement rates between nurses and BART on assigning scores to the six Braden-scale parameters were calculated with 39 convenience samples of patient data.

RESULTS:

Although several items required for the automated decision were not found from the documentation, the majority of the required data items were documented with feasible formats (i.e., coded lists or free text with nominal or numeric values) for algorithmic processing. When evaluated with 39 test cases, BART and the nurses showed varying levels of agreement (from "slight" to "substantial") on assigning scores for the six parameters of the Braden scale. They showed "fair" level of agreement with an "at risk" decision.

CONCLUSION:

BART has limitations that need to be addressed through future enhancements. However, it demonstrates potential for reuse of documented patient data to automatically populate pressure ulcer risk using the Braden scale.

PMID:
20869303
DOI:
10.1016/j.ijmedinf.2010.08.005
[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Elsevier Science
Loading ...
Support Center