HSR&D Home » Research » IIR 98-213 – HSR&D Study
Application of Clinical Decision Aids for Acute Upper GI Hemorrhage
Thomas F. Imperiale, MD
Richard L. Roudebush VA Medical Center, Indianapolis, IN
Funding Period: October 1999 - March 2003
Clinical decisions aids (CDAs) are tools to assist medical decision-making. Since CDAs either provide a probability of disease or outcome, or suggest a diagnostic or therapeutic course of action, they are often the basis for practice guidelines. Despite the proliferation of CDAs and practice guidelines, their direct application to clinical practice has been problematic. Several CDAs have been developed for risk stratification and management of acute upper gastrointestinal (GI) hemorrhage, although none is endorsed by the physician groups who care for these patients despite the fact that several have been validated internally - that is, in the same setting in which they were developed. Further, there are no data on a head-to-head comparison of how CDAs for acute upper GI hemorrhage perform in the same population. Moreover, none has been developed or tested in the VA hospital setting. This study focuses on both external validation and head-to-head comparison of six CDAs in the VA population.
The objectives of this project are to: 1) test six previously published CDAs to determine their external validity and to compare their performance when applied to the same population; 2) derive and internally validate a VA population-specific CDA using clinical and endoscopic data; 3) compare the best published CDAs with the derived population-specific CDA; and; 4) compare suggested versus observed length of stay (LOS) according to a CDA-derived algorithm. Research design is a prospective cohort study with VAMCs in Indianapolis, Seattle and Durham.
Veteran patients admitted for acute upper GI hemorrhage are followed by research nurses (RNs), who will record the clinical and endoscopic data required to score each patient for each of six published CDAs. The RNs follow patients’ hospital course for complications related to GI hemorrhage, such as rebleeding and other poor outcomes as designated by the individual CDAs. A CDA-derived algorithm is applied to determine suggested LOS. Thirty-day follow up status is assessed through telephone interview with patients. Patients who are hospitalized at other (non-VA) institutions within the follow up period sign a release of information to facilitate complete and accurate follow up information regarding cause for readmission.
Enrollment of veteran patients commenced October 1, 1999 at the Indianapolis and Durham sites and December 1, 1999 at the Seattle site. Currently, Indianapolis has 118 inclusions and 212 exclusions; Durham, 103 inclusions and 231 exclusions; Seattle, 128 inclusions and 198 exclusions. Post-hospital discharge follow up data collection is available for 283 (81%) of 351 patients. Among 330 analyzable subjects, some kind of poor outcome (rebleeding, urgent surgery, death during hospital stay, or new or worsening comorbidity) occurred in 102 (31%), rebleeding alone in 54 (16%), and hospital mortality in 12 (3.6%).
As efficiency of care and implementation of practice guidelines have become priorities for the Department of Veterans Affairs, this research will produce: 1) performance measures of published CDAs for acute upper GI hemorrhage in the VA hospital setting; 2) a VA-specific CDA; 3) a VA-specific guideline for management of acute upper GI hemorrhage; 4) an estimate of reduction of LOS with CDA application; and 5) a model for validation of CDAs in clinical practice.
External Links for this Project
Dimensions for VADimensions for VA is a web-based tool available to VA staff that enables detailed searches of published research and research projects.
If you have VA-Intranet access, click here for more information vaww.hsrd.research.va.gov/dimensions/
VA staff not currently on the VA network can access Dimensions by registering for an account using their VA email address. Search Dimensions for this project
DRA: Health Systems
DRE: Treatment - Observational
Keywords: Acute illness, Decision support
MeSH Terms: none