Related Experiment Videos
Automated attention flags in chronic disease care planning.
J R Warren1, J T Noone, B J Smith
1Advanced Computing Research Centre, University of South Australia, Adelaide. warren@cs.unisa.edu.au
The Medical Journal of Australia
|October 23, 2001
Summary
Computerised decision support software demonstrated moderate to good agreement with specialist recommendations for chronic respiratory disease management. This tool may improve general practitioners' access to care guidelines.
Area of Science:
- Health Informatics
- Respiratory Medicine
- Clinical Decision Support
Background:
- Chronic respiratory disease management requires adherence to complex care guidelines.
- General practitioners (GPs) act as care coordinators but may face challenges in consistently applying guidelines.
- Computerised decision support systems offer potential to enhance guideline implementation.
Purpose of the Study:
- To evaluate the effectiveness of computerised decision support software in managing chronic respiratory disease.
- To compare the agreement of specialist recommendations with those of GPs and decision support software.
Main Methods:
- Care guidelines for chronic obstructive pulmonary disease (COPD) were developed.
- Decision support software (Care Plan On-Line - CPOL) was created to flag guideline-based interventions.
- Decisions on nine additional services for 20 COPD patients were compared between respiratory specialists, GPs, and CPOL.
Main Results:
- Specialist agreement was moderately low (ICC=0.48).
- CPOL showed moderate to good agreement with specialists (kappa=0.72), outperforming GPs (kappa=0.49).
- CPOL was 87% accurate in indicating specialist decisions, with a 16% false-positive rate.
Conclusions:
- Automated decision support (CPOL) can enhance GPs' access to guideline intent in chronic respiratory disease management.
- Further research is needed to fully establish the value of computerised decision support in this context.