- Clinical Trials
- Research News
- Industry Trends
- Agency Actions
- Drug Safety Issues
- Approvals, Launches, & New Indications
- Health Care Reform
Electronic Decision-Support Tool Helps Reduce Deaths From Pneumonia in Emergency Departments
Tool could lead to improvements in pneumonia care and outcomes, researchers say (September 10)
Although guidelines for treating pneumonia exist, it is often difficult for these to be fully implemented in an emergency setting. Researchers at the University of Utah therefore developed an electronic decision-support tool, linked to the patient’s medical record, that automatically extracts data predicting the severity of pneumonia. The tool then provides recommendations regarding where the patient should be admitted, which diagnostic tests are best to use, and which antibiotics are most appropriate.
The researchers tested the effectiveness of the tool in pneumonia patients in seven emergency departments (EDs). An initial group of 2,308 patients was analyzed before the electronic tool was used. A subsequent group of 2,450 patients was assessed when four of the seven EDs used the electronic tool.
In both groups, the researchers looked at hospital admission rates, the length of hospital stays, deaths, secondary hospitalization rates, and adherence to guidelines.
The results showed a significant reduction in death rates in the EDs where the decision-support tool was used. The crude inpatient mortality rate fell from 5.3% to 3.5%, and after adjusting for severity, the relative risk of death was reduced by 25%.
Senior author Dr. Nathan Dean said: “Although doctors are free to choose at any time whether to follow the recommendations, we think that a tool that is individualized and integrated into the electronic health record is a much more efficient way of supporting decision-making and making treatment guidelines quickly accessible during an emergency situation.”
The new study will be presented Sept. 11 at the European Respiratory Society (ERS) Annual Congress in Barcelona.
Source: EurekAlert; September 10, 2103.