Even before COVID-19, pneumonia was a leading cause of death from infectious disease in the United States. However, consistently delivering guideline-driven, best-practice care for this illness in the ED can be challenging. The authors of a new study demonstrated that when deployed in the EDs of community hospitals, an electronic decision support tool for pneumonia can improve treatment while dramatically reducing mortality and inpatient utilization.1
Nathan Dean, MD, section chief of pulmonary and critical care medicine at Intermountain Medical Center in Salt Lake City, says while pneumonia is a common illness that frequently brings patients to the ED, it can be tricky to diagnose. “The symptoms overlap with a number of other diagnoses,” he says. “Also, there is a large number of data elements that go into making the diagnosis of pneumonia accurately.”
For instance, while pneumonia may produce cough, fever, shortness of breath, chest pain, a high white blood cell count, and congestion in the lungs, all these symptoms can be attributed to other diagnoses. To make that job easier, the electronic clinical decision support tool for pneumonia (ePNa) could help clinicians assemble all the required data from the electronic medical record (EMR) and think through whether the patient is suffering from pneumonia. The tool synthesizes more than 40 elements, calculates the likelihood the patient has pneumonia, then it displays all the factors.
When ePNa was tested at a group of 16 community hospitals in the Intermountain Healthcare System, the tool was set up to automatically place a P alert on the electronic tracking board whenever the likelihood of pneumonia was 40% or higher. That limit was established in consultation with emergency medicine providers who wanted to make sure they were not alerted all the time. “The physician does have to initiate the program. In the Cerner EMR [at Intermountain], that takes three different clicks to do that,” Dean explains.
Between December 2017 and June 2019, Dean and colleagues found 6,848 pneumonia cases. The tool was used in the care of 67% of eligible patients. Further, the tool was associated with a 38% relative reduction in mortality 30 days after diagnosis. Investigators observed the largest reduction in mortality occurred in patients admitted directly from the ED to the ICU.
Using the ePNa tool also was associated with fewer ICU admissions without safety concerns, a shorter mean time from ED admission to the commencement of antibiotics, and a 17% increase in outpatient disposition for patients. Dean believes ePNa “helps the physicians to feel more confident about sending patients home.” He adds that when ePNa recommended home care during the study, and physicians followed that guidance, the mortality rate was extremely low — in the half of a percent range.
However, Dean emphasizes ePNa is only part of the reason why outcomes improved. “The tool is open loop, meaning that at every point it kind of gathers, organizes, and displays [information], and it makes a recommendation,” he says. “But ... we reviewed thousands of ED records, and I am convinced that it is the combination of the bedside clinician with the decision support tool that makes for the best care and best outcomes.”
Dean acknowledges that whenever you implement a new tool, there are bound to be obstacles, and researchers did face some challenges with ePNa. “You have to convince [providers] that this will improve their care of patients and improve outcomes. Then, you have to show them how [the tool] launches,” he explains. “This is a very advanced electronic decision support tool.”
Jenna Rubin, RN, an emergency nurse, worked with EDs in the participating community hospitals to implement and use ePNa. She notes Intermountain was implementing a new EMR system at roughly the same time the ePNa was rolled out. In some cases, displeasure or frustration with the new EMR affected implementation of ePNa.
“I gave [clinicians] my personal number and my email address and told them if there is an issue, whatever it may be, please let me know. If I don’t know about it, I can’t fix it,” she shares. “Then I would try to my best to respond quickly.”
The regular back and forth with clinicians during the implementation period was not only effective from an educational standpoint, it also indicated ePNa was here to stay. “After that, we would do reviews at [each] facility on a monthly basis, looking at how many times the tool was used and how the tool worked properly,” Rubin says.
In one remarkable case, a patient presented to the ED after falling off a horse; ePNa indicated there was an 80% to 90% chance she had pneumonia. Caregivers disregarded the result, thinking the patient’s lungs were just bruised from her fall, and then sent her home on painkillers. However, when the patient later returned with worsening symptoms, the care team realized the tool was right. Going over such cases with clinicians helped boost utilization.
While Rubin held many face-to-face discussions about ePNa with physicians, she also spoke with the nurses at each facility, noting nurses play an important role in making sure they are charting all the important vital signs and other data that go into the ePNa calculations. Dean concurs nurse input is vital.
“The computer doesn’t know anything unless the nurse puts it in. [Nurses] have got to chart their lung exam, and they’ve got to chart whether the patient is confused,” Dean says.
Dean also stresses how nurses can nudge physicians to pull up the tool in appropriate cases. “Nurses play a role in reminding and facilitating. If they know what is going on and they are on board [with ePNa], that helps us get the physicians to use the tool,” he says.
Dean observes deployment was most challenging in smaller, rural hospitals with EDs mostly staffed by family practice clinicians. “We developed [ePNa] in a way that works better in EDs with ED physicians and where the whole process of care is such that the ED team does their thing, puts in their orders, and then an upstairs physician takes over,” he says.
The ePNa tool has been used in the EDs at four Salt Lake Valley Hospitals, none of which were part of the current study, for more than a decade. This is a more urban region where tool use on appropriate patients is in the 80% range. In smaller facilities, a family practice provider will see the patient in the ED and then take care of the patient on the inpatient side. That did not mesh as well with the tool because it ends with an ED physician order set.2
Dean is working to adapt the ePNa tool for urgent care clinics, which also see many patients with pneumonia. He notes this version may eventually work better for smaller, rural hospitals.
Elsewhere, work has commenced on a version of ePNa that can interface with any EMR. As part of a three-year grant, investigators plan to deploy this version of the tool at Vanderbilt Medical Center and another suburban hospital in the Vanderbilt Health System, both of which use the Epic EMR.
REFERENCES
- Dean NC, Vines CG, Carr JR, et al. A pragmatic stepped-wedge, cluster-controlled trial of real-time pneumonia clinical decision support. Am J Respir Crit Care Med 2022; Mar 8. doi: 10.1164/rccm.202109-2092OC. [Online ahead of print].
- Carr JR, Jones BE, Collingridge DS, et al. Deploying an electronic clinical decision support tool for diagnosis and treatment of pneumonia into rural and critical access hospitals: Utilization, effect on processes of care, and clinician satisfaction. J Rural Health 2022;38:262-269.
The authors of a new study demonstrated that when deployed in the EDs of community hospitals, an electronic decision support tool for pneumonia can improve treatment while dramatically reducing mortality and inpatient utilization.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content