Mass General system finds patients at risk for ventilator-associated pneumonia
Ventilator and EHR data is monitored in real time instead of manually recording and reviewing information.
Massachusetts General Hospital is using an automated system that better identifies patients at risk for ventilator-associated pneumonia than traditional surveillance methods, which rely on manual recording and interpretation of individual patient data.
MGH researchers developed an algorithm to provide automated, real-time monitoring of both ventilator settings and electronic health record data in order to reduce the time required to manually record and review the information.
In fact, an algorithm—jointly developed by MGH’s Division of Infectious Diseases, Infection Control Unit and Clinical Data Animation Center—was found to be 100 percent accurate in identifying at-risk patients when provided with the necessary data.
“Manual surveillance is vulnerable to human error,” conclude the authors in a study published online last week in the journal Infection Control & Hospital Epidemiology. “Automated surveillance is more accurate and more efficient for (ventilator-associated event) surveillance.”
Also See: Machine learning, EHR data helping to combat hospital infections
“We had to figure out how to get access to data from lots of different sources, including ventilators, laboratory values and medications, bringing it together to automate what was previously a fairly tedious and error-prone task,” says Brandon Westover, MD, in the MGH Department of Neurology, who is director of the Clinical Data Animation Center and co-senior author of the study.
Erica Shenoy, MD, of the MGH Division of Infectious Diseases, who is lead author of the study and hospital epidemiology lead for Clinical Data Animation Center, adds that manual surveillance of ventilator-associated pneumonia in the hospital is a labor-intensive process. With an automated system, “we can focus our efforts on the prevention piece and let the computer do the work that we would normally be doing using print outs and trying to apply the definition in a manual way,” notes Shenoy.
The algorithm determines whether criteria are met for three ventilator-associated events—ventilator-associated condition (VAC), infection-related ventilator-associated complication (IVAC) and possible ventilator-associated pneumonia (PVAP)—as defined by updated surveillance standards issued in 2013 by the Centers for Disease Control and Prevention’s National Health and Safety Network.
“People were previously doing it by hand to check the definition,” observes Westover. “What we did was wrote computer programs and created pipelines for gathering the data together to compute this automatically.”
“Any time you have manual input, you’re going to have the possibility of error, and it also can slow down your ability to do your detections,” concludes Shenoy. “What our team was able to do was actually pull the ventilator data directly off of the machines and incorporate that into the algorithm.”
According to Westover, among the ventilator data the algorithm leverages are the pressure required to keep a patient’s lungs open at the end of a breath and the percentage of oxygen being delivered to the patient. In addition, EHR data such as white blood cell counts and changes in body temperature are factored, he adds.
“Since the completion of the study, our institution has transitioned to a new EHR in which vital signs, including temperature, are documented electronically; thus, temperature data have been added to the algorithm,” state the authors.
MGH researchers developed an algorithm to provide automated, real-time monitoring of both ventilator settings and electronic health record data in order to reduce the time required to manually record and review the information.
In fact, an algorithm—jointly developed by MGH’s Division of Infectious Diseases, Infection Control Unit and Clinical Data Animation Center—was found to be 100 percent accurate in identifying at-risk patients when provided with the necessary data.
“Manual surveillance is vulnerable to human error,” conclude the authors in a study published online last week in the journal Infection Control & Hospital Epidemiology. “Automated surveillance is more accurate and more efficient for (ventilator-associated event) surveillance.”
Also See: Machine learning, EHR data helping to combat hospital infections“We had to figure out how to get access to data from lots of different sources, including ventilators, laboratory values and medications, bringing it together to automate what was previously a fairly tedious and error-prone task,” says Brandon Westover, MD, in the MGH Department of Neurology, who is director of the Clinical Data Animation Center and co-senior author of the study.
Erica Shenoy, MD, of the MGH Division of Infectious Diseases, who is lead author of the study and hospital epidemiology lead for Clinical Data Animation Center, adds that manual surveillance of ventilator-associated pneumonia in the hospital is a labor-intensive process. With an automated system, “we can focus our efforts on the prevention piece and let the computer do the work that we would normally be doing using print outs and trying to apply the definition in a manual way,” notes Shenoy.
The algorithm determines whether criteria are met for three ventilator-associated events—ventilator-associated condition (VAC), infection-related ventilator-associated complication (IVAC) and possible ventilator-associated pneumonia (PVAP)—as defined by updated surveillance standards issued in 2013 by the Centers for Disease Control and Prevention’s National Health and Safety Network.
“People were previously doing it by hand to check the definition,” observes Westover. “What we did was wrote computer programs and created pipelines for gathering the data together to compute this automatically.”
“Any time you have manual input, you’re going to have the possibility of error, and it also can slow down your ability to do your detections,” concludes Shenoy. “What our team was able to do was actually pull the ventilator data directly off of the machines and incorporate that into the algorithm.”
According to Westover, among the ventilator data the algorithm leverages are the pressure required to keep a patient’s lungs open at the end of a breath and the percentage of oxygen being delivered to the patient. In addition, EHR data such as white blood cell counts and changes in body temperature are factored, he adds.
“Since the completion of the study, our institution has transitioned to a new EHR in which vital signs, including temperature, are documented electronically; thus, temperature data have been added to the algorithm,” state the authors.
More for you
Loading data for hdm_tax_topic #better-outcomes...