How Mass General is using an automated system to identify patients at risk for pneumonia

Researchers at Boston-based Massachusetts General Hospital have developed an automated system that is 100 percent accurate in predicting patients' development of ventilator-associated pneumonia, according to a study published in Infection Control & Hospital Epidemiology.

Here are seven things to know.

1. The research team said its automated system — using an algorithm created in collaboration with Mass General's Division of Infectious Diseases, Infection Control Unit and the Clinical Data Animation Center — was 100 percent accurate in identifying at-risk patients when given necessary data. 

2. "Ventilator-associated pneumonia is a very serious problem that is estimated to develop in up to half the patients receiving mechanical ventilator support," co-senior author Brandon Westover, MD, PhD, said in a news release. "Many patients die each year from ventilator-associated pneumonia, which can be prevented by following good patient care practices, such as keeping the head of the bed elevated and taking measures to prevent the growth of harmful bacteria in patients' airways."

3. Traditionally, surveillance of patients who receive mechanical ventilation involves manual recording every 12 hours of ventilator settings, which are adjusted at different points during the day to meet the patient's needs. An infection control practitioner reviews these settings for signs of potential ventilator-associated pneumonia. 

4. To cut the time it takes these practitioners to manually record and review ventilator settings and medical charts, the research team created an algorithm to display automated, real-time monitoring of the ventilator settings and the patient's EHR information. Using that data, the algorithm determined whether the patient met the criteria for a ventilator-associated event. 

5. The researchers tested the automated system from January through March of 2015 in four Mass General intensive care units. More than 1,300 patients were admitted to the units during that time. Of these patients, 479 received ventilator support.

6. In a retrospective analysis that compared manual and automated surveillance of data collected from the patients, findings revealed the automated system was 100 percent accurate in detecting ventilator-associated events, distinguishing patients with such events from those without and estimating the development of ventilator-associated pneumonia. The researchers found the accuracy of manual surveillance for those three measures was 40 percent, 89 percent and 70 percent, respectively.  

7. "An automated surveillance system could relieve the manual effort of large-scale surveillance, freeing up more time for clinicians to focus on infection prevention," Dr. Westover said. "Automated surveillance is also much faster than manual surveillance and can be programmed to run as often as desired, which opens the way to using it for clinical monitoring, not just retrospective surveillance."

More articles on clinical leadership and infection control: 
Nurse found dead in car at Georgia hospital
Washington hospital agrees to $16K fine to settle hazardous waste violations
Sepsis shortfalls: How current sepsis care and measurement is lacking

© Copyright ASC COMMUNICATIONS 2019. Interested in LINKING to or REPRINTING this content? View our policies by clicking here.

 


IC Database-3

Top 40 Articles from the Past 6 Months