Automated monitoring of EHRs alerts physicians to clinical decline
- A study involving 300,000 hospitalized patients across 24 Banner Health hospitals over 1.5 years showed that an algorithm capable of real-time monitoring of electronic health records (EHRs) was able to alert physicians about a patient's clinical decline.
- The algorithm searches for a minimum of two of the four systemic inflammatory response syndrome criteria, and at least one of 14 acute organ dysfunction parameters.
- It was successfully implemented for the early identification of sepsis and lead author of the study Dr. Hargobind Khurana said the algorithm was a reason why the sepsis mortality rate at Banner is "better than what is expected nationally."
The algorithm was able to identify the majority of high-risk patients within 48 hours of admission, which led to early medical intervention.
Dr. Khurana also indicated there are additional benefits to EHRs, and that the algorithm enables alerts to be brought to the patient bedside "where they are needed most." The technology can help improve patient outcomes and has the potential to become more precise, according to Dr. Khurana.
A 2014 study published in JAMA found that although sepsis occurs in about 10% of hospital patients, it contributes to half of all hospital deaths.
The CDC estimates 750,000 hospitalized patients are affected by sepsis every year in the U.S.