CPOE: it sounds like a sci-fi robot, but it's actually the computerized physician order entry system that could revolutionize the way hospitals function. You may already be familiar with CPOE, as it's currently used in about 30% of U.S. hospitals. If not, the system expedites processes like ordering tests and filling prescriptions. These tasks are then made instantly—and even remotely—available to authorized hospital staff. Theoretically, it reduces the margin of error, but the merits of CPOE have not been studied in depth until recently.
In early May, Pediatrics published the results of a year-anda-half-long study conducted by Stanford University and Lucile Packard Children's Hospital in California, showing a 20% decrease in fatalities during that period. The study analyzed 100,000 Packard Children's Hospital discharges from January 2001 through April 2009. CPOE was introduced in 2007, and between that time and the study's conclusion, researchers observed the improvement. Researchers compared the observed mortality with the expected mortality, using data from 42 other nonprofi t pediatric hospitals comparable to Packard Children's.
Stanford representatives say their research is the first to show a connection between the implementation of CPOE and decreased fatalities; however, they also acknowledged that one study does not serve as conclusive evidence and that other health care initiatives at the hospital could have played a role in the improved mortality rate. Critics have questioned the value of CPOE since its fi rst use during the 1970s, and one Pittsburgh hospital using the system actually saw an increase in fatalities in their pediatric ICU in 2005.
Implementing CPOE can take years and requires ample training. Perceived risks include delays in workflow due to inexperienced staff members using the system, a lack of communication between doctors and nurses, default settings that may overlook unique patient needs, and even an over-reliance on the automation and subsequent false sense of security. CPOE also offers electronic "suggestions" when it senses a potential error, and critics have argued that a barrage of such messages will eventually lose their potency as health care professionals fi nd they can usually ignore them.
However, the study's results are still encouraging. The hospital appeared to experience an influx of children with more serious illnesses and conditions during the period, and fatalities continued to decrease. (Health care professionals might remember the 1999 Institute of Medicine report claiming medical errors caused 44,000–98,000 deaths every year in the United States, stimulating the adoption of computerized systems.)
Stanford researchers credit the hospital's success to the carefully planned implementation of the CPOE system, and they recommend a similar approach among health care providers going forward. According to the research, Packard Children's Hospital experienced two fewer deaths per 1,000 discharges—36 lives over 18 months.