Here’s a kind of story that makes you feel better about your EHR investment. A new journal article is reporting that researchers were able to find a source of Clostridium difficile within a hospital, not with elaborate big data analytics but simply by using basic EHR data.
According to the item, which appeared in JAMA Internal Medicine, a group of researchers examined EHR data on time and location to map roughly 435,000 patient location changes at the University of California San Francisco Medical Center. The effort was led by Russ Cucina, chief health information officer at UCSF.
After analyzing overall data, the researchers found a total of 1,152 cases of laboratory-documented CDI. The data indicated that CDI-positive patients moved through an average of four locations during their hospitalization, but that the CDI events came from a single location.
Researchers concluded that when patients were exposed to C. diff infections in the emergency department’s CT scanner, it was associated with a 4% incidence of CDI. They also noted that the association between CT exposure and CDI was still significant even after adjusting other influences such as antibiotic use and patients’ length of hospital stay. The association also remained significant when their sensitivity analysis extended the incubation period from 24 to 72 hours.
Having identified the CT as a potential vector of infection, the hospital next looked at how the that happened. It found that cleaning practices for the device didn’t meet the standards set for other radiology suites, and took steps to address the problem.
While healthcare leaders will ultimately use EHR data to make broad process changes, addressing day-to-day problems that impact care is also valuable. After all, finding the source of CDI is no trivial manner.
For example, a study recently concluded that ambulatory care organizations can do a pretty good job of analyzing their workflow by using EHR timestamp data.
Researchers had developed the study, a write up of which appeared in the Journal of the American Medical Informatics Association, to look at how such data be could be used in outpatient settings. Aware that many outpatient organizations don’t have the resources to conduct workflow studies, the researchers looked for alternatives.
During the research process, the team began by studying the workflow at four outpatient ophthalmology clinics associated with the Oregon Health and Science University, timing each workflow step. They then mapped the EHR timestamps to the workflow timings to see how they compared.
As it turned out, the workflow times generated by analyzing EHR timestamps were within three minutes of observed times for more than 80% of the clinics’ appointments. The study offers evidence that outpatient organizations can examine their workflow without spending a fortune, using data they already collect automatically.
Of course, hospitals will continue to do more in-depth workflow analyses using higher-end tools like big data analytics software. These efforts will provide a multidimensional picture that wouldn’t be available using only timestamp analysis. But for hospitals and clinics with fewer resources, timestamp analysis may be a starting point for some useful research.