CARS coverage: Twenty-three reasons reforms resulting from the IHE make sense

There are 23 good reasons for introducing concepts from the Integraing the Healthcare Enterprise project into the U.S. hospital system, according to Dr. Steven Horii, a radiologist at the University of Pennsylvania Medical Center.

That is the number of individual steps that can be eliminated from the process of ordering, carrying out, and interpreting an ultrasound scan at a institution that switches from a film- and paper-based system to a fully integrated information network, Horii told delegates at the 15th International Congress on Computer Assisted Radiology and Surgery (CARS 2001) in Berlin.

Horii described the clinical impact of IHE-based reforms introduced at his own institution. The process - from the physician ordering an ultrasound scan to receiving the radiologist's report - used to involve at least 32 separate tasks. Indeed, the number could rise to as many as 59 tasks at institutions without a computerized records system in the form of a HIS or RIS.

A new system that allowed automatic data input and full integration of the HIS and RIS networks reduced the number of tasks to nine. A huge number of routine, time-consuming chores involving the ward clerk, radiology department scheduler, transport department staff, sonography technician, radiologist, and others were taken out of the loop.

Such changes in workload provides major benefits in productivity, cutting the time required for a report to appear on patient's chart from up to three days to a matter of minutes.

"Every step in the process is a potential delay point and every manual data entry point is a potential source of error," Horii said.

Studies have shown that there is a 20% chance of a mistake being introduced each time data is keyed in manually. Some mistakes may be trivial, but others could easily lead to misidentifying the patient.

Horii warned, however, that the records system must be fully integrated if the hospital is to achieve the expected efficiency gains. A conventional PACS requiring manual data entry at critical stages would be unlikely to have the expected impacts on the sonographer's workload. Any benefits from eliminating the need to process film would be offset by the time spent keying in data. In a hospital like UPMC, which carries out 20,000 procedures a year, spending an estimated five minutes inputting and accessing data wastes 207 eight-hour technician shifts.

IHE-based reforms can avoid potential problems in a number of clinical scenarios. These include assigning a temporary ID to an unconscious patient brought into the emergency room or simplifying the procedure when a whole-body CT scan must be examined by several different radiologists, Horii said.

The system introduced at UPMC is not yet fully functional but has already shown its potential to improve productivity and reduce errors. It has cut in half the time that sonographers spend interacting with the information system, for example.

Still, legitimate concerns remain that need to be addressed by institutions contemplating the introduction of a similar system, Horii said. Relying on a single database could be risky in the event of a system crash, and contingency plans should be designed for an emergency paper-based system. Staff training at all levels is also important. There is no benefit in providing results inside 25 minutes if the physician is still mentally attuned to waiting for days to receive a printed report.