Case study: identifying potential problems at the human/technical interface in complex clinical systems
The deployment of computerized physician order entry (CPOE)/Electronic Medical Records (EMR) systems in the healthcare organizations continues to grow in an effort to increase the level of medical care. While the use of such technologies hold promise in the reduction of some medical errors, this paper highlights how unanticipated use of CPOE/EMR system functions may result in the occurrence of adverse drug events.
The case study examined in this article involves that of a 76-year old patient currently receiving warfarin (coumadin) therapy on an outpatient basis, due to a history of deep vein thrombosis (DVT). The patient’s condition warranted an adjustment in warfarin dosage from 2 MG/day to 4mg of warfarin for 6 days/week and 2mg of warfarin one day/week. The change in dosage was noted in the outpatient EMR chart notes but the initial dosage of “WARFARIN (COUMADIN) NA 2MG TAB TAKE AS DIRECTED BY COUMADIN CLINIC BY MOUTH EVERY DAY TO PREVENT BLOOD CLOTS” was left unchanged on the patient’s medication list.
Deterioration in the patient’s medical condition, required that he be hospitalized on three occasions. During the first two hospital visits, the medical staff failed to recognize the change in the patient’s medication dosage as documented in the outpatient chart notes. Making use of the copy and paste function of the CPOE system, the initial dosage of “WARFARIN (COUMADIN) NA 2MG TAB TAKE AS DIRECTED BY COUMADIN CLINIC BY MOUTH EVERY DAY TO PREVENT BLOOD CLOTS” was simply copied from the outpatient EMR medication list into the hospital CPOE chart. As a result, the patient received insufficient warfarin therapy. The change in medication dosage was finally discovered during the patient’s third hospitalization, after full review of the patient’s entire outpatient chart.
The authors basically attribute the occurrence of this adverse drug event to the unexpected manipulation of systems features by clinicians in order to reduce end user burden and facilitate ease of use. Instead of generating a new prescription in the appropriate medication list template, outpatient staff documented the change using text in the chart notes. Additionally, hospital staff most likely felt it was easier to access and copy the existing medication order from the outpatient medication list rather than actually searching through the entire outpatient chart for that information.
The authors believe such behaviors may have been anticipated had the institution employed process analysis tools such as cause-and-effect, fishbone analysis and failure mode and effect analysis used in “human factors engineering”. Such tools allow one to estimate the risk of the occurrence of system errors by identifying weaknesses in both human and technological processes used in delivering health care. The authors suggest that this analysis be ongoing, given the increased complexity of the available technology and healthcare organization infrastructure. While technology may use as a tool in decreasing adverse patient occurrences, all involved in its implementation must be cognizant of the additional vulnerabilities it may introduce.