(Implementation of rules based computerised bedside prescribing and administration: inervention study
Reducing errors made by emergency physicians in interpreting radiographs: longitudinal study
Two studies in this week's issue look at how changes in the way things are done can have a significant impact on safety.
Nightingale and colleagues from the Department of Medicine at the University of Birmingham describe a rules based system for the prescribing and recording of drugs given to patients. The system can be accessed from the bedside through wireless terminals. During 11 months of monitoring, the system prevented 58 unsafe prescriptions and gave over 700 high level warnings. Complete and legible prescriptions have eliminated transcription errors and most of the staff on the 64 bed renal unit, where the system was tested, felt that it was better than handwritten prescriptions.
As Dr Reinertsen points out in his editorial, medication error is the most common single preventable cause of injury to patients: baggage handlers, he says, make fewer mistakes with our luggage than do medical staff with drugs given to patients in hospital.
In a study from the Department of Emergency Medicine at Overlook Hospital in New Jersey, Espinosa and Nolan show the impact of learning from mistakes, and of cooperation between clinical disciplines. Monthly team meetings, in which all emergency staff were involved, reviewed mistakes made in the interpretation of x-ray pictures. An error file was set up and used for teaching. A system was then devised for daily use all year round in which every film was interpreted initially by the emergency room doctor and then reviewed within 12 hours by a radiologist, as a quality control measure.
This replaced a system in which responsibility for the initial reading varied between the emergency room doctor and the radiologist, depending on the time of day and day of the week. The error rate more than halved after the first intervention and fell further when the system was re-configured. The authors estimated that the number of potential mistakes fell from 19 per 1000 cases to just three.
Contacts:
Dr Peter Nightingale, Wolfson Computer Laboratory, Department of Medicine, University of Birmingham. Email: P.G. Nightingale@bham.ac.uk
Dr James Espinosa, Department of Emergency Medicine, Overlook Hospital, Summit, New Jersey, USA. Email: jim010@aol.com