Researchers from Brigham and Women’s Hospital (BWH) and Massachusetts General Hospital analyzed nearly 4,000 computer generated prescriptions received in a commercial pharmacy chain and found that one in ten contained at least one error, a third of which had the potential to cause a patient harm. This rate is similar to the error rate for prescriptions that are written by hand. They also found that the number, type and severity of errors varied by computerized prescription systems suggesting some systems may be better than others. The findings appear in the June 30, 2011 issue of the Journal of the American Medical Informatics Association.
The researchers analyzed 3,850 computer-generated prescriptions received by a commercial outpatient pharmacy chain in three states over a four week period. They found that 452 prescriptions contained 466 total errors, with 163 (35 percent) of the errors potentially causing a patient harm, also known as a potential adverse drug event (ADE). There were 95 errors that the researchers considered significant and 68 that were considered serious. None of the errors were life-threatening.
Prescribing error rates and potential for ADEs varied widely among the electronic prescribing systems in use with error rates ranging between systems from 5.1 to as high as 37.5 percent. Different systems were associated with different error types. For example, the most common errors in one system were omitting how long a prescribed treatment should be taken,. In another system the most common errors were inappropriate abbreviations.
“With more than 3 billion prescriptions written annually in the U.S. alone, this could amount to 385 million errors each year, with 128 million of them having the potential to cause patient harm,” said Jeffrey Rothschild, MD, from the Center for Patient Safety Research and Practice at BWH and a co-author on the paper.
The researchers recommend several strategies to help eliminate errors, suggesting “forcing functions,” which would not allow prescriptions to be filled if information is missing, specific decision support to check maximum dosages and calculators to resolve inconsistent quantity errors by eliminating redundant data entry.
“By providing better training and adding some interventional tools to the existing computerized prescribing system, medication errors could be significantly lowered and hopefully even eliminated,” said Karen Nanji, MD, from the Department of Anesthesia, Critical Care and Pain Medicine at Massachusetts General Hospital and lead author of the study. “Implementing a computerized prescribing system without a comprehensive set of checks in place will not lower medication errors.”
The research was supported by a grant from the Agency for Healthcare Research and Quality and in part from the Harvard Risk Management Foundation.