SAN FRANCISCO — Half of 83 prescriptions for analgesics written for 77 children being sent home from the hospital contained one or more errors, including 2 prescriptions with errors that could have significantly harmed the patients, Benjamin H. Lee, M.D., said at the annual meeting of the American Academy of Pediatrics.
Unbeknownst to the primary medical or surgical physicians who wrote the discharge analgesic prescriptions, investigators from the Johns Hopkins Medical Institutions' (Baltimore) pediatric pain service secretly monitored the prescriptions during the study and rewrote any they considered dangerous, so no children were harmed, said Dr. Lee of Johns Hopkins.
“We were surprised at this potential adverse drug event rate of 2.4%. That's not insignificant in this small patient series,” he said, adding, “I don't think this is something that's limited to Johns Hopkins Hospital.”
Discharge time is a vulnerable period for inpatients, who lose the safety net of pharmacists, nurses, and multiple physicians who look at medication orders while the patient is hospitalized, he explained. When the patient is sent home with analgesics, a single physician writes the prescription and the discharge orders, which are reviewed usually by a single nurse, with no pharmacists or other providers involved.
The two dangerous prescriptions in the study were for opioids. One included a 10-fold overdose error. The other included instructions for a long-acting medication that could lead a patient to take multiple doses all at once. All patients got prescriptions for opioids at discharge, and 7% also received NSAIDs.
Most of the prescription errors were not clinically significant; the study used a strict definition of error. The two most common causes of errors, however, illustrate problems that could lead to patient harm: a lack of any identification of weight or weight-based dosing in the prescription, and incomplete information about dispensing of the medication.
For patients weighing less than 40 kg, no weight was recorded on 45% of analgesic prescriptions. Investigators found discrepancies between the written prescriptions and the discharge data form in 10% of cases. Physicians wrote an incorrect name or patient identifier in 4% of analgesia prescriptions.
A separate study is underway to see if using a computerized prescription-writing program that includes weight-based dosing for pediatric patients will reduce errors and improve patient safety, Dr. Lee said.
The current results echo those of a similar study by Dr. Lee and his associates of errors in 122 prescriptions written for 75 patients in the hospital's day surgery unit, which included mostly adults, with some pediatric patients. They found an error rate of 57% including potentially dangerous errors in 2.5%.
“If this is happening at a tertiary-care hospital with a lot of oversight, with providers who are very comfortable taking care of critically ill patients, this is probably a phenomenon that extends to other institutions and other settings,” Dr. Lee said.