First study to examine chemotherapy errors in ambulatory care for cancer patients

In one of the first studies to examine chemotherapy errors in ambulatory care for cancer patients, researchers at Dana-Farber Cancer Institute (DFCI) and Brigham and Women's Hospital (BWH) have found that about three percent of chemotherapy orders in three outpatient infusion clinics studied contained mistakes.

Most of the errors were intercepted by nurses and pharmacists before reaching patients, and none were life-threatening or caused patient harm; but the results show that room for improvement exists even in hospitals with strong error-prevention programs, the authors say.

The research, reported in today's online version of the journal Cancer, was made possible by Dana-Farber and BWH leaders' decision to share drug-order and patient-safety records with investigators. Both hospitals are established leaders in efforts to reduce medication errors and heighten patient safety. The findings of the study have prompted both Dana-Farber and Brigham and Women's to make changes in the hospital's automated drug order-entry system to further lessen the chance of mistakes.

"Our results show that while safeguards such as computerized order-entry systems - used at both Dana-Farber and Brigham and Women's - significantly reduce drug-order errors, additional improvements are still possible, and necessary" says the study's co-lead author, Tejal Gandhi, MD, MPH, of Brigham and Women's.

Adds co-lead author Sylvia Bartel, RPh, MHP, of Dana-Farber, "DFCI's leadership supported the in-depth review of all medication orders to gain information about potential system defects. The goal was to utilize the results of the study to make system improvements and ensure a safe medication process for our patients."

Previous studies have estimated that about five percent of drug orders for hospitalized patients have errors, but much less scientific attention has been given to the prevalence of such mistakes in outpatient settings. While medication errors often don't harm patients, the complexity of some chemotherapy regimens, and the toxicity of the drugs, makes it particularly important to minimize mistakes in ordering and administering cancer therapies.

In the current study, Gandhi, Bartel, and their colleagues from BWH, Dana-Farber, and the Harvard School of Public Health reviewed more than 10,000 medication orders from Dana-Farber's adult and pediatric ambulatory oncology infusion clinics, which used a computerized or paper medication-ordering system, respectively. Using a strict definition of error, they found that three percent of the orders contained errors, one-third of which were deemed serious.

Rating the errors by severity, researchers determined that 82 percent of the errors in adults and 60 percent in children had potential for harm to patients. Pharmacists and nurses caught 45 percent of the potentially harmful errors before they reached patients, and none of the errors actually caused patient harm.

In the adult clinics, which used a computer-aided ordering system, and the pediatric clinic, where a paper-based ordering system was in place, the most frequent errors involved omitted or incorrect dosages and failure to discontinue orders.

To reduce the chances of future medication order errors, officials at DFCI have instituted several changes. In the pediatric clinic, orders are now placed via computer. In the adult clinics, physicians now use a more sophisticated computer application with more of the drug-ordering information embedded within it. For medications that are usually given in tandem, the program now requires physicians to order both at the same time.

Co-authors of the study are Lawrence Shulman, MD, Deborah Verrier, RN, Angela Cleary, RN, of Dana-Farber; Elisabeth Burdick, MS, Jeffrey Rothschild, MD, MPH, and David Bates, MD, MSc, of Brigham and Women's; and Lucian Leape, MD, of the Harvard School of Public Health.