Pharmacist Participation in Medical Rounds Reduces Medication Errors

Kimberly K. Scarsi, Michael A. Fotis, Gary A. Noskin


Am J Health Syst Pharm. 2002;59(21) 

In This Article


Medication errors are a significant problem in U.S. hospitals. Because of the complexity of this issue, a multi-disciplinary approach is required to effectively minimize the potential for medication errors. The specialized training and education of pharmacists allow them to serve as medication experts and potentially prevent medication errors. These data lend more evidence to the compelling body of literature supporting the active participation of pharmacists in medical rounds as a means of reducing the frequency and duration of medication errors.[5]

An important observation was that nearly 80% of the patients in the control group had a medication error. The participation of a pharmacist in medical rounds can reduce this number substantially. To our knowledge, this study is the first to document that pharmacist participation in medical rounds not only decreases the frequency of errors but also the duration that an error persists once it has occurred. Not surprisingly, errors still occurred in the intervention group, even with a pharmacist present. These errors were typically identified and addressed shortly after the order was written and often before any doses of medication were administered to patients. We attribute this difference to the greater knowledge the pharmacist had of the patient's medical problem, history, and current goals of care. This increased knowledge about the patient made it much more likely that the pharmacist would recognize a potential error and take the appropriate action to prevent the error from occurring or continuing.

These data have had a significant impact on our institution. After this evaluation, two additional pharmacists attended medical rounds on the general medicine service. An automated medication administration record has also been implemented on the general care units of our institution. We anticipate that this automated record will significantly reduce the number of administration errors related to time-and food-dependent medications. Also, a patient safety team has been assembled, including physician, nursing, administration, and pharmacist staff, to further evaluate the broad scope of medical errors and the best way to reduce them in our institution.

During this evaluation, 37 errors were documented via the MedMARx system; 20 of those errors were reported in the intervention group. This demonstrates the difficulty in thoroughly and accurately documenting medication errors in most institutions. Since this study, the pharmacy and therapeutics committee, along with the patient safety team, has facilitated the documentation of errors in our institution. Specifically, these committees have been working toward identifying the systems failures, not the parties involved, that may contribute to errors. Emphasis has also been placed on error reporting with ease to encourage more error reporting in a time-efficient manner.

An active pharmacist presence on rounds was associated with a reduction in the rate and duration of medication errors. This improvement in safety may have been due to the extra time spent evaluating these patients by the pharmacist on rounds. We do not argue that extra pharmacist time would contribute to a reduction in errors; however, a collaborative effort allows the pharmacist to be fully informed about patient-specific issues and have greater access to other clinical decision-makers who affect patient safety.

There are several limitations inherent to retrospective evaluations of this type that must be considered that could impact the generalizability of the data. We used subjective evaluation of medication errors as interpreted by trained pharmacists. The blinding of the pharmacist reviewers and the definitions of medication errors (as set forth by IOM) helped compensate for this limitation. Also, the detection of medication errors relies on the information recorded in the medical record, which may result in underestimating some types of errors. However, the concurrence of prospectively and retrospectively collected data in our trial seems to indicate that, at least in this case, underestimation was minimal. Finally, we evaluated a small sample size in an academic medical center, which may not be representative of all hospitalized patients. The general medicine service was highly motivated, which facilitated interactions among all health care providers. Care was taken to minimize potential bias when possible. Patients were matched based on admission date, age, sex, duration of stay, and number of medication orders written to minimize the potential for a difference in the frequency of errors resulting from outside factors.

Pharmacist participation in medical rounds reduces the frequency of medication errors as well as their duration once they are identified.

Presented at the ASHP Midyear Clinical Meeting, Las Vegas, NV, December 6, 2000.