Dedicated Medication Nurse Did Not Reduce Error Rates

Abstract & Commentary

Synopsis: Use of a dedicated medication nurse had no effect on the total number of medication errors based on observations performed in medical, surgical, and mixed medical-surgical units.

Source: Greengold NL, et al. The impact of dedicated medication nurses on the medication administration error rate: A randomized controlled trial. Arch Intern Med. 2003;163:2359-2367.

Medication administration is an activity prone to errors. A variety of potential causes have been posed, including the proliferation of new drugs, increased patient acuity, and increased workload due to the nursing shortage. This study was performed to determine whether use of a dedicated medication nurse would reduce error rates by allowing the nurse to focus on medication administration. The study was conducted simultaneously at an academic community hospital (hospital A) and a university teaching hospital (hospital B) over a 12-week interval. There were 2 study units at each institution. At hospital A, the 2 study units admitted either medical or surgical patients. At hospital B, both units admitted mixed medical and surgical patients. The nurses were 16 volunteers recruited from unit staff and randomly assigned to function as either a medication nurse or general nurse for a 6-week block of time. Each medication nurse received 8 hours of training. Trained observers watched and recorded all errors including those related to drug, dose, formulation, route, rate of administration, reconstitution, administration technique, and omitted drugs. Each medication nurse was responsible for 16-18 patients and general nurses an average of 6 patients.

The total error rate was 15.7% for medication nurses and 14.9% for general nurses (P = 0.84). Error rates for medication nurses were higher in about half the weeks. When the two institutions were compared, the total error rate was higher for medication nurses at hospital B (19.7%) than hospital A (11.2%) (P < 0.04) but not different for general nurses (15.0% vs 14.7%, respectively). The most common medication errors were administration technique (6.4%), dose preparation (1.4%), omitted drugs (0.9%), and incorrect dosage (0.8%). There was no significant difference in errors between medication and general nurses on medical units, but there was a lower (P < 0.01) rate of errors for medication nurses on surgical compared to medical floors. There was no known association between the errors observed and patient outcome.

Comment by Leslie A. Hoffman, PhD, RN

The premise of this study was that a simple change in work design (dedicated medication nurse) and educational intervention might decrease the number of medication errors. In fact, the intervention had no effect the majority of the time. With the exception of the surgical units (where fewer medications were administered), the error rate was unchanged or tended to be higher with a medication nurse. At one time, team nursing was the predominant form of work organization, and a dedicated medication nurse was part of this work design. Team nursing was largely abandoned due to the belief that patient care is best carried out when the nurse is responsible for all aspects of patient care and can analyze patient response from a holistic perspective. In contrast, the medication nurse focuses on the process of "passing out pills" and may miss important signs and symptoms that suggest the need for a change in the plan of care.

It was interesting that significant differences in favor of the medication nurse only occurred in surgical units where the number of medications per patient is usually less. Study data support this potential, as Greengold and associates calculated the potential for error based on the number of medications administered. There were 75 opportunities per day for medical units and 48 opportunities per day on surgical units.

In commenting on their findings, Greengold et al posed 2 possible explanations for their failure to find a difference in error rates. First, there were differences in medication administration systems with the system that was judged more complex in the hospital (B) with higher error rates for medication nurses. This, however, would affect both groups. Second, they posed that there may be a threshold number of medications, activities, and patients that one nurse can manage, above which the error rate increases. Therefore, adding a medication nurse with responsibility for up to 18 patients provided no advantage. The latter explanation seems the more logical and correct for medication as well as general nurses.

Dr.  Hoffman is Professor Medical-Surgical Nursing Chair, Department of Acute/Tertiary Care University of Pittsburgh School of Nursing