Medication errors in nursing homes and small hospitals.

An observation method for measuring the rate of medication errors, which can be used as an outcome indicator of a medication system's quality, was evaluated in a national sample of long-term care facilities (LTCFs) and small hospitals. Trained nurse and pharmacist observers observed nurses administer medications during the three-hour period surrounding the peak medication workload on one day in national sample of 58 LTCFs and 10 hospitals. Opportunity-for-errors (OE), defined as the total number of doses ordered plus the unauthorized doses given, were counted, as well as all medication errors. The error rate was calculated as the proportion of errors in total OEs. The reliability of the method was evaluated on seven days by comparing the results of a researcher and observer, who were observing the same nurse. The mean-medication-error rate was 12.2 and 11.0% in the LTCFs and hospitals studied, respectively. Three LTCFs and four hospitals had error rates of zero. Only 31% of the LTCFs and 40% of the hospitals would pass a medication-error limit standard of 6%. The authors concluded that the observation method was promising. Recommendations for further study included: (1) implementation of a one-year project to evaluate observer efficiency after becoming proficient with the method, (2) improvement of the reliability measure, and (3) examination of the relationship of medication errors with structure and process variables.