Repeat medication errors in nursing homes: Contributing factors and their association with patient harm.

BACKGROUND: Medication errors are highly prevalent in long-term care facilities and are responsible for preventable injury. Repeat medication errors, or identical events occurring multiple times in the same patient, may be particularly preventable.

OBJECTIVES: This study assessed the factors that contribute to repeat medication errors and the association between repeat medication errors and patient harm.

METHODS: In this cross-sectional analysis, medication error reports submitted by licensed nursing homes to North Carolina's Medication Error Quality Initiative-Individual Error Web-based incident reporting system were analyzed for fiscal years 2006-2008. When reporting errors, the sites were asked whether the event was identically repeated within the same patient. Repeat medication errors were defined as identical events in terms of patient characteristics, drug involved, error type, potential cause, phase of the medication care process, and personnel involved. Repeat errors were compared with nonrepeat errors. Multivariate logistic regression was used to explore whether certain patient or error characteristics were related to a higher likelihood of repeat errors, and a similar analysis was used to explore whether repeat errors were related to patient harm.

RESULTS: Of the total 15,037 errors reported by 294 unique nursing homes, 5615 (37.3%) were repeated one or more times. Among the repeat errors, the associated event within each error was repeated a mean (SD) of 10.7 (14.3) times. Wrong dosage (65.1% [3654/5615]) and wrong administration (10.2% [571/5615]) were the most frequent repeated events. In multivariate analysis, repeat errors occurred less frequently among younger residents (aged <75 years) than among older residents (aged >or=75 years) (odds ratio [OR] = 0.85; 95% CI, 0.79-0.93) and among residents able to direct their own care compared with cognitively impaired residents (OR = 0.87; 95% CI, 0.81-0.95). Patient harm was reported in only 1.2% (68/5615) of repeat errors and 0.6% (55/9422) of non-repeat errors. A multivariate analysis of patient harm found that repeat errors were more likely to be harmful than were nonrepeat errors (OR = 2.11; 95% CI, 1.43-3.11).

CONCLUSIONS: Repeat medication errors in nursing homes are a common occurrence and have greater odds of being associated with harm than do nonrepeat errors. Future patient-safety research should focus on factors related to repeat errors.