A study published recently in the Annals of Emergency Medicine looked at whether the use of a trained, professional interpreter with limited-English or non-English-speaking patients in hospital emergency departments (ED’s) reduced the incidence of medication errors and other mistakes. The study compared situations in which a trained interpreter was present with situations with an amateur translator and situations with no interpreter. Patients who had the assistance of professional interpreters suffered a far lower rate of risky translation errors, particularly errors related to medication dosages. The findings could provide a helpful model for hospitals seeking to reduce pharmacy misfills due to language barriers, and also for other communication barriers like hearing impairment.
The researchers conducted a “cross-sectional error analysis” of ED visits captured on audiotape over the course of thirty months. The visits occurred at two Massachusetts pediatric ED’s. They reviewed fifty-seven meetings between doctors or other medical professionals and Spanish-speaking patients and caregivers with limited English proficiency. Twenty of the visits utilized professional interpreters, and twenty-seven had “ad hoc interpreters.” The remaining ten visits occurred with no interpreters.
In reviewing the fifty-seven meetings, the researchers counted nearly two thousand translation errors. Eighteen percent of those errors had the potential to affect the patient’s treatment. The percentage of potentially harmful errors, when broken down among the three types of meetings, revealed that meetings with professional interpreters had just over half the rate of errors as the other meetings. While twelve percent of the translation errors in visits with professional interpreters were potentially harmful, the rate for visits with amateur translators was twenty-two percent. The rate for visits with no interpreter was twenty percent. Interestingly, the visits with no interpreter at all had a lower error rate than those with an ad hoc translator.