Training Medical Interpreters to Improve Language Assistance: Lessons from 6-years of training workshops for interpreters throughout Los Angeles County

An estimated 50 million people in the U.S. do not speak the same language as their health care provider and 23 million are considered limited English proficient (LEP). To deter discriminatory and unsafe services, federal and state laws mandate language assistance services free of charge to all LEP patients at all points of medical care. Interpreting by a bilingual individual either in person of by telephone is the primary means of round-the-clock, rapid access to the multitude of languages providers may encounter. Currently, there are no training methods sanctioned by national and state association of medical interpreters. Several popular programs exist to train bilingual people to interpret in health care, but no published research exists to demonstrate their effectiveness in preparing and improving interpreting skills. However, some state laws are requiring organizations to show evidence of training completion by persons who interpret (e.g., California's Senate Bill 853). A two-phase study was conducted with a training program to improve interpreting in health care with Los Angeles County, CA. The training content was based on the California Standards for Health Care Interpreting (2002) and other published guidelines for effective interpreting. Phase 1 was a retrospective analysis of pre-training surveys of 865 participants between 2003 and 2008. This is the largest sample of participants and longest duration of interpreter training to be studied in the U.S. The analysis examined characteristics of interpreters and correlates of interpreter readiness (e.g., history of training, interpreting, level of language proficiency and medical terminology). Phase 2 was a prospective evaluation of 88 participants in 2008 to determine changes in pre-post training in knowledge and skills recommended by California Health Care Interpreting Standards, and predictors of improvement based on variables examined in Phase 1. A 13-item addendum to the original pre-training survey and implemented again post-training. For the cumulative sample, the average number of years interpreting was 6 (Range 0 – 59). The most common language interpreted was Spanish (14%). Statistically significant (P < .05) improvement was found for 2 of 13 items related to standards of quality interpreting: knowledge of the civil right for interpreting and for the correct interpretation of a diabetes management scenario. This study represents very early but important steps in the research of training interpreters as a means to prevent health care disparities for patients with limited English proficiency.