Retrospective analysis of the effect of limited english proficiency on abdominal MRI image quality

Abstract

Purpose

To evaluate the effect of English proficiency on abdominal MRI imaging quality.

Methods

Three equal-sized cohorts of patients undergoing 3T abdominal MRI were identified based on English proficiency as documented in the EMR: Primary language of English; English as a second language (ESL)/no translator needed; or ESL, translator needed (42 patients per cohort for total study size of 126 patients). Three radiologists independently used a 1–5 Likert scale to assess respiratory motion and image quality on turbo spin-echo T2WI and post-contrast T1WI. Groups were compared using Kruskal–Wallis tests.

Results

For T2WI respiratory motion, all three readers scored the Translator group significantly worse than the English and ESL/no-Translator groups (mean scores across readers of 2.98 vs. 3.58 and 3.51; p values < 0.001–0.008). For T2WI overall image quality, all three readers also scored the Translator group significantly worse than the English and ESL/no-Translator groups (2.77 vs. 3.28 and 3.31; p values 0.002–0.005). For T1WI respiratory motion, mean scores were not significantly different between groups (English: 4.14, ESL/no-Translator: 4.02, Translator: 3.94; p values 0.398–0.597). For T1WI overall image quality, mean scores also were not significantly different (4.09, 3.99, and 3.95, respectively; p values 0.369–0.831).

Conclusion

Abdominal MR examinations show significantly worse T2WI respiratory motion and overall image quality when requiring a translator, even compared with non-translator exams in non-English primary language patients. Strategies are warranted to improve coordination among MR technologists, translators, and non-English speaking patients undergoing abdominal MR, to ensure robust image quality in this vulnerable patient population.