Dear Editors,

We read with interest Lopez et al.'s recent article.1 Patient–provider communication during hospitalization and at the time of discharge is increasingly recognized as important to quality of care and safety for all patients.2 This field of inquiry is of importance as the number of limited English proficiency (LEP) patients in our country grows, and as the authors document once again, interpreters are greatly underutilized in health care. The question asked by the authors is of great import—how does the use of an interpreter impact length of hospital stays (LOS) and post-discharge outcomes? However, we question the interpretation and importance of their findings, given the significant limitations of their measure of LEP, the fact that there is no measure of the providers’ use of non-English language, and the very weak measure of interpreter use.

The authors’ measure of LEP was based on patients’ documented language preference at the time of hospital admission. Karliner et al. document well that the optimal method of identifying LEP patients should be a two part process. Patients should be asked how well they can discuss their symptoms in English and this should be followed by a language preference question.3 Lower specificity of the language preference question alone could lead to misclassification of some patients as LEP who are, in fact, able to effectively communicate in English with their physicians. Further, the authors did not consider the possibility that some of the patients in the study may have had physicians who spoke their language. This language-concordant care could have resulted in better communication and shorter length of stay.4 There was also no documentation of the presence or absence of English proficient family members or use of phone interpretation. Finally, as we understand it, patients were included in the interpreter service utilized group if patients received interpreter services as little as one time. As LOS was, on average, more than 5 days, it would be highly unlikely that one-time use of an interpreter would have any impact on LOS or other outcomes.

We recognize that many of these limitations were a consequence of conducting a retrospective study. These results should be interpreted within the context of these limitations and within the greater context of the existing literature, which overwhelmingly supports the use of professional interpreters as important to the quality of care delivered to LEP patients.5