Methodological problems in comparing English-speaking and Spanish-speaking patients' satisfaction with interpersonal aspects of care.

OBJECTIVES: The authors examine the reliability and validity of English and Spanish versions of a patient satisfaction measure, the Interpersonal Aspects of Care (IAC) Examiner Scale.
METHODS: The examiner scale was administered to 234 Spanish-speaking and 250 English-speaking patients seen in the emergency department of a large public hospital. Reliability and validity coefficients were calculated for both administrations. Differences in item and scale response distributions were compared using two methods of response dichotomization.
RESULTS: In general, the examiner scale was reliable and valid. However, the Spanish version of the scale was significantly less reliable and valid. Depending on the method of item response dichotomization, significant differences between Spanish-speaking and English-speaking patients were found for the majority of the individual scale items. This was due primarily to a tendency for Spanish-speaking patients to respond "good" to items more frequently than did English-speaking patients.
CONCLUSIONS: The results bring into question the use of certain types of response formats and the practice of dichotomizing responses when evaluating patient satisfaction in Spanish-speaking patients.